VOLUME 68 2017 TOME 68
EDITORIAL BOARD – COMITÉ DE RÉDACTION 2017
Editors-in-Chief Associate Editors
BRIANNA CARMICHAEL NATACHA CONNELLY BOSSÉ
ASHLEY GODFREY CATHERINE HUTCHENS 2017 CanLIIDocs 175 KELCIE WHITE Rédactrices en chef SONNY XUE
Rédactrices et rédacteur adjointes
Honourary Editor-in-Chief Faculty Advisors
THE HON. G. V. LA FOREST ANNE WARNER LA FOREST JANE THOMSON Rédacteur en chef honoraire Conseillères academiques
Business Manager Translator
CHELSEA BRAKE NATACHA CONNELLY BOSSÉ Gestionnaire Traductrice
TO BE CITED AS: (2017) 68 UNBLJ MODE DE RÉFÉRENCE: (2017) 68 RD UN-B
ISSN: 0077-8141 Copyright 2017 Droits d’auteur University of New Brunswick Law Journal © Revue de droit de l’Université du Nouveau-Brunswick Printed in Canada – Imprimé au Canada Since 1947, the University of New Depuis 1947, la Revue de droit de Brunswick Law Journal has published l’Université du Nouveau-Brunswick a articles, comments, and reviews in wide- publié des articles, des commentaires, et ranging areas of law. The Journal is des avis juridiques dans plusieurs produced by students from the Faculty of domaines du droit. La Revue est publiée Law at the University of New Brunswick par les étudiants de la Faculté de Droit de with the advice of one or more faculy l’Université du Nouveau-Brunswick sous members. The objective of the Journal is la direction d’un ou plusieurs membres du to promote academic discussion of current corps professoral. L’objectif de la Revue legal issues, problems, and philosophies. est de promouvoir les discussions académiques concernant les The Journal expresses its gratitude to the préoccupations, problèmes et philosophies Law Society of New Brunswick, the New juridiques actuels. Brunswick Law Foundation, the Faculty of Law at the University of New La Revue désire exprimer sa Brunswick, and patrons for their ongoing reconnaissance envers le Barreau du encouragement, support, and assistance. Nouveau-Brunswick, la Fondation pour l’avancement du droit du Nouveau- The opinions expressed in the Brunswick ainsi que ses donateurs pour contributions do not necessarily represent leur encouragement, leur aide et leur the views of the Journal. soutien constants.
2017 CanLIIDocs 175 Les opinions exprimées dans les contributions ne reflètent pas nécessairement les opinons de la Revue.
Copyright Droits d’auteur Copyright © 2017 University of New Droits d’auteur © 2017 Revue de droit de Brunswick Law Journal. All rights l’Université du Nouveau-Brunswick. Tous reserved. Unless otherwise stated, no part droits réservés. Sauf dispositions of the Journal may be reproduced in any contraires, aucune partie de la Revue ne form or by any electronic or mechanical peut être reproduite sous toute forme que means, including information storage and ce soit ou par aucun moyen mécanique ou retrieval systems, without permission informatique, y compris le stockage from the Editors-in-Chief. Reproduction d’information et systèmes de recherche de of this material without authorization by données, sans la permission des any method of duplication whatsoever is a Rédacteurs ou Rédactrices en chef. Toute violation of copyright. reproduction de ce matériel sans avoir obtenu l’autorisation par toute méthode Except as otherwise provided, the author duplicative est une violation des droits of each article in this volume has granted d’auteur. to the Journal the non-exclusive licence to publish, reproduce, distribute, and use Sauf dispositions contraires, l’auteur de the article in print and electronic form. chaque article de ce volume a donné un The Journal authorizes the reproduction droit non-exclusif à la Revue de publier, of articles to be made for classroom use, reproduire, distribuer et utiliser les articles provided that the author and the Journal en format papier et électronique. La Revue are identified and proper notice of autorise la reproduction des articles à des copyright is affixed to each copy. fins académiques, tant et aussi longtemps que l’auteur de l’article et la Revue soient identifiés et que l’avis relatif au droits d’auteur soit inscrit sur chaque copie.
Subscriptions and Back Issues: Abonnements et anciens numéros:
Subscription Price: $15.00 CDN per volume Abonnement : 15,00 $ CA par volume Back Issues: $12.00 CDN per volume Anciens numéros : 12,00 $ CA par volume Index, Vols. 1-36: $10.00 CDN Index, Volumes 1-36 : 10,00 $ CA
Please address your request to: Veuillez communiquer avec :
Editors-in-Chief Rédacteurs en chef University of New Brunswick Law Journal Revue de droit de l’UN-B Ludlow Hall, PO Box 4400 Ludlow Hall, Case postale 4400 University of New Brunswick University of New Brunswick Fredericton, NB E3B 5A3 Fredericton (N-B) E3B 5A3 Email: [email protected] Courriel : [email protected]
Correspondence and Submissions: Correspondance et soumissions :
The Journal welcomes the submission of La Revue accepte des manuscrits non unsolicited manuscripts. Manuscripts sollicités. Les manuscrits doivent être must be written in either English or écrits en français ou en anglais et doivent French and must be submitted être soumis sur support électronique en electronically in Microsoft Word format. format Microsoft Word. Les références Citations should conform to the Canadian doivent être conformes au Manuel 2017 CanLIIDocs 175 Guide to Uniform Legal Citation, 8th ed canadien de la référence juridique, 8e éd, (Toronto: Thomson Carswell, 2014). Toronto, Thomson Carswell, 2014. La Preference is given to submissions that préférence sera accordée aux manuscrits are fewer than 60 pages double-spaced. de mois de 60 pages, rédigés à double Manuscripts submitted simultaneously to interligne. Les manuscrits présentés à other journals are not accepted. d’autre revues de façon simultanée ne seront pas acceptés.
All correspondence and submissions La correspondance et les soumissions should be addressed to: doivent être adressés à :
Editors-in-Chief Rédacteurs en chef University of New Brunswick Law Journal Revue de droit de l’UN-B Ludlow Hall, PO Box 4400 Ludlow Hall, Case postale 4400 University of New Brunswick University of New Brunswick Fredericton, NB E3B 5A3 Fredericton (N-B) E3B 5A3 Email: [email protected] Courriel : [email protected]
To be cited: (2017) 68 UNBLJ Reference : (2017) 68 RD UN-B
ISSN: 0077-8141 ISSN : 0077-8141
Printed and bound by KKP in Fredericton, Imprimé et relié par KKP à Fredericton, New Brunswick, Canada. Nouveau-Brunswick, Canada.
2017 CanLIIDocs 175 UNIVERSITY OF NEW BRUNSWICK LAW JOURNAL
REVUE DE DROIT DE L’UNIVERSITÉ DU NOUVEAU-BRUNSWICK
EDITORS’ PREFACE Brianna Carmichael & Ashley Godfrey…………………………………… 1
PART I: IVAN C. RAND MEMORIAL LECTURE AND PANEL
SELECTING SUPREME COURT JUSTICES: IS TRUDEAU’S SUNNY WAY A BETTER WAY? Peter H. Russell……………………………………………………………. 3
REFORMING JUDICIAL APPOINTMENTS: CHANGE AND CHALLENGE Rosemary Cairns Way…………………………………………………… 18
LE BILINGUISME : UNE EXIGENCE RAISONNABLE ET ESSENTIELLE POUR LA 2017 CanLIIDocs 175 NOMINATION DES JUGES À LA COUR SUPRÊME DU CANADA Michel Doucet……………………………………………………………. 30
VIRTUES AND SHORTCOMINGS IN CONSTITUTIONAL DYNAMISM: COMMENTARY ON PRIME MINISTER TRUDEAU’S 2016 CREATION OF AN INDEPENDENT ADVISORY BOARD FOR SUPREME COURT OF CANADA APPOINTMENTS AND ON THE INSTRUCTIONS FOR MAKING RECOMMENDATIONS FOR APPOINTMENT John D. Whyte……………………………………………………………. 36
THE FALSE DICHOTOMY BETWEEN REGIONAL REPRESENTATION AND OTHER FORMS OF DIVERSITY: REIMAGINING A REPRESENTATIVE COURT A. Wayne MacKay………………………………………………………... 46
PART II: KEYNOTE: VISCOUNT BENNETT MEMORIAL LECTURE
A JUDICIARY CLEAVED: SUPERIOR COURTS, STATUTORY COURTS AND THE ILLOGIC OF DIFFERENCE The Hon. Justice David Stratas…………………………………………... 54
PART III: FORUM – ISSUES IN ADMINISTRATIVE AND CONSTITUTIONAL LAW
THE SIGNAL AND THE NOISE IN ADMINISTRATIVE LAW Paul Daly…………………………………………………………………. 67
THE TIME HAS COME: STANDARD OF REVIEW IN CANADIAN ADMINISTRATIVE LAW Jonathan M. Coady………………………………………………………. 87
RENOVATING JUDICIAL REVIEW Matthew Lewans………………………………………………………… 109
IDENTIFYING THE REVIEW STANDARD: ADMINISTRATIVE DEFERENCE IN A NUTSHELL The Hon. Joseph T. Robertson………………………………………….. 145
BUNGLED POLICE EMERGENCY CALLS AND THE PROBLEMS WITH UNIQUE DUTIES OF CARE Bruce Feldthusen………………………………………………………... 169
THE COMPLICATED INTERSECTION OF POLITICS, ADMINISTRATIVE AND CONSTITUTIONAL LAW IN NUNAVUT’S ENVIRONMENTAL IMPACTS ASSESSMENT REGIME Daniel W. Dylan………………………………………………………… 202
PERFECTLY LEGAL, BUT STILL BAD: LESSONS FOR SEX WORK FROM THE DECRIMINALIZATION OF ABORTION Jula Hughes……………………………………………………………... 232
2017 CanLIIDocs 175 THE RIGHT TOOL FOR THE JOB?: FREEDOM OF ASSOCIATION UNDER PROVINCIAL HUMAN RIGHTS CODES Keir Vallance……………………………………………………………. 254
SOME INITIAL THOUGHTS ON WILSON V. ATOMIC ENERGY OF CANADA LTD AND EDMONTON (CITY) V. EDMONTON EAST (CAPILANO) SHOPPING CENTRES LTD Diana Ginn……………………………………………………………… 285
REVISITING THE APPLICATION OF SECTION 7 OF THE CHARTER IN IMMIGRATION AND REFUGEE PROTECTION Gerald Heckman………………………………………………………... 312
PART IV: COMMENT
THE DUTY OF FAIRNESS IN THE INVESTIGATIVE STAGE OF THE ADMINISTRATIVE PROCEEDINGS Lucie LaBoissonnière…………………………………………………… 357
PART V: STUDENT SUBMISSION
MARITIME LAW: SOVEREIGNTY IN THE ARCTIC Kyle Mercer……………………………………………………………... 365 EDITORS’ PREFACE
Forum: “Issues in Administrative and Constitutional Law”
The University of New Brunswick Law Journal has a longstanding history of topical and engaging volumes; it is our hope that Volume 68 represents a continuation of that tradition. The Editorial Board chose “Issues in Administrative and Constitutional Law” as our forum topic for Volume 68.
While the major theme of Volume 68 is administrative and constitutional law, our two annual lecture series, both published by the Journal, presented a “theme within a theme”: the Canadian judiciary. The first of these lectures was the Ivan C. Rand Memorial Lecture, given on October 20, 2016 by Professor Peter Russell. Professor Russell’s lecture, entitled “Selecting Supreme Court Justices: Is Trudeau’s Sunny Way a Better Way?”, was accompanied by a discussant panel of four experts. Their pieces appear in these pages alongside Professor Russell’s and we are honoured to print their contributions. 2017 CanLIIDocs 175
Justice David Stratas of the Federal Court of Appeal gave the Viscount Bennett Memorial Lecture on February 9, 2017 entitled, “A Judiciary Cleaved: Superior Courts, Statutory Courts, and the Illogic of Difference”. Justice Stratas’s contribution provided the centrepiece for Volume 68 of the Journal. We are thrilled to publish his submission, which assisted the Editorial Board in soliciting superb contributions from scholars and practitioners of administrative and constitutional law from across the country (and one ocean).
Volume 68’s Editorial Board is proud to continue the tradition of printing a submission by a student of the University of New Brunswick’s Faculty of Law which was established in Volume 67. Student submissions are reviewed and selected by the Editorial Board. This year, our student submission comes from third year student Kyle Mercer (JD Class of 2017). We are pleased to print his piece within.
The Journal would not be possible without the dedication and hard work of many hands and minds. We are grateful to the University of New Brunswick, its Faculty of Law, its professors and staff, and to the Law Society of New Brunswick. We would also like to thank those who contributed their excellent articles and our reviewers who ensured each article was of publishable quality. Finally, we would like to thank our Associate Editors for their tireless dedication, excellent work, and constant willingness to help.
We hope each reader will find something interesting, instructional, and inspiring within these pages.
Brianna Carmichael & Ashley Godfrey Editors-in-Chief, Volume 68 2017 CanLIIDocs 175
SELECTING SUPREME COURT JUSTICES: IS TRUDEAU’S SUNNY WAY A BETTER WAY?
Peter H. Russell*
On August 2, 2016, Prime Minister Trudeau announced a new process for selecting Supreme Court of Canada justices. The lack of a guarantee in the new process for observing the convention of regional representation on the Court – in this case replacing retiring Justice Thomas Cromwell with another jurist from Atlantic Canada – was the main focus of public concern. That concern may well have been the main reason for suggesting the topic for the Ivan Rand Memorial lecture that I had the honour to give at the University of New Brunswick in October 2016. With the announcement on Monday, October 17 that Prime Minister Trudeau has decided to have Newfoundland & Labrador’s Judge Malcolm Rowe appointed, that concern is over – for now. Though I will certainly discuss the regional representation 2017 CanLIIDocs 175 convention later in my talk, I would first like to place the Trudeau government’s reform of the selection process in a broad international and Canadian context and then look closely at all features of the new process.
The International Movement for High Court Reform
Over the past half century most of the world’s constitutional democracies have recognized the need for some check and balance on the discretion of political heads of government in selecting members of their country’s highest constitutional court. This idea has accompanied the adoption of constitutional or semi-constitutional bills of rights in many democracies: the judges who interpret and apply constitutional limits on elected governments should be selected and appointed through a process that is open and transparent and not controlled or dominated by the government that is subject to these judicially enforced limits on its actions.1
The most common approach to meeting this concern has been to establish judicial nominating committees to find outstanding candidates and recommend one or more of them for appointment. Final power to determine who will be appointed remains with the political, elected head of government, as is appropriate in a democracy, but the president or prime minister is required to choose a candidate recommended by the nominating commission.
* Peter H Russell is a Professor Emeritus of political science at the University of Toronto and is a leading scholar in the fields of Canadian politics and law. The following article reflects Professor Russell’s Ivan C. Rand Memorial Lecture, delivered at the University of New Brunswick Faculty of Law in October 2016. – Eds.
1 See Peter H Russell, “Judicial, Recruitment, Training and Careers,” in Peter Cane & Herbert M Kritzer, eds, The Oxford Handbook of Empirical Legal Research (Oxford: Oxford University Press, 2010) 322. 4 UNBLJ RD UN-B [VOL/TOME 68
Reforming the judicial selection process in Canada
In Canada interest in limiting the discretion of elected heads of government in the selection of Supreme Court justices came well before the “rights revolution”. It initially focussed on the Court’s role in interpreting the federal division of powers between governments rather than enforcing citizens’ rights against governments. In 1949 when Parliament was debating amendments to the Supreme Court Act to end appeals to the Judicial Committee of the Privy Council, Léon Balcer, a Conservative MP, thought it questionable that the tribunal that would be settling disputes between the two levels of government was so thoroughly a creature of the federal government.2 Another Quebec MP, Wilfred La Croix, proposed that four Supreme justices (on a bench to be expanded from seven to nine) be nominated by provincial governments.3
Later on, when our country plunged into those endless attempts at re-doing our Constitution, proposal after proposal gave the provinces a role in selecting Supreme Court justices. The most recent of these was section 19 of the
Charlottetown Accord that would have required the federal government to name 2017 CanLIIDocs 175 judges from lists submitted by the governments of the provinces and territories. Of course, as you know all too well, none of those proposals ever became law. And some of you may be saying to yourself “wasn’t that a good thing?” From the time the Supreme Court was established in 1875, right up to the early years of this century, Canada carried on with a very simple unreformed process of filling vacancies on the Supreme Court of Canada. Let me briefly describe the unreformed process.
Although in law the appointment of a Supreme Court justice is made by the Governor-in-Council (i.e., the cabinet), in practice it is the Prime Minister who decides who is to fill a vacancy. The Minister of Justice has always assisted the Prime Minister in making his or her decision by looking for promising candidates in the region of the justice who is leaving the Court. In the modern period, the minister has had the help of a special assistant who performs a function similar to that of a chief scout for a sporting franchise. In the search for good candidates various soundings and consultations would take place – sometimes with the Supreme Court’s chief justice, sometimes with provincial attorneys general, usually with provincial governments with which the federal government has friendly political relations and which is in the region in which the vacancy occurred, as well as with groups and individuals who might have strong views or useful information to impart. Even I, a non-lawyer political science professor, was consulted on one occasion. And of course, the government has always received heaps of unsolicited advice on who to appoint or who not to appoint, from groups and individuals. This is known as lobbying.
The central role of the Prime Minister in naming Supreme Court justices differs from the process of filling vacancies in the section 96 provincial courts and
2 House of Commons Debates, 21st Parl, 1st Sess, Vol 1 (11 October 1949) at 661.
3 House of Commons Debates, 21st Parl, 1st Sess, Vol 1 (27 September 1949) at 313. 2017] SELECTING SUPREME COURT JUSTICES 5 the federal courts. These too, by law, are Governor-in-Council appointments, but when the Justice Minister brings names to the cabinet, cabinet ministers can weigh in and argue for or against a proposed appointment especially if it is to a vacancy on a court in their province. Appointments to the Supreme Court, on the other hand, like the selection of Governors General, provincial Lieutenant Governors and Territorial Commissioners, have been considered so important that the Prime Minister must be the key and final decision-maker in selecting the appointee.
It is essential to note that the process I have described was never written into the Supreme Court Act or any other legal instrument. The whole process, including the Prime Minister’s role, remains in the informal, so-called “unwritten” part of our constitutional system. Usually we call these rules and practices constitutional conventions. Although, given the variations in the consulting part of the process, calling the process a convention implies more coherence and consistency than has occurred. The one constant of this conventional practice has been the unbridled discretion of the Prime Minister in deciding who will serve on the Supreme Court of Canada.
2017 CanLIIDocs 175
Reform in the modern ea
The defeat of the Charlottetown Accord in 1992 brought an end to efforts to reform the process of Supreme Court reform by constitutional amendment. But well before then, in the 1980s, there was strong interest in professional and academic organizations in reforming the judicial selection process not only for the Supreme Court but for section 96 courts and the Federal Court. The impetus for this, in part, came from new methods of judicial selection introduced by many provinces and Yukon, that aimed at removing political patronage as the dominant influence on appointments to provincial and territorial courts, the lowest trial courts in the court hierarchy but the courts where most Canadians have their first hand experience with the administration of justice. Again, though there were institutional variations across the jurisdictions, reform involved the introduction of independent nominating bodies that would recommend lists of candidates from which government would be required to choose. The aim was to make professional merit rather than political affiliation the primary criterion of selection. In 1987, the Canadian Bar Association published a report critical of “undue political favouritism” in the federal judicial appointment system and advocated reform along the lines of reform at the provincial level.4 The Association of Canadian Law Teachers, much earlier than this, championed reforms to change judicial selection from a patronage-ridden system to one based on a search for excellence.
In 1988, the Mulroney government responded to this pressure by introducing Judicial Appointment Advisory Committees to assist it in filling vacancies on section 96 courts, with a committee for each province, and for federal courts (the Canadian Tax Court, the Court Martial Appeal Court and the Federal
4 Canadian Bar Association, The Appointment of Judges in Canada (Ottawa: Canadian Bar Foundation, 1985). 6 UNBLJ RD UN-B [VOL/TOME 68
Court of Canada).5 But these advisory committees, unlike their provincial counterparts, were not nominating bodies. They were (and still are) screening bodies asked to respond to lists of legally qualified candidates sent to them by the Commissioner of Federal Judicial Affairs. So long as the committees could respond by designating some on their list as “highly recommended” rather than simply “recommended”, considerations of merit could enter into the advice they gave government. But when the Harper government reduced the committees’ mandate to advising simply whether a candidate was qualified or not qualified, considerations of merit was removed from the federal advisory process. The advisory committee system never applied to the selection of Supreme Court justices.
By the 1990s in the context of appointing Supreme Court of Canada justices, the importance of the Court’s role in interpreting the Charter of Rights and Freedoms had supplanted federalism as the focus of concern. The politicians and the public now knew that whomever gets appointed to the Supreme Court of Canada has tremendous power on controversial matters of great interest to the public, such as abortion, gay rights, prostitution, police powers and criminal justice. There was growing interest in a system of selection that was open and known, and not entirely 2017 CanLIIDocs 175 subject to the whims of the Prime Minister.
The first glimmer of reform came in 2004, when – out of the blue – the Martin government announced that Justice Minister Irwin Cotler would appear before an ad hoc committee of seven MPs and two members of the bar to answer questions about the persons the government had chosen to fill two Ontario places on the Supreme Court, Rosalie Abella and Louise Charron. The event was a political flop. The two Conservatives on the committee complained about being asked to rubber-stamp persons the government had already decided to appoint. And indeed Abella and Charron were appointed. The Globe & Mail’s lead editorial declared the process a “sham”.6
For their next Supreme Court appointment, in 2005 – to fill the vacancy created by the retirement of Justice Major of Alberta - the Liberals introduced a more elaborate process. This time the Minister of Justice, Irwin Cotler, sent a list of eight candidates to a nine-person Advisory Committee consisting of four MPs (one from each party), a retired judge chosen by the Canadian Judicial Council, a lawyer chosen by the law societies of the Prairie provinces, a representative of the three Prairie provincial governments, and two lay persons of “integrity and distinction” from the Prairie region chosen by the Justice Minister. The committee’s mandate was to assess the persons nominated by the minister and winnow the list down to three. The Minister of Justice would recommend one of these three to the Prime Minister. If that person were not appointed, Mr. Cotler would offer an explanation to the House of Commons Justice Committee.7
5 Peter H Russell & Jacob S Ziegel, “Federal Judicial Appointments: An Appraisal of the First Mulroney Government’s Appointments and the New Judicial Advisory Committees” (1991) 41 U Toronto LJ 4.
6 Donald R Songer, The Transformation of the Supreme Court of Canada (Toronto: University of Toronto Press, 2008) at 16–17.
2017] SELECTING SUPREME COURT JUSTICES 7
This advisory committee, it should be noted, was not a nominating committee, but a committee to select the best among the government’s nominees. It should also be noted that implicit in the structure of the new process – the legal and provincial government representation – was observance of the convention of regional representation. Since there was already a British Columbia jurist, Beverley McLachlin, on the court, the appropriate governments and law societies to be represented were those of the Prairie provinces.
The process was completed by January 2006. But by then the country was in the midst of a federal election resulting in the Harper Conservatives forming a minority government. Vic Toews, the new Conservative Justice Minister, announced that Prime Minister Harper had chosen to appoint Justice Marshall Rothstein, a Federal Court judge with a professional career in Winnipeg and one of the three names submitted by the Advisory Committee. So the Harper Conservatives completed the process introduced by the Liberals. But they did a little more than that: they added a new wrinkle. Marshall Rothstein, the designated new Supreme Court justice, would be interviewed by an ad hoc committee of MPs on live television. The committee would be chaperoned by law professor Peter Hogg, to make sure its 2017 CanLIIDocs 175 questioning did not venture into the justice-designate’s views on any issues that might come before the Court. It was a sort of “getting to know you” session for the politicians and the people.
This new process with the Conservatives add-on was a pale imitation of the American system of filling Supreme Court vacancies. As in Canada, in the United States, it is the government of the day that does the nominating, and then elected legislators who react to the government’s choice. The big difference, of course, is that in the American system the Senate has the constitutional power to advise and consent (or refuse consent) for the President’s nominee, while Canadian parliamentary committees have no power to reject the Prime Minister’s chosen candidate.
After the Rothstein appointment, the Harper government dropped the process the Liberals had established in 2005 and casually, from time to time, used truncated versions of what it had added to the Liberal process – small committees of MPs, always with a government majority, reviewing, in private, the person or persons the Prime Minister was considering for appointment. Some variant of this process was used to fill five vacancies, but not for the other three appointments that the Harper government made.8 With so much adhocery, and no checks and balances on the government’s nominating and selection process, it could not be said that up to 2016 Canada had reformed its system of selecting Supreme Court justices.
7 Peter H Russell, Rainer Knopff, Tom Bateman & Janet Hiebert, eds, “Introduction,” in The Court and the Constitution: Leading Cases (Toronto: Emond Montgomery, 2008) at 15.
8 Erin Crandall & Andrea Lawlor, “Courting Controversy: The House of Commons’ Ad Hoc Process to Review Supreme Court Candidates” (2015) 38:4 Can Parliamentary Rev 35 at 38.
8 UNBLJ RD UN-B [VOL/TOME 68
The Trudeau government’s new process
With this international and Canadian background of reform experience as context, let me now examine the new process the Trudeau government has put in place. At the centre of the process is the Advisory Board for Supreme Court Appointments.9 This new institution is most definitely a nominating body. It has a mandate to “actively seek out qualified candidates” and encourage them to apply and to provide the Prime Minister with “recommendations of at least three, but up to five, qualified and functionally bilingual candidates” for his or her consideration.10 The position is to be advertised, so that lawyers and judges who are interested can apply, but the Advisory Board can seek out candidates who it thinks are promising and encourage them to apply, as can the federal, provincial and territorial governments.
The Board has seven members, four of whom are chosen by judicial and legal organizations. The one judicial member is Richard Scott, retired Manitoba Chief Justice, named by the Canadian Judicial Council. The three lawyers are Susan Ursel from Toronto, named by the Canadian Bar Association, Jeff Hirsch of
Winnipeg, named by the Federation of Canadian Law Societies and Camille 2017 CanLIIDocs 175 Cameron, a legal scholar and Dean of the University of Dalhousie’s Schulich School of Law, named by the Council of Canadian Law Deans. The other three members, two of whom are non-lawyers, are nominated by the Minister of Justice. They are Lili-Anne Peresa, President and Executive Director of Centreaide of Greater Montreal, Stephen Kakfiw, former premier of the Northwest Territories and President of the Dene Nation, and Kim Campbell, former Canadian Prime Minister and federal Justice Minister in the Mulroney government. Campbell is the Board’s chair.
With one qualification, the Advisory Board’s membership measures up reasonably well to the norms of Canadian diversity. Women are in the majority outnumbering male members four to three. There is at least one member from each of what counts as Canada’s “regions” in filling Supreme Court vacancies. The Board’s ethnic diversity represents the Canada’s multiculturalism, but while there is one Aboriginal member, there is no one from French Canada; as a result, it fails to reflect Canada’s multinational character.11
9 Prime Minister of Canada, “Prime Minister announces new Supreme Court of Canada judicial appointments process” (2 August 2016), online:
10 Office of the Commissioner for Federal Judicial Affairs Canada, “Terms of Reference of the Advisory Board”, (2 August 2016), online:
11 For an account of how Canada has become a multinational, multicultural country, see Peter H Russell, Canada’s Odyssey: A Country Based on Incomplete Conquests (Toronto: University of Toronto Press, forthcoming in 2017). 2017] SELECTING SUPREME COURT JUSTICES 9
With only two non-lawyers the Board might seem to be over-dominated by legal professionals. In this respect it contrasts sharply with Ontario’s Judicial Appointments Advisory Committee (of which I was the founding chair) which since its beginning in 1989 has had a requirement that a majority of its 13 members be non-lawyers. I think that makes sense for selecting members of courts where so many citizens experience justice first-hand and whose justices frequently do not show sufficient respect for the people who appear before them or for their family members and friends in the courtroom.
The new Canadian Advisory Board is more like its counterpart in the United Kingdom. The five-person commission that makes recommendations on appointments to the UK’s Supreme Court consists of the Court’s President (its Chief Justice), deputy president and three members of the appointing committees or boards established for Scotland, Northern Ireland and lower courts in England.12 There is no requirement that any of those three be non-lawyers. The United Kingdom, I should note, is at the extreme end of the reform spectrum in trying to rid the judicial appointment process of the influence of political patronage and political ideology.
Fortifying judicial independence was the overriding goal of the reforms that took 2017 CanLIIDocs 175 place under Tony Blair’s Labour government and that have been carried on under Conservative administrations. The UK appointing commission gives only one name to the Minister of Constitutional Affairs who must give reasons to parliament if he or she does not accept the “recommendation”.
At the other end of the spectrum are South Africa’s 23-person Judicial Services Commission, 11 of whose members are elected politicians, three of whom must be from opposition parties,13 and Israel’s nine-person committee for judicial appointments and promotions, three of whom are elected politicians, traditionally one from the opposition plus two ministers.14 Both these judicial selection bodies serve as nominating bodies not only for their country’s highest constitutional court but for judges of all other courts (excluding South African magistrates).
The absence of any MPs from the Canadian Board’s composition drew some initial outrage from Conservative politicians. But elected politicians are by no means excluded from the new process. Once the Committee has settled on its short list, the Minister of Justice is to consult with the Chief Justice of Canada, relevant provincial and territorial attorneys general, relevant cabinet ministers and opposition justice critics, as well as the relevant House and Senate committees. After the Prime Minister makes his or her selection from the short list submitted by the Advisory Board, the Minister of Justice and Chairman of the Advisory Committee will appear
12 See Kate Malleson, “The New Judicial Appointments Commission in England and Wales: New Wine in Old Bottles?” in Kate Malleson & Peter H Russell, eds, Appointing Judges in an Age of Judicial Power: Critical Perspectives from around the World (Toronto: University of Toronto Press, 2006) 39 at 46.
13 François du Bois, “Judicial Selection in Post-Apartheid South Africa”, in Malleson & Russell, supra note 12, 280 at 284.
14 Eli M Salzberger, “Judicial Appointments and Promotions in Israel: Constitution, Law and Politics”, in Malleson & Russell, supra note 12, 241 at 248.
10 UNBLJ RD UN-B [VOL/TOME 68 before the House Standing Committee on Justice and Human Rights to explain how the chosen nominee meets the statutory requirements and the criteria. Further to that, the nominee is to participate in a moderated question and answer session with members of the House Standing Committees on Justice and Legal Affairs and the Senate Standing Committee on Legal and Constitutional Affairs.15 Representatives of the Bloc Québécois and the Green Party are to have an opportunity to participate in this part of the process. This means that parliamentarians will at least get to meet with the justice-designate, an opportunity not extended to them in the on and off private meetings with MPs under the Harper government. My own view is that MPs are not likely to have the time or capacity to engage in the work of looking for the best candidates for service on the Supreme of Canada. Those who do know or have heard of outstanding lawyers or judges whom they would like to see considered can encourage them to apply. I fear that at least some MPs would be too driven by partisan or ideological considerations if they were involved in assessing the pool of candidates that the Board assembles.
The reference to consulting on the short list with relevant provincial and territorial attorneys general held out a glimmer of hope that the new process would 2017 CanLIIDocs 175 observe the convention of regional representation on the Supreme Court bench. Politically speaking, that glimmer of hope was clearly not enough to overcome unease in Atlantic Canada.
Let me now discuss the stated criteria for appointment from which regional representation is missing. The Advisory Board must be guided by two institutional criteria: the Supreme Court should be “gender balanced” and “reflect the diversity of Canadian society.”16 The appointment of a fifth male to a bench of four men and four women cannot reasonably be said to upset the gender balance. As for diversity, except for the absence of a justice with an Aboriginal background, the existing Supreme Court bench measures up fairly well. Two of the Ontario justices, Rosalie Abella and Michael Moldaver, have a Jewish Eastern European background, and the third, Andromache Karakatsnasis, is from Canada’s Greek Eastern Orthodox community. This means that justices who are neither British or French in their ancestry form over 30 per cent of the existing bench, which is roughly in line with their proportion of Canada’s current population. But while that may satisfy the multicultural dimension of Canadian society, it does not serve Canada’s multinational structure. Aboriginal peoples should not be thought of as minority cultures. They are nations within, with governments of their own and their homelands in Canada. It has taken a long time for the country to accept this fact and build recognition of it into Canada’s Constitution. The Supreme Court plays a huge role in adjudicating Aboriginal rights cases. The absence of an Aboriginal jurist with life experience and deep knowledge of Aboriginal law and tradition is a major flaw in the Court’s present composition.
15 Office of the Commissioner for Frederal Judicial Affairs Canada, “Frequently Asked Questions” (2 August 2016), online:
16 FJA, “Terms of Reference”, supra note 10.
2017] SELECTING SUPREME COURT JUSTICES 11
It may very well be that it was the Advisory Board’s effort to find a well- qualified Aboriginal person that prompted Mary-Ellen Turpel-Lafond, an outstanding Aboriginal scholar and jurist from western Canada, to make it clear that she would not be available to fill the vacancy created by Justice Cromwell’s retirement. “I never go where I am not wanted,” Turpel-Lafond told The Globe and Mail. 17 In making this statement she recognized and honoured regional representation for Atlantic Canada on the Supreme Court, and did the country a favour in not making one representational priority trump another. It would be very difficult to find a well-qualified, bilingual, Aboriginal jurist in Atlantic Canada.
Let me now turn to bilingualism, which is set out in the new policy almost as if it were a legal requirement of eligibility for appointment, which of course it is not. Much progress has been made in making the Supreme Court a bilingual institution since I did my study of the Court for the Royal Commission on Bilingualism and Biculturalism in the 1960s.18 Then, I found that with no instantaneous translation facility and most justices from English-speaking Canada being unilingual, Francophone advocates from Quebec who wanted to present their case to the Court in their first language were at a serious disadvantage. Equally 2017 CanLIIDocs 175 unacceptable was the fact that the Court’s Official Reports contained English versions of all decisions but did not report all decisions, including some constitutional decisions, in the French language. Both of those lamentable institutional defects, as a result of the Commission’s work, were soon remedied. Is the country now ready to take the next step and require that all who sit on its highest court be functionally bilingual?
The definition of functionally bilingual on the Commissioner for Federal Judicial Affairs Canada website requires that all members of the Court be able to read written submissions and hear oral arguments without translation services, but justices can speak in their first language when questioning counsel in the oral hearing – a reasonable modification, I would say.19 Given the size of the pool of lawyers and judges from which Supreme Court justices are drawn and the popularity of French immersion over many decades in English-speaking Canada, I do not think the requirement of functional bilingualism will unduly restrict the availability of outstanding candidates to fill positions on the Court. The Deputy Commissioner of Federal Judicial Affairs told The Globe & Mail that Malcolm Rowe surpassed the level of competence required.20 The one qualification I would make is that English/French bilingualism should not be a requirement for an outstanding
17 Sean Fine, “Trudeau waffles on approach to appointing top judge”, The Globe & Mail (28 September 2016) A1 [Fine, “Trudeau waffles”].
18 Peter H Russell, The Supreme Court of Canada as a Bilingual and Bicultural Institution (Ottawa: Queen’s Printer, 1969).
19 Office of the Commissioner for Frederal Judicial Affairs Canada, “Qualifications and Assessment Criteria” (2 August 2016), online:
20 Sean Fine, “PM makes judicial activist first Supreme Court nominee”, The Globe & Mail (17 October 2016) A1.
12 UNBLJ RD UN-B [VOL/TOME 68
Aboriginal candidate who is already fluent in his or her native tongue and one of Canada’s official languages.
The new process directs the Advisory Board to observe one further institutional criterion: there should be a balance of public and private law expertise among the Court’s members.21 In Canada, like other countries with institutions based on the common law Westminster tradition, our highest court is supreme in all areas of law. That is not true in the United States, other Commonwealth countries or civil law countries. That means the Canadian Supreme Court must have strength in all fields of law. That is why it makes such good sense to give the Chief Justice of the Court an opportunity to be consulted on the Advisory Board’s short list. This will enable her or him to inform the Board on how candidates they are recommending serve the functional needs of the Court in various areas of law.
Criteria that speak to institutional merit are not to be satisfied at the expense of individual merit. The criteria posted online set out, more thoroughly than any writing, official or unofficial, I have ever seen the personal and professional qualities that are expected of a Supreme Court justice. They include, deep and 2017 CanLIIDocs 175 demonstrated legal knowledge (“the chief consideration”), analytical skill, clarity of expression, commitment to public service, moral courage, independence of mind, and personal integrity.22 My brief summary does not do justice to the care with which this statement of personal qualifications has been written. This part of the new process makes it clear that the Advisory Board is being asked to look in the large pool of lawyers who are legally qualified to sit on the Court – meaning they have ten years of practice at the bar or service on a superior court – for truly outstanding candidates. It is indeed a search for excellence.
When Kim Campbell, Chair of the Advisory Board, appeared with Justice Minister Jody Wilson-Raybould before a panel of parliamentarians on October 24, 2016 to explain the process that led to the selection of Malcolm Rowe, Canadians learned much more about the new process than had been disclosed up to that time.23 Campbell reported that the Advisory Board had received 31 applications from across Canada. She said the Board had actively sought out candidates who, if they wished to be considered, would still be required to apply. We don’t know how many of the 31 applicants were persons that the Board sought out and how many applied on their own initiative. Applicants were required to set out their qualifications and reasons for being interested in serving on Canada’s highest court. Transparency was well served by giving the public an opportunity to read the selected candidate’s presentation of his qualifications. The Board examined applicants’ legal writings (presumably, in the case of judicial applicants, that would include their judicial opinions), as well as professional and community references. It also conducted hour-long interviews with the top ten candidates. “Our goal,” said Ms Campbell, “was to create a list that would
21 FJA, “Qualifications”, supra note 19.
22 Ibid.
23 Tonda MacCharles, “Ex-PM backs Rowe despite ruling”, Toronto Star (25 October 2016) A7. 2017] SELECTING SUPREME COURT JUSTICES 13 keep the prime minister up at night trying to figure out which one of these excellent people to appoint.”24
The day after the explanation of the process to the House committee, the justice-designate appeared before a panel of MPs and Senators and an audience of 150 University of Ottawa law students. This was not a carefully managed question and answer session like the one which Marshall Rothstien underwent in 2006. Conservative Senator Denise Batters and NDP leader Thomas Mulclair asked sharp questions about Rowe’s decision as a Newfoundland and Labrador Court of Appeal judge not to overturn an acquittal of a defendant in a rape case despite serious errors by the trial judge in allowing in evidence about the victim barred by rape-shield legislation.25 Some Conservatives questioned Rowe on the statement he had made in his written application that judges in their adjudicative work inevitably “make law.” That idea, which has long been accepted by most serious students of the judicial process, was a “no-no” for Harper’s Conservatives. In his response, Rowe did not back away from acknowledging the creative aspect of adjudication, especially at the Supreme Court level, but expressed sensitivity to the need for caution in developing fields of law, such as Aboriginal rights. The Court, he said, must “not get ahead of 2017 CanLIIDocs 175 governments and indigenous leaders who bear the prime responsibility of engaging in “negotiations and dialogue” and bringing along the public.”26 This part of the new process was an informative encounter that gave the country a fair indication of the direction in which this new member of their highest court will develop Canadian law, but it was not a “nomination hearing” as some media described it. One journalist wrote that a House of Commons committee would meet the next day “to indicate whether it supports the nomination…”27 Unlike the United States where the Senate must confirm the President’s nominee, the Prime Minister’s nominee does not require confirmation by Canada’s Parliament. On October 28, Prime Minister Trudeau announced Malcolm Rowe’s appointment.28
Regional Representation
Having given an overview of what the new process calls for, let me now turn to the one point on which it has nothing to say – regional representation. The composition of the Supreme Court from its establishment in 1875 has had a pattern of regional
24 Ibid.
25 See Alyshah Hasham, “New Supreme Court nominee under fire for rape trial ruling”, Toronto Star (21 October 2016) A1.
26 Tonda Maccharles, “Supreme Court nominee Malcolm Rowe surprises observers in questioning”, Toronto Star (25 October 2016), online:
27 Ibid.
28 “Justice Malcolm Rowe formally appointed to Supreme Court of Canada”, The Canadian Press (28 October 2016), online:
Representation of other regions has always depended on constitutional convention. But constitutional convention has not been a tightly defined set of rules. The first Supreme Court had two Ontario justices and two from the Maritimes (New Brunswick’s Justice Ritchie and Nova Scotia’s Justice Henry) along with the two mandatory Quebec appointees. I assume the west was left out because the fledgling provinces of Manitoba and British Columbia did not yet have enough legal talent for a Supreme Court appointment. The west got its first justice with the appointment of Justice A.C. Killam in 1903. Maritime jurists filled two places on the Court until 1906, when Charles Fitzpatrick, a Quebecer and Wilfrid Laurier’s Justice Minister, had himself appointed to fill the place vacated by Justice Sedgwick of Nova Scotia. When Chief Justice Davies, Prince Edward Island’s only Supreme Court member ever (and a former premier of the province), left the Court in 1924, he was not 2017 CanLIIDocs 175 replaced by a Maritimer. From 1924 until the appointment of New Brunswick’s Justice Crocket in 1932, there was no Maritimer on the Supreme Court bench, even though it had been expanded to seven members in 1927. When Justice Rand joined the Court in 1943, the Maritime contingent was back to two. Justice Crocket left the Court in 1948, leaving Rand as the only Maritime justice.
Finally, from 1949 until today, the pattern of regional representation, with one exception, has remained in place: three from Quebec, three from Ontario, two from the west and one from Atlantic Canada. The one exception occurred in 1979, when the vacancy created by the retirement of Ontario Justice Wishart Spence was filled by Justice William McIntyre from the British Columbia Court of Appeal. Joe Clark’s Progressive Conservative government had secured Ontario’s consent for this temporary departure from the standard pattern. The one exception, I think, proves the rule. Ontario is the only province that might be able to accept a reduction of its representation on the Supreme Court. Even then, Prime Minister Clark was wise to get Ontario to accept the appointment of a British Columbia judge to one of its places on the Court, and Ontario was gracious to accede to the request in order to accommodate a province that had not had a justice on the Court since 1962 and tends to see itself as a fifth region of Canada rather than just one of four western provinces. I cannot remember a whimper of protest from the Ontario people.
Failure to fill the existing vacancy on the Supreme Court with a jurist from Atlantic Canada turned out to be politically unthinkable for Prime Minister Trudeau. The reason for regional representation on the Supreme Court is not functional. Atlantic Canada does not have a distinct body of law that requires a
29 See Peter H Russell, The Judiciary in Canada: The Third Branch of Government (Toronto: McGraw Hill Ryerson, 1987) at 138–139, Table 5.1.
30 Supreme Court Act, RSC 1985, c S-26, s 6. 2017] SELECTING SUPREME COURT JUSTICES 15 justice with knowledge of it. The reason for the constitutional convention is political justice in a country that has always had deep sectional divisions. The sense of injustice in Canada’s smallest region which, if the convention had not been observed, would not just have had its representation on the Court reduced but eliminated, was clearly intense. The politicians and people of Atlantic Canada made that point clear. So did a unanimous House of Commons, including its Liberal caucus. The Prime Minister was listening.
The Trudeau government did give the assurance that it was “committed to include candidates from Atlantic Canada on the short list for the position.”31 If Justin Trudeau had let us down and had not selected an Atlantic Canadian person from that list, what would the consequences have been? In the 1981 Resolution to Amend the Constitution,32 the Supreme Court of Canada made it clear that courts can identify a convention, but do not enforce convention. The penalties for breach of a constitutional convention are political. In this case, the penalty for the Liberal Party in the next election would have been significant. Political support that was so quickly garnered can just as quickly be withdrawn.
2017 CanLIIDocs 175
Is the New Process Constitutional?
Concern about the constitutionality of the new process for selecting Supreme Court justices stems from the Supreme Court’s decision in the 2014 Senate Reference case. In that case the Court found the Harper government’s legislation to have Senators selected though consultative provincial elections unconstitutional. A key rationale for this finding was that such a change “would fundamentally alter the architecture of the Constitution”.33 The Court went on to say that the “entire process by which Senators are selected” is subject to the general procedure for amending the Canadian Constitution – the federal Parliament plus seven provinces representing 50 per cent of the population.34 That language sounded so sweeping that some constitutional scholars suggested it would rule out changes in how Supreme Court justices or Senators are selected even if the changes are effected through informal means by modifying constitutional conventions rather than through federal legislation such as the Harper government tried to use for Senate reform.35
I think it is most unlikely that the new procedure of appointing Supreme Court justices, if challenged in the courts, would be found to be unconstitutional. The new process is an addition to the constitutional convention that the Prime Minister
31 Fine, “Trudeau waffles”, supra note 17.
32 Resolution to Amend the Constitution, [1981] 1 SCR 753, sub nom Reference Re Amendment of the Constitution of Canada (Nos 1, 2 and 3), 125 DLR (3d) 1.
33 Reference re Senate Reform, 2014 SCC 32 at para 54, [2014] 1 SCR 704.
34 Ibid at paras 64–65.
35 Dennis Barker & Mark D Jarvis, “The End of Informal Constitution Change in Canada?” in Emmett MacFarlane, ed, Constitutional Amendment in Canada (Toronto: University of Toronto Press, 2016) 185. 16 UNBLJ RD UN-B [VOL/TOME 68 advises the Governor-in-Council on whom to appoint to the Court. The purpose of conventions in our constitutional system is to provide guidelines or rules on the proper use of legal powers. The new judicial selection process limits the discretion of the Prime Minister to advise who should be appointed. It is designed to improve the information on which the Prime Minister’s choice is based and make it more accountable.
The Supreme Court reform is in line with similar reforms to the Prime Minister’s role in making vice-regal appointments and filling Senate vacancies. In 2012, with very little fanfare, Prime Minister Harper established the Advisory Board on Vice-Regal Appointments to assist him in selecting Canadians for appointment to the positions of Governor General, provincial Lieutenant Governors and Territorial Commissioners.36 Last December, the Trudeau government established an Advisory Board to assist the Prime Minister in selecting Canadians to be summoned by the Governor General to fill Senate vacancies.37 These new advisory bodies, like the new process for selecting Supreme Court justices, aim at reforming selection processes that up to now appear to have been dominated by political patronage by making them merit-based and more accountable. All three of these recent reforms have not been 2017 CanLIIDocs 175 put into legislation but remain in the informal conventional part of our constitution. I think it most unlikely that a majority of the Supreme Court would find any of them unconstitutional.
One further reform would enhance these informal changes in the selection of vice-regal office-holders, Senators and Supreme Court of Canada justices. This is the adoption of a reform carried out in New Zealand and Great Britain aimed at making the so-called “unwritten” rules, practices and principles of their constitutions publicly accessible in a succinct, online succinct statement of them, called Cabinet Manuals. I have been a champion of bringing this reform to Canada.38 Strong support for the idea has come from members of the Canadian Bar Association, the Canadian Political Science Association, leading journalists, as well as from senior members of all five of our parliamentary parties. I am hopeful that Justin Trudeau’s government will take up the idea and make it a sesquicentennial project. Nothing could do more to improve the constitutional literacy of Canadian citizens.
36 “New panel to ensure ‘non-partisan’ vice regal appointments”, The Canadian Press (5 November 2012), online:
37 Government of Canada, “Government Announces Immediate Senate Reform” (3 December 2015), online:
38 Peter H Russell & Cheryl Milne, Adjusting to a New Era of Parliamentary Government: Report of a Workshop on Constitutional Conventions (Toronto: Asper Centre for Constitutional Rights, 2011). 2017] SELECTING SUPREME COURT JUSTICES 17
Conclusion
Let me close by giving a positive answer to the question posed in the title of this lecture. Yes, Justin Trudeau’s new sunny way of selecting Supreme Court justices is a better way than anything we have had before or than has been proposed in the past. It measures up to the best reforms in selecting provincial and territorial judges and to reforms instituted in other constitutional democracies, though it is sui generis, as are all those other efforts at eliminating the unbridled discretion of elected leaders in staffing the courts. Every jurisdiction expresses its own constitutional culture and experience in reforming the way judges are selected. Let us hope that the reformed process for selecting Supreme Court justices is a harbinger of sunny ways changing the process of filling vacancies in the provincial and territorial superior courts and courts of appeal, and the federal courts. Converting that process to a merit-based system and one that is not totally controlled by federal politicians promises even more benefits for Canadians than reform at the Supreme Court level.
There is certainly room for improving the new process of selecting Supreme
Court justices. This first use of the process should lead to some changes, especially 2017 CanLIIDocs 175 including the regional representation convention as one of the stated criteria. If the Trudeau government does this, it should ask the Advisory Board to confine its next search to the appropriate region rather than conducting a national search as it did in the first application of the new process. One of the great benefits of practices of government being regulated by convention is that they are much easier to change than legislation or the written Constitution.
The Advisory Board, I am pleased to see, is in place for five years. No adhocery here! This means it will be involved in replacing Chief Jurstice Beverley McLachlin who is now 73, and possibly Justice Abella who is in her 71st year, as well as other justices who, like Thomas Cromwell, decide to retire before reaching the mandatory retirement age of 75. The new process will have a work out over the next few years. If, following the 2019 election, a new party takes power in Ottawa, it will have discretion to decide whether to follow the new process, perhaps with some modifications, or not. That is the nature of constitutional conventions: governments are not bound by them unless they think they should be. That is their weakness. But their strength is their sensitivity to the changing winds of democratic politics. I have a hunch that the Advisory Board on Supreme Court Appointments, probably with some ongoing modifications, is here to stay.
REFORMING JUDICIAL APPOINTMENTS: CHANGE AND CHALLENGE
Rosemary Cairns Way*
Introduction
The Ivan C. Rand Memorial Lecture Series has a history of topicality.1 On October 20, 2016, Professor Peter H. Russell delivered the 23rd lecture,2 a thoughtful analysis of the reformed Supreme Court of Canada selection process announced by Prime Minister Trudeau in August 2016. I was delighted to participate as a member of the discussant panel. Three days before the lecture, the Prime Minister announced the nomination of Justice Malcolm Rowe to fill the seat vacated by the Honourable Justice Thomas Cromwell. The nomination provided concrete fodder for the lecture and discussion, particularly with respect to two issues which garnered public 2017 CanLIIDocs 175 attention in the lead-up: the requirement of functional bilingualism, and the immediate fate of the convention of regional representation. The icing on the cake came on the day of the lecture, when the federal government announced substantial changes to the federal judicial appointment process. These changes were intended, in their words, to “increase the openness, transparency, accountability, and diversity of Canada’s judiciary.”3 It was hard not to wonder if the lecture planners had access to insider knowledge.
In his 2008 Rand lecture, “Judicial Appointments, Democratic Aspirations, and the Culture of Accountability”,4 Professor Lorne Sossin noted the historic complacency which has surrounded judicial appointment in Canada. For many, a tradition of judicial excellence meant that “fixing judicial appointment truly is a solution in search of a problem.”5 In his lecture, Professor Sossin challenged this
* Professor, Faculty of Law, University of Ottawa.
1 Recent iterations of the lecture have addressed Indigenous rights and cyberbullying. See John Borrows, “Unextinguished: Rights and the Indian Act” (2016) 67 UNBLJ 3, and A Wayne MacKay, “Law as an Ally or Enemy in the War on Cyberbullying: Exploring the Contested Terrain of Privacy and Other Legal Concepts in the Age of Technology and Social Media” (2015) 66 UNBLJ 3.
2 Peter H Russell, “Selecting Supreme Court Justices: Is Trudeau’s Sunny Way a Better Way?” (2017) 69 UNBLJ 3.
3 See online:
4 (2008) 58 UNBLJ 11.
5 Ibid at 12. 2017] REFORMING JUDICIAL REVIEW 19 view, arguing that the system of appointment was both “inconsistent with the independence of the judiciary” and out-of-step with contemporary political norms of transparency and accountability. 6 The timing of his critique was not a coincidence. In 2006, Prime Minister Stephen Harper entered office with a promise to bring transparency and accountability to judicial appointments. In this, he was capitalizing on reforms to the Supreme Court process initiated by the previous liberal government,7 as well as reforms to the section 96 process introduced by Prime Minister Mulroney in 1988. Prime Minister Harper’s commitment to reforming Supreme Court appointments was inconsistent and ultimately ad hoc,8 but he did make relatively dramatic shifts in both the composition and powers of the Judicial Advisory Committees (JACs) which screened section 96 appointments.9 These changes were controversial and were subjected to sustained critique by the legal academy, the legal profession, policy makers and even the judiciary.10 The debate appears to have catalyzed the far-reaching reforms to judicial appointments announced by the Trudeau government in August and October of 2016. There is no doubt that the politicization of judicial appointment has not ended with the defeat of the Harper government.11 Rather, the politics have shifted. The question now is whether the current reforms are cut from the same (but differently patterned) cloth, 2017 CanLIIDocs 175 or whether they are in fact more consistent with our democratic aspirations and constitutional commitments. In my view, the answer to this will depend on the nature of the constraints these reforms impose on the mostly unfettered executive prerogative to appoint judges. Do they, in theory and in practice, tend to enhance the constitutional guarantees of judicial independence and impartiality? This is a big question, and one which cannot be answered this early. My present aim is far more modest; I intend to offer a preliminary assessment of the ways in which the current reforms are informed by and consistent with a commitment to judicial diversity. 12
6 Ibid.
7 See Professor Russell’s discussion of this history in this volume. For a comprehensive analysis see Adam Dodek, “Reforming the Supreme Court Appointment Process, 2004-2014: A Ten Year Democratic Audit” (2014) 67 SCLR (2d) 111, online:
8 Ibid.
9 See my discussion of these changes in Rosemary Cairns Way, “Deliberate Disregard: Judicial Appointments under the Harper Government” (2014) SCLR (2d) 43 at 55–59, online:
10 See e.g. Carissima Mathen, “Choices and Controversy: Judicial Appointments in Canada” (2008) 58 UNBLJ 52; Sossin, supra note 4. Even the Canadian Judicial Council participated in the debate; see Canadian Judicial Council, News Release, “Judicial Appointments: Perspective from the Canadian Judicial Council” (20 February 2007), online:
11 For a fascinating range of views on the impact of the Harper government on the politics of judicial appointments, see “Stephen Harper and the Judiciary”, Policy Options, online:
12 The language we use is important, and the language of diversity carries political and ideological baggage. The work of Professor Sonia Lawrence is especially instructive here. She suggests that representation may be a more useful term which “more squarely confronts the ways in which a homogenous — or otherwise non-representative — bench threatens impartiality, by calling attention to the disparity between the judges and the judged.” See Sonia Lawrence, “Reflections: On Judicial Diversity 20 UNBLJ RD UN-B [VOL/TOME 68
There is no doubt that they are a far cry from the “deliberate disregard of diversity” 13 displayed by the former government.
There are three parts to my discussion. First, I explain why diversity matters to judging. Second, I consider how diversity ideals inform the new Supreme Court of Canada appointment process, and examine the conflicting challenges presented by a commitment to diversity measured on axes of region, language and identity. Third, I briefly examine the ways in which the section 96 reforms reflect the same public commitment to diversity.
Why Diversity Matters
The last decades have seen the emergence of a remarkable professional and intellectual consensus on the importance of a judicial appointments process which takes account of diversity. In 2012, Chief Justice Beverley McLachlin publicly recognized the need for “a bench that better mirrors the people it judges.”14 In
August of 2013, the CBA reiterated its long-standing call for increased diversity on 2017 CanLIIDocs 175 the bench, pointing out that “the low number of women and members of racialized and other minority groups appointed to the federal courts does not reflect the gender balance or diversity in the Canadian population.”15 The urgent need for Aboriginal judges has been highlighted by the CBA and the Indigenous Bar Association,16 and the fact that this need persists at a time when there is a judicially acknowledged crisis of criminal justice legitimacy for aboriginal peoples17 makes it especially urgent.
Until October of 2016, the calls for change had little apparent impact on the federal appointments process. Canada’s federal judiciary remains overwhelmingly white and male, at the same time as Canadian society grows increasingly diverse.
and Judicial Independence” in Adam Dodek & Lorne Sossin, eds, Judicial Independence in Context (Toronto: Irwin Law, 2010) 193 at 207. I agree. Nevertheless, diversity is the language currently chosen by the governments.
13 Supra note 9.
14 The Right Honourable Beverley McLachlin, “Judging: the Challenges of Diversity” (Remarks delivered at the Judicial Studies Committee Inaugural Annual Lecture, Edinburgh, Scotland, 7 June 2012), online:
15 Canadian Bar Association, Resolution 13-04-A, “Equality in Judicial Appointments” (17 August 2013), online:
16 Canadian Bar Association, Resolution 05-01-A, “Recognition of Legal Pluralism in Judicial Appointments” (13 August 2005), online:
17 R v Gladue, [1999] 1 SCR 688, 171 DLR (4th) 385; R v Ipeelee 2012 SCC 13, [2012] 1 SCR 433. 2017] REFORMING JUDICIAL REVIEW 21
The number of women on the federal bench has crept upwards at a glacial pace.18 As of April 30, 2016, only 35% of the federal bench was female. Statistics on indigeneity and race are even more troubling. In a five-year study of federal appointments, from 2009-14, I concluded that Aboriginal judges were being appointed to superior courts at a rate of barely more than 1%, while visible minority judges were appointed at a rate of half that.19 Meanwhile, almost 20% of Canadians are members of visible minority communities. In large urban centres like Toronto and Vancouver, visible minorities account for almost 50% of the population. 20 Aboriginal peoples make up 4% of the Canadian population, and the population is growing.21 There is clear evidence that the demographics of the legal profession are changing, although the profession is not as diverse as the general population. 22 Nevertheless a substantial pool of exceptionally talented women, aboriginal, and visible minority lawyers are qualified for appointment.
Why does diversity matter? The significance of diversity to judging depends on a claim about who judges are – products of lived experience and what judges do – apply the law and exercise discretion. Even the Chief Justice has acknowledged that “a variety of subjective influences — our beliefs about the world 2017 CanLIIDocs 175 and about human nature, our emotions, and our sense of justice — are inescapably part of judicial decision-making.”23 The more diverse the bench, the better the
18 Kirk Makin, “Appointments of female judges slump under Harper's Tories” The Globe and Mail (11 November 2011), online:
19 Supra note 10 at 61–64. See also, Rosemary Cairns Way, “Words are not Enough”, Policy Options (5 October 2015), online:
20 Statistics Canada, Immigration and Ethnocultural Diversity in Canada, Catalogue No 99-010-X, online: <12.statcan.gc.ca/nhs-enm/2011/as-sa/99-010-x/2011001/tbl/tbl2-eng.cfm>.
21 Statistics Canada, Aboriginal Peoples in Canada: First Nations People, Métis and Inuit, Catalogue No 99-011-X, online: <12.statcan.gc.ca/nhs-enm/2011/as-sa/99-011-x/99-011-x2011001-eng.cfm#bx6>.
22 Michael Ornstein, Racialization and Gender of Lawyers in Ontario (Toronto: The Law Society of Upper Canada, 2010). The report concludes: “The legal profession in Ontario is changing dramatically. The number of lawyers who are women, Aboriginal and members of a visible minority continues to grow, transforming the face of a profession that until the early 1970s was primarily White and male. … Leading the transformation is an extraordinary increase in the percentage and number of women lawyers. Accounting for just 5 percent of Ontario lawyers in 1971, growth in the number of women lawyers has continued unabated for 35 years. In 2006 women accounted for nearly 60 percent of the youngest lawyers and 38 percent of all lawyers in Ontario. … In the last decade, gains in the representation of women are attributable largely to increased numbers of racialized women. Racialized women account for no less than 16 percent of all lawyers under 30, compared to just 5 percent of lawyers 30 and older; racialized men account for 7 percent of lawyers under 30, compared to 6 percent of lawyers 30 and older. The percentage of Ontario lawyers who were Aboriginal was unchanged between 1981 and 2001, but increased from 0.6 to 1.0 percent between 2001 and 2006.” In December of 2016, the Law Society of Upper Canada approved the Challenges Faced by Racialized Licensees Challenges Working Group’s final report, with 13 recommendations to address issues of systemic racism in the legal professions. Copies of the report and the recommendations are available online:
23 Supra note 16. 22 UNBLJ RD UN-B [VOL/TOME 68 quality of judgment. Increasing the range of perspectives and experiences among the judiciary will increase the likelihood of truly impartial judgment – judgment that does not unintentionally replicate the perspectives and values of a limited subset of human experience. A diverse bench increases the judiciary’s capacity to be both individually and structurally (institutionally) impartial.24 It also provides a public guarantee that appointments are animated by the constitutional norm of antidiscrimination. A homogenous bench suggests an appointments process which disproportionately denies opportunities to indigenous peoples, racialized individuals, women, and other members of equality-seeking groups. This is not a claim about intention. Rather it is a claim that executive discretion constrained only by an uncritical allegiance to merit has the potential to reinforce an unrepresentative status quo, while at the same time resisting substantive change. In the absence of transparency, the only way to assess the process is to observe the results. And when the results are an unrepresentative bench, the public has a right to worry that the institution charged with the delivery of impartial justice and the protection of the rule of law may be institutionally incapable of delivering on these promises. Justice, in a diverse society, is more likely to be both done and seen to be done, when the institution dispensing justice reflects that diversity. 2017 CanLIIDocs 175
Diversity Objectives and the New Supreme Court Process
When asked why his cabinet had equal numbers of men and women, Prime Minister Trudeau famously responded, “Because it’s 2015!” The Prime Minister has delivered a remarkably consistent message on the importance of equality and diversity since being elected. Ministerial mandate letters were made publicly available in November 2015.25 The mandate letters committed the government at large to “transparent, merit-based appointments” which would help ensure gender parity, and a better reflection of “Indigenous Canadians and minority groups in positions of leadership.”26 The Minister of Justice’s mandate letter made specific reference to the appointment of Supreme Court Justices, and on August 2, 2016, the government followed through on its promise by announcing a thoroughly revamped appointment process,27 which it used for the subsequent appointment of Justice Malcolm Rowe.28
24 The importance of structural impartiality is examined by Sherrilyn A Ifill, “Racial Diversity on the Bench: Beyond Role Models and Public Confidence” (2000) 57 Wash & Lee L Rev 405 at 411.
25 All of the letters are available online:
26 Each letter includes this commitment.
27 Online:
28 Online:
2017] REFORMING JUDICIAL REVIEW 23
Where, and how, do diversity ideals figure in this new process?29 As Professor Russell has explained, the centerpiece is the creation of an independent and non-partisan advisory board tasked with assessing applications and providing a short-list of candidates to the Prime Minister.30 The seven members of the Advisory Board represent the judiciary, the legal profession, the academy, and the public. The Minister of Justice nominates three public members, at least two of whom are from outside the legal community. The government describes the Board member selection process as attentive to “gender balance, diversity (including linguistic diversity), and regional balance,” and biographies of the Board members, who continue to serve for up to five years (renewable), demonstrably reflect these values. The purpose of diversifying the Board is to ensure that “diverse perspectives are brought to bear on the ultimate goal of identifying the best candidates.” The Board is required to make recommendations to the Prime Minister of no less than three and no more than five candidates, each of whom is “functionally bilingual” and who otherwise meets the criteria for appointment. In addition to receiving applications, the Board is specifically empowered to “actively seek out qualified candidates” and to consult with the Chief Justice of Canada and other key stakeholders as they see fit. The
Board is specifically tasked with supporting the government’s intent to achieve “a 2017 CanLIIDocs 175 gender-balanced Supreme Court of Canada that also reflects the diversity of members of Canadian society, including Indigenous peoples, persons with disabilities and members of linguistic, ethnic and other minority communities including those whose members’ gender identity or sexual orientation differs from that of the majority.” The Board is required to provide an assessment of how each recommended candidate meets the requirements of the Supreme Court Act31 and the extent to which they meet the established criteria, along with any additional reasons in support of their candidacy.
In addition, the Board is obligated to issue a public report on its activities within one month of an appointment. On November 25th 2016 the Advisory Board report was made public.32 Virtually unprecedented, the report is the most transparent and descriptive commentary on the appointment of a justice to the Supreme Court of Canada ever willingly provided to the Canadian public.33 It describes the assessment
29 The new process is thoroughly explained on the website of the Office of the Commissioner for Federal Judicial Affairs, online:
30 The Terms of Reference of the Advisory Board provide that the short-list does not bind the Prime Minister. The short list does not bind the Prime Minister (1). The Prime Minister may ask the Advisory Board to provide the names of additional qualified candidates (7). The Government has indicated that its intention is to nominate an individual from the shortlist. See “Frequently Asked Questions”, online:
31 Supreme Court Act, RSC 1985, c S-26, ss 5–6.
32 The report is available online:
33 A non-authorized, and politically embarrassing description of the selection process which lead to the ultimately unsuccessful nomination of Justice Marc Nadon was published by the Globe and Mail on May 14, 2014. See Sean Fine, “The secret short list that provoked the rift between Chief Justice and the PMO”, The Globe and Mail (23 May 2014), online:
If the Advisory Board is the centerpiece of the new process, the statement of qualifications and criteria for appointment, prepared by Professors Adam Dodek, Charles-Maxime Panaccio and Carissima Mathen, members of the Public Law Group at the University of Ottawa, are the backbone. The criteria reflect the role of the Supreme Court in a mature constitutional democracy, and are functionally linked to the Court’s core functions: resolving disputes between a wide range of parties, 2017 CanLIIDocs 175 communicating effectively with the public, upholding the constitution and protecting the rule of law. Broadly divided into three areas, personal skills and experience, personal qualities, and the institutional needs of the court, the criteria are, in the words of Professor Russell, the “most thorough” statement ever prepared on the personal and professional qualities expected of a Supreme Court justice.34 Three criteria relate directly to the diversity rationale. The first is the capacity to be aware of, and synthesize information about the social context in which legal disputes arise, as well as a sensitivity to changes in social values which relate to the cases before the Court. The second is an ability to appreciate a diversity of views, perspectives and life experiences, including those relating to groups historically disadvantaged in Canadian society. The commentary accompanying this criteria recognizes that judges invariably “draw on common sense and experience,” that the judicial perspective must be “neither too narrow nor resistant to change”, and must include the “capacity to empathize with persons who come from backgrounds that are very different from one’s own.” Finally, the criteria note that, at an institutional level, the Supreme Court must reasonably reflect the diversity of Canadian society, a diversity which is not yet “fully reflected in its institutions.” A reasonably reflective Court will benefit from a “range of viewpoints and perspectives” and promote “public confidence in the administration of justice as well as in the appointment process.”35
These thoughtful criteria are made operational by a detailed questionnaire which foregrounds diversity. Candidates are offered an option to self-identify, which is explicitly linked to the government’s diversity objectives. In addition to the usual materials related to education, professional and employment history, legal experience and expertise, and legal skills, the questionnaire requires candidates to write five
34 Russell, supra note 2.
35 The qualifications and assessment criteria for the new appointment process are described online:
In short, all of the materials made public as part of the new appointment process are explicitly informed by a commitment to diversity functionally linked to judicial excellence and public trust. They inspire legitimate confidence that 2017 CanLIIDocs 175 appointments to the Supreme Court will be more transparent, inclusive, and consistent with our democratic aspirations than ever before. Of course, this confidence relies on the willingness of the Prime Minister to stand by his public position, a confidence sorely tested by the previous government. And it will require political courage, as the public debate surrounding the Rowe appointment suggests.
Two questions dominated public discussion of the new process. The first was the fate of the convention of regional representation. The second, perhaps less controversial, was the requirement of functional bilingualism. Both questions demonstrate how challenging it can be to take account of diversity, especially when embodied in only nine individuals.37 I share the view of many that the requirement of functional bilingualism may well require governments to make difficult trade-offs between bilingualism, excellence, and other forms of diversity. I also worry, as Frances Wooley has argued, that there may be an inherent elitism in the bilingualism requirement,38 although I acknowledge that, by definition the pool of potential candidates for the Supreme Court is unavoidably and necessarily, elite. The Minister of Justice, when asked, suggested that those who contemplate an eventual application to sit on the Supreme Court should achieve functional bilingualism as soon as possible.39 This suggestion fails to take into account the ways in which opportunities
36 Justice Rowe’s completed questionnaire, redacted only for privacy, is available online:
37 But see contra Michel Doucet, “Le biliguisme: une exigence raisonnable et essentielle pour la nomination des juges à la Cour supreme du Canada” (2017) 68 UNBLJ 30, and the powerful arguments made by Sébastian Grammond & Mark Power, “Should Supreme Court Judges be Required to be Bilingual?” (2011) Institute of Intergovernmental Relations, School of Policy Studies, Queen’s University, SC Working Paper 2011–02, online:
38 Frances Woolley, “A Supreme Folly” (20 October 2016), ABlawg: The University of Calgary Faculty of Law Blog (blog), online:
26 UNBLJ RD UN-B [VOL/TOME 68 to learn a second language are unequally distributed, and on whom this task will disproportionately fall. Professor Russell has identified the particular challenges of the bilingualism requirement for Aboriginal peoples, and suggests that fluency in an Aboriginal language should be a sufficient equivalent. I disagree. It would be a cruel irony if the unlawful and heartbreaking loss of aboriginal languages occasioned by the residential schools tragedy ended up disqualifying a unilingual Indigenous candidate for the Supreme Court of Canada.40 I think there is a powerful substantive equality claim, at the least, that functional bilingualism should not be required for Indigenous candidates. The bilingualism requirement, like any other kind of job qualification, has potentially discriminatory adverse impacts which merit concern and respect. Interestingly, the government made functional bilingualism non- negotiable, but was, as I discuss below, prepared to abandon the convention of regional representation. This suggests an ordering of the criteria essential to judicial excellence. The political price of that ordering was made apparent in the lead-up to the nomination.
The convention of regional representation required the appointment of a judge from the Atlantic region to replace Nova Scotia’s Thomas Cromwell. It was 2017 CanLIIDocs 175 clear from the outset that the government was prepared to contemplate ignoring the convention. The explicit reason for doing so was an unwillingness to limit the search for “outstanding individuals,” a search which the Minister claimed, at the hearing into the nomination, was consistent with “the values of Canadians today, [which] supports a modern Supreme Court of Canada that is reflective of and responsive to those values.”41 The problem, for a government overtly and publicly committed to diversity and to reconciliation with Indigenous peoples, was that their apparent desire to appoint the first Indigenous, or visible minority judge to the Supreme Court of Canada would be made appreciably more difficult if they were limited to Atlantic Canadian candidates. In fact, to be accurate, those goals would be difficult to meet in most parts of the country, given the current demographics of the Canadian judiciary. The government’s refusal to pre-emptively limit applications triggered a court challenge in Atlantic Canada,42 and an unanticipated resolution in the House of
39 At the special Justice and Human Rights Committee hearing examining the process by which Justice Rowe was appointed, Minister Wilson-Raybould encouraged “all of those individuals out there that meet the statutory requirements ... to brush up on their French if they are wanting to apply to be the next Supreme Court justice." See “Functionally bilingual requirement here to stay, says Wilson-Raybould”, CBC News (24 October 2016), online:
40 Kristy Kirkup, “Top court’s bilingual rule a barrier to indigenous judges: Sinclair, Bellegarde”, The Globe and Mail (22 September 2016), online:
41 Minister Wilson-Raybould as quoted by Sean Fine, “Liberals stick to countrywide Supreme Court selection process”, The Globe and Mail (24 October 24 2016), online:
42 Peter Zimonjic, “Atlantic Canada lawyers challenge Trudeau on changes to Supreme Court appointment process”, CBC News (19 September 2016), online:
Commons which urged the government to abide by the regional convention.43 The resolution was unanimously supported by all 270 members of the House, effectively thwarting any desire the government had to look outside Atlantic Canada, and eventually leading to the appointment of Justice Malcolm Rowe.
I have no doubt that Mr. Justice Rowe will be an outstanding Supreme Court judge, who will make an important contribution, not only because of his qualifications as a jurist and public servant, but because of the particular perspective he will bring as a resident of Newfoundland and Labrador. I do worry, however, that the politics of regional diversity has wrongly stymied the justice of the need for racialized Canadians to see themselves represented on the Court, and, perhaps even more importantly, the legitimate claim of Indigenous people. In my view, the necessity of Indigenous representation on the Supreme Court implicates institutional legitimacy. Reconciliation requires the appointment of an Indigenous jurist to the Court because Indigenous legal systems are a legitimate part of an increasingly tri- jural Canada. They must be honoured, implemented, and acknowledged as part of a national process of reconciliation. The Prime Minister’s ministerial mandate letters said as much, providing: “[n]o relationship is more important to me and to Canada 2017 CanLIIDocs 175 than the one with Indigenous Peoples. It is time for a renewed, nation-to-nation relationship with Indigenous Peoples, based on recognition of rights, respect, co- operation, and partnership.”44 That renewed relationship requires an indigenous presence on our highest court. In my view, we need to look closely and critically at the convention of regional representation. Are the diversity values served by regional representation truly critical to institutional impartiality? Or, are they more political (historic) than legal? Should that matter? And, in what circumstances do we need to recognize that previously unacknowledged (or suppressed, or colonized) perspectives may be more functionally significant to the evolution of Canadian law than regionalism? The impending retirement of Chief Justice McLachlin will present the current government with another opportunity to appoint a Supreme Court jurist, and many hope that the government will respond with the historic appointment that is institutionally required. However, the political nature of the decision makes delay risky. If 2016 has taught us anything, it is that there is no such thing as certainty in politics.
Diversity and the Federal Judiciary
The government’s commitment to judicial diversity in federal appointments was apparent even before it began making formal changes to selection processes in August of 2016. In June 2016, the government announced fifteen section 96 appointments which clearly demonstrated a deliberate shift in priorities. This was confirmed by the Minister’s public acknowledgement that “our judicial system is
43 Kate Simpson, “MPs unanimously support regional representation for Supreme Court”, CBC News (27 September 2016), online:
44 Online:
The cornerstone of the section 96 reforms are changes intended to “strengthen the role of Judicial Advisory Committees”47 (JACs). The government has reversed the changes made in 2006 by: 1) restoring the right of judicial members on the JACs to vote; 2) removing the representative of law enforcement; and 3) re- instating the highly recommended category. Each JAC will consist of seven volunteer members representing the bench, the bar, and the general public and include: one nominee of the provincial or territorial law society, one nominee of the 2017 CanLIIDocs 175 provincial or territorial branch of the CBA, one judge nominated by Chief Justice of the province; one nominee of the provincial Attorney General and three nominees of the federal government explicitly described as representing the ‘general public.’ Committee members are selected by the federal government from either a list of three nominees provided by the relevant nominating authority, or, for the public representatives, through a new, application-based process. Diversity aspirations are threaded throughout.48 Members of the JACs are selected “with a view to achieving a gender-balanced Committee that also reflects the diversity of members of each jurisdiction, including Indigenous peoples, persons with disabilities and members of linguistic, ethnic and other minority communities, including those whose members’ gender identity or sexual orientation differs from that of the majority.” Along with their assessment of professional competence and overall merit, Committee members are overtly charged with attempting to create a candidate pool which is similarly reflective of the jurisdiction. The criteria for appointment seem largely unchanged, but the completely revamped questionnaire is the same as the one completed by applicants for the Supreme Court of Canada. The option to self-identify is intended, presumably, to allow the government to fulfil its promise of collecting and publishing “statistics and demographic information on both applicants for and appointments to judicial office to measure whether Canada is meeting its diversity
45 Department of Justice Canada, News Release, “The Government of Canada announces judicial appointments in the province of British Columbia” (17 June 2016), online:
46 Department of Justice Canada, News Release, “Government of Canada announces judicial appointments and reforms the appointments process to increase openness and transparency” (20 October 2016), online:
47 Ibid.
48 An overview of the new process is found online:
Conclusion
In an editorial published on the day of Justice Rowe’s nomination to the Supreme Court, the Globe and Mail hailed the nomination as a triumph of “qualifications” 50 over “identity politics.” This is a familiar, but simplistic and counter-productive 2017 CanLIIDocs 175 analysis of the complex compromises required by a commitment to diversity on a nine-member court. Judicial appointment in Canada has always been and remains political in a manner consistent with Canadian political traditions. Professor Sossin has suggested that the inevitable political preferences shaping selection processes should not be “the only or primary ones for appointment.”51 I agree. The reforms to judicial appointment put in place in 2016 have the potential to replace pure partisanship with politics of a different sort, politics rooted in legal and constitutional norms … not identity politics, but constitutional politics. In my view, a commitment to diversity which is explicitly linked to the constitutional value of impartiality, and an appointment process which is increasingly transparent and infused with lay participation will augment the judiciary’s capacity to operate as independent guarantors of the rule of law. It remains to be seen whether the government can sustain the political will to live up to the public promise of these reforms. I, for one, am cautiously optimistic.
49 For a comprehensive discussion of the importance of data collection see Lorne Sossin and Sabrina Lyon, “Diversity and Data in the Canadian Legal Community” (2014) 11 JL & Equality 85.
50 “Ignoring identity politics, Trudeau makes a Supreme choice”, Editorial, The Globe and Mail (17 October 2016), online:
51 Lorne Sossin, “Seeing Through Judicial Appointments” Policy Options (5 October 2016); supra note 11.
LE BILINGUISME : UNE EXIGENCE RAISONNABLE ET ESSENTIELLE POUR LA NOMINATION DES JUGES À LA COUR SUPRÊME DU CANADA
Michel Doucet, c.r.*
INTRODUCTION
Le 2 août 2016, le gouvernement canadien annonçait la mise sur pied d’un comité consultatif indépendant sur la nomination des juges à la Cour suprême du Canada.1 Le comité avait comme mandat de présenter au premier ministre des recommandations2 fondées sur le mérite. Les conditions en vue d’une nomination à la Cour suprême sont prévues aux articles 5 et 6 de la Loi sur la Cour suprême.3 En plus de ces conditions, le gouvernement canadien s’est engagé à ne nommer que des 2017 CanLIIDocs 175 juges qui sont « effectivement bilingues ». Le juge, qui sera retenu en vertu de ce processus doit pouvoir lire des documents, comprendre un plaidoyer sans devoir recourir à la traduction ou à l’interprétation et il ou elle doit pouvoir discuter avec un avocat pendant un plaidoyer en français ou en anglais. Bien que l’interprétation de cette exigence ait pu soulever certains doutes,4 personne ne remettra en question le fait que le juge Malcom Rowe, qui fut nommé à la Cour suprême à la suite de ce processus, rencontre le critère de bilinguisme exigé.
La Cour suprême du Canada est le tribunal de dernière instance au pays. Les parties qui comparaissent devant la Cour ont le droit constitutionnel5 et législatif6 d’employer l’une ou l’autre des langues officielles dans leurs plaidoiries orales et écrites, et les décisions de la Cour sont publiées simultanément dans les deux langues
* Professeur titulaire et directeur de l’Observatoire international des droits linguistiques à la Faculté de droit de l’Université de Moncton.
1 « Nouveau processus de nomination des juges de la Cour suprême du Canada », en ligne : Justin Trudeau, premier ministre du Canada
2 L’annonce du processus de sélection parle de recommandations non contraignantes, mais il va sans dire qu’une recommandation n’est pas contraignante.
3 Loi sur la Cour suprême, LRC 1985, c S-26.
4 Voir, par exemple, « Juges de la Cour suprême : bilingues sans parler français? », en ligne : Radio- Canada.ca
5 Partie I de la Loi constitutionnelle de 1982, constituant l’annexe B de la Loi de 1982 sur le Canada (R- U), 1982, c 11, art 19 [Charte].
6 Loi sur les langues officielles, LRC 1985, c 31 (4e suppl), art 14 [LLO].
2017] LE BILINGUISME : UNE EXIGENCE RAISONNABLE 31 officielles.7 Or, l’article 16 de la LLO, qui prévoit que les juges des tribunaux fédéraux doivent comprendre la langue officielle du procès sans l’aide d’un interprète, crée une exception pour la Cour suprême du Canada. L’article prévoit que le juge des « tribunaux fédéraux autres que la Cour suprême du Canada » qui entend l’affaire doit comprendre les plaideurs directement, sans l’aide d’un interprète, dans la langue officielle dans laquelle ils ont choisi de procéder. L’exception défavorise les plaideurs qui souhaitent procéder en français devant la Cour suprême, car, comme l’écrivait si bien Peter Russell, en 1969 : while fluency in English appears to have been a necessary qualification for membership on the Court’s bench, fluency in French has not.8 Ainsi, le plaideur francophone qui souhaitait procéder dans sa langue doit accepter de plaider son dossier par l’intermédiaire d’un interprète afin que les juges unilingues de la Cour puissent le comprendre.
Sébastien Grammond et Mark Power ont très bien décrit, dans un article traitant du bilinguisme des juges de la Cour suprême du Canada,9 les désavantages auxquels font face les avocats qui décident de plaider en français. Ils notent, entre autres, que les mémoires et les autres documents écrits présentés à la Cour par les parties ne sont pas traduits. Les juges unilingues anglophones n’ont donc pas accès 2017 CanLIIDocs 175 directement aux arguments écrits, aux éléments de preuve et à la jurisprudence déposés en français. L’importance de cet aspect de la plaidoirie ne saurait être sous- estimé. Le mémoire écrit de l’avocat ou de l’avocate constitue le premier contact qu’il ou qu’elle a avec les juges. Le mémoire écrit représente l’occasion de bien exposer sa position et de convaincre d’ores et déjà les juges qui entendront l’affaire. Le plaideur qui procède dans une langue qu’un ou des juges ne comprennent pas part donc sur un pied d’inégalité par rapport à ceux et celles qui décident de procéder en anglais, langue que tous les juges comprennent.
Bien que les interprètes de la Cour suprême du Canada fassent, règle générale, un excellent travail, il n’en demeure pas moins qu’il leur est souvent impossible de saisir toutes les nuances des arguments présentés et même parfois de suivre les échanges, souvent rapides et intenses, entre les juges et les avocats. J’ai personnellement eu l’occasion de prendre connaissance des limites de l’interprétation simultanée à la Cour suprême du Canada dans un dossier que nous avons perdu à 5 contre 4.10 Sans prétendre que l’interprétation simultanée soit la raison de ce résultat, je dois admettre qu’après avoir écouté sur le Cable Public Affair Network mes plaidoiries en français traduites en anglais, je me suis sérieusement posé des questions à savoir ce que les juges unilingues anglophones avaient compris. À plusieurs reprises, l’interprète était dans l’impossibilité de suivre les échanges. Il se
7 Ibid, art 20.
8 PH Russell, « The Supreme Court of Canada as a Bilingual and Bicultural Institution », Documents of the Royal Commission on Bilingualism and Biculturalism, Ottawa, Queen’s Printer, 1969 à la p 61.
9 S Grammond et M Power, « Should Supreme Court Judges be Required to be Bilingual? » Special Series on the Federal Dimensions of Reforming the Supreme Court of Canada, Institute of Intergovernmental Relations, School of Policy Studies, Queen’s University, SC Working Paper 2011-12.
10 Charlebois c Saint John (City), [2005] 3 RCS 563, 2005 CSC 74.
32 UNBLJ RD UN-B [VOL/TOME 68 référait aussi au paragraphe 16(1) lorsque dans mon argumentaire, je faisais référence à l’article 16.1 de la Charte. Je me suis alors posé la question à savoir si j’avais bien rendu service à ma cliente en utilisant le français pour mes plaidoiries : un doute que l’on ne devrait jamais avoir lorsqu’on plaide un dossier devant le plus haut tribunal d’un pays officiellement bilingue comme le Canada. D’ailleurs, Peter Russell avait également fait référence à ce problème dans son texte pour la Commission royale d’enquête sur le bilinguisme et le biculturalisme en indiquant que plusieurs avocats francophones préféraient plaider leur cause en anglais afin de s’assurer d’être compris par la Cour.11
Pourquoi les avocats francophones devraient-ils s’adresser à certains juges de la Cour suprême du Canada par l’intermédiaire d’un interprète, alors que ce n’est pas le cas pour les avocats anglophones? Tous les juristes sont conscients de l’importance des plaidoiries orales dans le régime juridique canadien. La plaidoirie orale permet aux avocats de préciser leurs arguments écrits à la lumière des questions qui leur sont posées par la Cour. Ils ne suivent pas nécessairement à la lettre l’ordre de présentation de leur mémoire écrit. Ils cherchent souvent par l’argument oral à clarifier certains points et à articuler leur argumentaire d’une manière que la 2017 CanLIIDocs 175 plaidoirie écrite ne le permet pas. Ce n’est pas pour rien que les deux formes de communication sont employées lors d’une audience. Elles ont toutes deux des forces et des faiblesses. La plaidoirie écrite permet de bien détailler l’affaire, d’appuyer le tout par des références précises et, encore plus important, permet au lecteur de s’arrêter et de réfléchir, au besoin, lors de la lecture du document. La plaidoirie orale, quant à elle, est plus dynamique et elle permet à l’orateur de cibler les éléments clés et de répondre directement aux questions des juges. Priver une partie de l’une ou l’autre de ces deux formes d’expression est une atteinte sérieuse au droit fondamental d’ester en justice ou d’accès à la justice.
Si la situation permettant à des juges unilingues anglophones de siéger à la Cour suprême du Canada devait perdurer, nous serions en droit de nous demander comment un plaideur pourrait se sentir égal à son collègue anglophone s’il doit, pour se faire comprendre par le plus haut tribunal du pays, passer par l’intermédiaire d’un interprète? Comme l’a si bien dit le juge en chef Dickson dans l’arrêt Société des Acadiens c. Association of Parents : « Dans une salle d’audience, c’est en parlant qu’on communique avec le juge ou les juges ».12 Il est donc primordial, pour qu’il y ait une égalité réelle entre les deux communautés de langues officielles, que le juge ou les juges comprennent directement la langue choisie par le justiciable. À la Cour suprême du Canada, seul le plaideur anglophone a ce privilège. Le plaideur francophone, pour sa part, doit accepter la présence d’un intermédiaire, sur lequel il n’a aucun contrôle, entre lui et le juge.
Dans le contexte d’une plaidoirie, l’interprétation simultanée n’est pas idéale. Comme le souligne Varady : Translation simply cannot fully mirror both
11 Voir Russell, supra note 8.
12 [1986] 1 RCS 549 au para 25 [Société des Acadiens].
2017] LE BILINGUISME : UNE EXIGENCE RAISONNABLE 33 arguments and the art of advocacy. It cannot reflect every emphasis, gambit of persuasion, or undertone. Often, the arguments are not reflected clearly either.13
Pourquoi est-il si difficile pour certains de comprendre l’importance d’instaurer un régime qui permettrait au plus haut tribunal du pays de fonctionner entièrement dans les deux langues officielles. La Cour suprême du Canada n’est-elle pas appelée à interpréter des textes constitutionnels et législatifs adoptés dans les deux langues officielles et dont les deux versions ont également force de loi ?14 Le principe de la valeur égale des lois signifie que les juristes, anglophones et francophones, ont l’obligation de lire la législation dans ses deux versions linguistiques, sinon ils ne pourront être certains du véritable sens d’une loi, car l’une n’est pas simplement la traduction de l’autre. Dans un tel contexte, il importe que tous les acteurs judiciaires, y compris les juges, aient une connaissance de l’autre langue, sinon ils ne font qu’interpréter une partie de l’intention du législateur.
Exiger le bilinguisme comme condition à la nomination des juges à la Cour suprême du Canada n’a pas seulement pour but de faciliter l’accès à cette institution pour la communauté francophone, mais envoie également un message clair que les 2017 CanLIIDocs 175 francophones ont suffisamment de valeur au sein de la société canadienne pour que la plus haute cour du pays leur soit accessible dans leur langue, et sans intermédiaire. En faisant du bilinguisme un critère de sélection des juges de la Cour suprême du Canada, le gouvernement canadien répond à une obligation importante dans un contexte d’égalité linguistique. Or, cette reconnaissance ne doit pas seulement demeurer à titre de recommandation, elle doit devenir la norme.
Ceux et celles qui s’opposent à l’idée d’exiger le bilinguisme pour une nomination à la Cour suprême du Canada font valoir que ce critère de sélection éliminerait de nombreux candidats compétents. Ils n’ont cependant pas compris que si la connaissance des deux langues officielles est nécessaire à l’exercice de la fonction de juge, le candidat unilingue n’est tout simplement pas aussi compétent que le candidat bilingue.
Certaines personnes soutiennent que de grands juges, comme Bora Laskin ou Brian Dickson, anciens juges en chef de la Cour suprême du Canada, n’auraient jamais pu être nommés à la Cour en raison de leur unilinguisme. Nous pourrions répondre en prenant pour exemple la nomination de femmes à la Cour suprême du
13 T Varady, Language and Translation in International Commercial Arbitration, The Hauge, TMV Asser Press, 2006 aux pp 49–50, cité dans Grammond et Power, supra note 9 aux pp 5–6. Voir également, Syndicat des travailleurs et travailleuses des postes c Société canadienne des postes, 2012 CF 110, [2012] ACF no 119 (QL) où la Cour fédérale a émis le commentaire suivant au para 46 : « Il m’apparait que l’arbitre des offres finales doit être capable de lire la convention collective et les offres finales dans les deux langues officielles. Enfin, forcer une partie à la convention collective, ses représentants et ses témoins à procéder ou à témoigner à l’audition contre leur gré dans l’autre langue officielle est non seulement injuste et préjudiciable, mais à terme, ceci pourrait justifier la Cour de casser la décision finale [rendue] par l’arbitre des offres finales désigné par la Ministre. » Si une telle conclusion s’applique à un arbitre dans le cadre d’un arbitrage, elle devrait d’autant plus s’appliquer aux juges de la Cour suprême du Canada.
14 Charte, supra note 5, art 18. 34 UNBLJ RD UN-B [VOL/TOME 68
Canada qui n’existe que depuis les années 1980. Est-ce qu’on oserait affirmer que le fait d’avoir nommé des femmes à la Cour prive des hommes compétents d’une nomination? La société change et ses institutions doivent également évoluer.
CONCLUSION
L’exigence de bilinguisme pour les juges à la Cour suprême du Canada ne m’apparait non seulement raisonnable mais essentiel. J’ajouterai que le degré de compréhension du juge qui veut siéger au plus haut tribunal du pays doit aller au- delà de la simple compréhension littérale de la langue employée par l’avocat ou l’avocate. Le juge ou la juge doit être en mesure d’apprécier tout le sens des arguments qui lui sont présentés dans l’une ou l’autre des deux langues officielles du pays. Les juges à la Cour suprême du Canada doivent atteindre ce niveau de perfectionnement pour que les droits linguistiques d’un plaideur aient un sens dans le cadre de procédures judiciaires.
Le système judiciaire canadien affichait ses vraies couleurs en n’exigeant 2017 CanLIIDocs 175 pas que les juges à la Cour suprême du Canada soient bilingues. Il envoyait le message que les droits linguistiques au Canada, qu’en disent la Constitution et les lois, ne sont pas réellement importants et qu’à défaut de s’y conformer, on créera pour eux une exception. En ajoutant cette exigence au processus de nomination des juges, le gouvernement canadien vient corriger cette perception erronée.
Pour ce qui est de la façon de déterminer le niveau de compétence linguistique des candidats à la Cour suprême du Canada, un système d’évaluation des compétences linguistiques dans les deux langues officielles doit être mis en place. La connaissance d’une langue comporte quatre niveaux :
(1) la compréhension de la langue écrite; (2) la compréhension de la langue parlée; (3) la capacité de s’exprimer oralement dans la langue en question; et (4) la capacité d’écrire dans cette langue.15
Afin de réaliser l’égalité linguistique, les juges nommés à la Cour suprême du Canada devraient, dès leur nomination, avoir atteint ce niveau de connaissance des deux langues officielles.
J’ose espérer que l’exigence selon laquelle les juges à la Cour suprême du Canada doivent être « effectivement bilingues », ne demeure pas qu’une recommandation, mais devienne une obligation, inscrite en bonne et due forme dans un texte de loi pour en assurer la pérennité. Depuis 2008, plusieurs projets de loi émanant de députés visant cet objectif ont été déposés à la Chambre des communes.
15 Le juge en chef Monnin de la Cour d’appel du Manitoba, tel qu’il est cité dans Société des Acadiens, supra note 12 au para 172.
2017] LE BILINGUISME : UNE EXIGENCE RAISONNABLE 35
En mai 2008, le projet de loi C-54816 proposait de modifier l’article 16 de la LLO pour que l’exception prévue pour la Cour suprême du Canada soit enlevée et que la Cour soit assujettie à la même obligation que les autres tribunaux fédéraux. En juin 2008, le projet de loi C-55917 proposait plutôt de modifier l’article 5 de la Loi sur la Cour suprême en y incorporant une obligation de compréhension des deux langues officielles pour les juges nommés à la Cour. Ces deux projets de loi sont morts au Feuilleton. Le projet de loi C-232,18 présenté en 2009, qui proposait la même modification que le projet de loi C-559, est celui qui s’est rendu le plus loin : après avoir été adopté en troisième lecture par la Chambre des Communes, il s’est cependant buté à la majorité conservatrice du Sénat. En 2014, à la suite d’une nouvelle tentative, le projet de loi C-20819 fut également défait en deuxième lecture à la Chambre des Communes. Deux nouveaux projets de loi ont été déposés en 2015, mais sans donner plus de résultats.20
S’il ne s’agissait que du principe de l’équité dans les procédures judiciaires, il serait suffisant que le juge comprenne assez bien la langue employée par les parties pour que le procès soit équitable, et pour ce faire il pourrait avoir recours, au besoin,
à l’interprétation simultanée. Si, par contre, il est question de l’égalité des langues 2017 CanLIIDocs 175 officielles devant les tribunaux, pareille façon de procéder est insuffisante et inconstitutionnelle. Pour réaliser l’égalité, il est essentiel que les principaux acteurs du système judicaire que sont les juges de la Cour suprême du Canada soient « effectivement bilingues » ou functionally bilingual, comme le dit la version anglaise de l’annonce du premier ministre, c’est-à-dire que le ou la juge doit pouvoir « étudier les documents d’un cas, comprendre les faits et saisir les subtilités des arguments présentés en cour »21 dans les deux langues officielles.
16 Canada PL C-548, Loi modifiant la Loi sur les langues officielles (compréhension des langues officielles — juges de la Cour suprême du Canada), 2e sess, 39e lég, 2007-2008.
17 Canada PL C-559, Loi modifiant la Loi sur la Cour suprême (compréhension des langues officielles), 2e sess, 39e lég, 2008.
18 Canada PL C-232, Loi modifiant la Loi sur la Cour suprême (compréhension des langues officielles), 3e sess, 40e lég, 2010-2011.
19 Canada PL C-208, Loi modifiant la Loi sur la Cour suprême (compréhension des langues officielles), 1ère sess, 41e lég, 2011.
20 Canada PL C-203, Loi modifiant la Loi sur la Cour suprême (compréhension des langues officielles), 1ère sess, 42e lég, 2015 et Canada PL S-209, Loi modifiant la Loi sur les langues officielles (communications et services destinés au public), 1ère sess, 42e lég, 2015.
21 Commissariat aux langues officielles, « Le bilinguisme et la Cour suprême », en ligne :
VIRTUES AND SHORTCOMINGS IN CONSTITUTIONAL DYNAMISM: COMMENTARY ON PRIME MINISTER TRUDEAU’S 2016 CREATION OF AN INDEPENDENT ADVISORY BOARD FOR SUPREME COURT OF CANADA APPOINTMENTS AND ON THE INSTRUCTIONS FOR MAKING RECOMMENDATIONS FOR APPOINTMENT
John D. Whyte1
Background
Although some may question this view, the Supreme Court of Canada is a 2 constitutional court in three different ways. First, it is a constitutional court in that it 2017 CanLIIDocs 175 is recognized in the 1982 Constitution3 through the identification of powers and structures that cannot be amended except in accordance with the constituted procedures for amending the Constitution of Canada. It is true that the Court was created through an ordinary act of the Parliament of Canada4 enacted under an authority conferred on it by the Constitution Act, 1867.5 It might seem, therefore, that the Supreme Court is not a constitutionally entrenched institution since, in the ordinary course, Parliament has the power to repeal its own statutes. While the position that the court is not constitutionally entrenched is sometimes advanced, this is very likely a wrong conclusion. Under section 42 (d) of the Constitutional Act, 1982 there is a general bar to the making of any amendments relating to the Supreme Court other than through the formal constitutional amending procedure required under section 42(d) – or in the case of changes to the composition of the Court under the requirement for unanimous federal and provincial consent as stipulated in section 41(d) of the Constitutional Act. In principle, there is no bar to constitutionally entrenching an institution and its features through the mere reference to it in the Constitution. The legislative status of the instrument that originally created and empowered such an institution does not determine the effect of giving it constitutional protection. Constitutions, while legal in form are conceived in light of
1 Professor Emeritus, Faculty of Law, Queen’s University; Professor Emeritus, Department of Political and International Studies, University of Regina.
2 See e.g. Peter W Hogg, “Appointment of Thomas A. Cromwell to the Supreme Court of Canada” in Nadia Verelli, ed, The Democratic Dilemma: Reforming Canada’s Supreme Court (Montreal & Kingston: McGill-Queen’s University Press, 2013) 13 at 24. See also Warren J Newman, “The Constitutional Status of the Supreme Court of Canada” (2009) 47 SCLR 429.
3 Constitution Act, 1982, s 41(d) and s 42(d), being Schedule B to the Canada Act 1982 (UK), c 11.
4 Supreme and Exchequer Courts Act, SC 1875, c 11.
5 Constitution Act, 1867 (UK), 30 & 31 Vict, c 3, s 101, reprinted in RSC 1985, Appendix II, No 5. 2017] VIRTUES AND SHORTCOMINGS 37 diverse social and political realities and aspirations and; therefore, carry with them the imperative for contextualized application.
The second way that the Supreme Court is a constitutional court is that it carries ultimate responsibility for the interpretation and application of the nation’s constitutional provisions. When a court is labelled a constitutional court what is generally meant is that it is a judicial body, the sole function of which is to adjudicate constitutional issues – to take ultimate responsibility for deciding the scope of constitutional limits on governmental powers. The Supreme Court of Canada, however, is a general court of appeal with responsibility for hearing appeals (although usually only when it grants leaves to appeal) in any case, no matter the nature of the legal claim, from any Court of Appeal in Canada. In fact, it decides more non-constitutional cases than constitutional cases. Those other cases are important in clarifying – and developing – legal norms that shape transactions and relations across a broad range of regulatory, commercial and social relations in Canada. However, from the perspective of the Supreme Court’s visible national role, its most notable function is to decide constitutional cases. As a result, constitutional decision-making has been the area of the Court’s activity that has produced most of 2017 CanLIIDocs 175 the public attention that is paid to the Court. The Supreme Court has, since its creation in 1870, been making constitutional decisions that have been vitally important to the exercise of government in Canada. These cases have until the last third of a century dealt with conflicts between the federal and provincial governments over regulatory authority and, as well, with issues over the jurisdiction of the nation’s superior courts to hear jurisdictional and procedural challenges to the regulatory activities of provincial and federal governments. But it has been the constitutionalization of basic human rights and minority’ rights in 1982,6 and the court’s decisions on the meaning and application of the Charter, that has drawn a much broader political and popular interest to the Court’s work and, hence, broader awareness of the court’s composition and workings – and, hence, the significance of the federal government’s appointment decisions.
Charter cases raise questions that engage the broadest and deepest values of people; they touch on every person’s idea of the good society and present issues on which there is seldom an expectation of common understanding or common position. Questions such as how to maintain the integrity of identities or faiths, or what just and equal treatment requires, or when the demand for accommodation is unreasonable, or what limits on free expression are vital to social solidarity, are all moral questions for a state and its people. These are questions for which there is seldom public indifference and the Court, in deciding such issues, is involved in mediating differences of opinion that are close to the people’s deepest commitments. The Court’s public reputation has become that of a constitutional court and, in particular, a human rights court.7
6 Canadian Charter of Rights and Freedoms, Part I of the Constitution Act, 1982, being Schedule B to the Canada Act 1982 (UK), 1982, c11.
38 UNBLJ RD UN-B [VOL/TOME 68
The third way in which the Supreme Court is a constitutional court is that, although not created by constitutional provision, it has attracted a number of constitutional or quasi-constitutional constraints with respect to the federal government’s relationship with it, especially with respect to appointments. The most notable of the politically constituted restraints is the practice of allocating seats on the Court to regions of the country – two members from the four Western provinces, three from Ontario and one from Atlantic provinces, in addition to the legislated and, now, the constitutionally entrenched allocation of three seats from Quebec.8
Although the Supreme Court has become constitutionally bound since its creation, this does not mean that there is no room for innovation in the exercise of the governmental power to make appointments to it. While constitutions are generally designed to be rigid and only amendable with the consent of a large number of (diverse) jurisdictions, this does not mean that national self-national determination with respect to the rules, structures and relationships of governing bodies can never be exercised. There is often room for the organic development of the norms under which the state conducts its internal political relationships between orders, divisions and branches of government while acting with fidelity to the 2017 CanLIIDocs 175 constitutional order. Although these practices, if clearly and purposefully articulated and invariably followed, might create a constitutional convention which could become politically binding, there can be much experimentation in the exercise of powers that responds to new realities and expectations that do not create constitutional obligations. Mature constitutional democracies are able to change political practice and, in this way, give flexibility to their basic rules without undertaking the difficult process of formal constitutional change.9
With respect to appointing judges to the Supreme Court there have been two developments over the past decade and a half that represent organic responses to the new Charter driven political sensibilities. They have created a strongly presumptive – and, possibly now, an unavoidable – condition for exercising the appointment power.10 The first is that governmental executive power to appoint Supreme Court judges should be constrained by a process of identifying persons who
7 It would, however, be a mistake to overlook the widespread popular engagement with Supreme Court cases that dealt with issues of constitutional reform based on the nature and the constraints of Canadian federalism. The court’s reputation as a constitutional court was certainly confirmed by its decisions in the many reference cases that, from the perspective of the federal principle, dealt with such questions as the role of provinces in making requests to the United Kingdom Parliament for amendments to the constitution, the necessity of Quebec’s consent to constitutional amendments under the pre-patriation convention relating to amendments, the rules relating to provincial secession from the Canadian federation and constitutional reforms relating to the Senate of Canada.
8 Supreme Court Act, RSC 1985, C S-26, s 6.
9 See Akhil Reed Amar, America’s Unwritten Constitution: The Precedents and Principles We Live By (New York: Basic Books, 2012) at 477: “We must look back backward in time and claim our constitutional inheritance, and we must also look forward in time and make our constitutional donation. [This] second responsibility does not reside on the clear surface of any explicit constitutional text...”
10 For a description of this process, see David Schneiderman, Red, White and Kind of Blue?: The Conservatives and the Americanization of Canadian Constitutional Culture (Toronto: University of Toronto Press, 2015) at ch 5 (234–252), “Appointing Justices: Supreme Court Nominees and the Press”. 2017] VIRTUES AND SHORTCOMINGS 39 are suitable for appointment and from which the government should ordinarily select the person it will appoint. This development tracks a long-standing political sense that the executive’s power to appoint judges should be constrained through a process that seeks to remove the influence of party loyalty or membership, filters potential nominees for professional competence and ensures that the personal character of persons recommended matches the judicial ideals of restraint, responsibility and an impartial concern for any person or interest who comes before the court. This process also has the potential virtue of discovering persons eminently suitable for judicial appointment who would not otherwise come to the attention of the appointing power. To this end, it is often the case that bodies responsible for nominating candidates for appointment will advertise for people to make application for appointment. The idea of a committee based process for creating a short list of potential nominees for appointment was first adopted in making the Supreme Court appointment to replace Justice John Major in 2005. It was then abandoned, or reduced to a half-hearted form, during the appointments made by Stephen Harper’s Conservative government, but was then then adopted again by Justin Trudeau’s Liberal government in making the 2016 appointment to replace Justice Thomas
Cromwell. 2017 CanLIIDocs 175
The second innovation in making Supreme Court appointments has been the creation of a process for some degree of Parliamentary review of persons nominated for appointment. This development was a response to the higher political sensitivity to Supreme Court nominations once the Court’s Charter of Rights decisions became conspicuous. The established view that governments appointed judges based on competence and integrity and not on any sense of how a candidate would approach specific types of legal issues that might come before the Court began to be questioned. This awareness of political predisposition was undoubtedly a product of the court’s Charter jurisprudence, not necessarily because Supreme Court justices became more blatant in their foundational values and beliefs, but because under the Charter there is no avoiding disclosure by judges of attitudes that the whole population is able to grasp and evaluate – values such as the responsible scope of personal liberty, what processes are fair for persons restrained or regulated by governmental action and when does differential treatment (or the absence of differential treatment) amount to inequality. Adoption of a process of parliamentary review of persons identified for appointment was initially adopted by the Harper government in the appointment of Justice Rothstein in 2006. The practice was not consistently followed by that government after its initial use. It was used in the appointments of Justices Karakatsanis (2011), Moldaver (2011), Wagner (2012) and Nadon (2013) (Nadon was later disqualified from sitting as a result of the Supreme Court opinion that he was constitutionally ineligible for appointment.)11 The parliamentary process was not used in the Harper government appointments of Justices Cromwell (2008), Gascon (2014), Côté (2014) and Brown (2015). This history clearly, belies the claim of a virtual political entrenchment of this process.
11 Reference re Supreme Court Act, ss. 5 and 6, 2014 SCC 21, [2014] 1 SCR 433.
40 UNBLJ RD UN-B [VOL/TOME 68
Nevertheless, it seems likely that future appointments will involve some form of parliamentary engagement. As David Beatty wrote in 1990, the “monopoly [over appointments] enjoyed by the executive branch can no [longer] be defended.”12
Supreme Court of Canada Judicial Appointment Process
On August 2, 2016 the process for making an appointment to the Court to replace Justice Thomas Cromwell was announced by Prime Minister Trudeau.13 It had four main features – qualified persons could apply for appointment, a committee – called the Advisory Board – would be created to develop a list of persons that it considers suitable for appointment, with the proviso that the government would not be bound to nominate a person from that list, stipulations with respect to the factors of representativeness that the committee should take account of were identified and a weak process of parliamentary review of the person that the Prime Minister intended to appoint was announced. All of these represent a degree of political constraint on the government’s power to appoint Supreme Court judges, although not a binding constraint; the Trudeau plan carefully avoided any hint of constitutional amendment 2017 CanLIIDocs 175 that would arise from making a formal alteration of the statutory (and, now, likely constitutional) power of appointment. The second feature is expressed in the Terms of Reference as a constraint on the Advisory Board but, of course, when that feature is connected to the presumption of making an appointment based on the Advisory Board’s recommendation, it becomes a derivative constraint on the government.
The Trudeau plan responds to the sense that Supreme Court appointments have become significant to Canadian public regulation and, therefore, a structure of higher accountability and restraint is desirable. The new appointment process was also a response to a degree of public dissatisfaction over some of the Supreme Court appointments made by his predecessor, Prime Minister Stephen Harper, and to dissatisfaction over his highly exigent and erratic – seemingly opportunistic – ad option of the consultation and review process that was initiated by Prime Minister Paul Martin. Further, notwithstanding a Canadian tradition of not criticizing Supreme Court appointments, at least two of his appointments received public criticism. This helped add to the anxiety that appointments to the Court might be made on bases other than a record of strong commitment to neutrality or commitment to the integrity of the legal process. Prime Minister Trudeau was faced with heightened public sensitivity with respect to Supreme Court appointments.
Elsewhere there had already been a strong development of regimes for restraining appointments to other courts – provincial lower courts, federal courts and provincial superior and appeal courts. However, the process created in 2016 for
12 David Beatty, Talking Heads and the Supremes: The Canadian Production of Judicial Review (Toronto: Carswell, 1990) at 263 [emphasis added].
13 Office of the Prime Minister, News Release, “Prime Minister Announces New Supreme Court of Canada Judicial Appointment Process” (2 August 2016), online:
Supreme Court of Canada appointments was less motivated by the desire to arrive at sounder appointments than it was a response to the particular pressures for greater governmental accountability in making appointments to the nation’s Supreme Court and, in particular, the nation’s court of last resort in human rights and minority rights cases – and other constitutional cases.
The 2016 process was adopted in response to the need to condition the exercises of a constitutional power. In that way, it is a constructed restraint for political purposes; it is a refinement of a constitutional process in the face of perceived and experienced dangers of a unilateral power. It works as a constitutional refinement without itself becoming constituted. In light of this undoubted statecraft purpose of restraining an executive authority that has become too great to sit comfortably within the liberal democratic paradigm of accountability, this process – or another process with similar features – is likely to persist and become part of political conventionality. That is not to say that the 2016 process is close to becoming a constitutional convention but it is purposive in precisely the way that political practices become conventions. For this reason, it is appropriate to look carefully at its elements with a view to determining whether they serve effectively 2017 CanLIIDocs 175 the underlying purpose of preserving the integrity of the Supreme Court of Canada.
The feature of an open invitation to apply for a Supreme Court appointments is, from a functional perspective, relatively insignificant, although, of course, highly attractive from a cosmetic perspective. It is certainly true that the route into a Supreme Court judgeship should not track the methods of club membership and that there are undoubtedly talented and highly able lawyers who lie below the radar scan for making appointments. Nevertheless, it is not likely that anyone will ever be appointed to the Supreme Court who has not been a part of a network of esteemed lawyers (or judges) working within a national structure of legal scholarship, legal practice, courts, governmental branches and agencies or, it must be admitted, political parties. The open application regime is not misguided; it is simply unlikely to lead to an alteration in who it is that will receive a Supreme Court appointment.
A similar analysis might also apply to the creation of a committee to recommend persons for appointment. The likelihood of such a committee straying outside the reasonably well-recognized community of persons in the class of potential Supreme Court judges depends on the composition of the nominating committee, or Advisory Board. This means that the government has considerable ability to steer the results of the nomination process. If it appoints to the Advisory Board those whose expertise has been developed in well-known and well established circles of law and legal practice then the short-list of persons suitable for consideration for appointment would not differ much from the list that, for example, a committee of lawyers made up of adherents to the governing party would produce.
42 UNBLJ RD UN-B [VOL/TOME 68
There are, though, two potential advantages that could flow from having a nominating committee. First, if the members of the committee appointed by the government have no ties to, no public adherence to, and no wish to gain the favour of, the governing party, its members will likely ignore party affiliation in selecting persons to nominate and this would serve to enhance public confidence in any appointment. The other benefit that can arise from a nominating committee occurs if the committee is comprised of persons from non-establishment legal and social communities. Sometimes nominating committees are specifically charged with finding candidates from communities that have not been widely represented in judicial appointments at any level and, in fact, that is an explicit part of the mandate given to the Advisory Board that was created by the August 2016 process.14
The question of representativeness is always central to public bodies’ legitimacy. The only firmly established element of representation with respect to the Supreme Court has been provincial or regional representation. Interestingly enough the Terms of Reference for the Advisory Board did not include geographic representation. What is identified is the importance of bilingualism, expressed in the 15 form of stating that it is a requirement for appointment. The Terms of Reference 2017 CanLIIDocs 175 also stated that in making its list the Advisory Board should take account of the government’s desire to achieve a Supreme Court that is gender-balanced as well as one that reflects Canadian diversity including linguistic, ethnic and minority communities and gender identity and sexual orientation.16 Judicial diversity in a nation’s highest court, especially when it is a relatively small court – normally for reasons of allowing full-court collegial decision-making, as is the case in Canada – is a tricky aspiration. This is not because the idea of representativeness must take second place to notions of judicial merit and institutional legitimacy, since the two goals are hardly unrelated, but because the size of the court inevitably makes such representation partial and only achievable in any comprehensive way cumulatively over very long periods of time – if ever. When the very strong tradition of provincial or regional representation is added to the diversity representation goal there may be very little realism behind this declaration of intent.
This is not to deny that institutional legitimacy in any public body depends on all of the members of a society that are subject to its jurisdiction and its decisions being able to see that persons of their identity are eligible to serve in that institution and are legitimate aspirants for inclusion in the state’s structures. It should also be
14 Three of the seven persons that Prime Minister Trudeau appointed to the Advisory Board that was created under the August 2, 2016 Court Judicial Appointment Process, ibid, could be said to have come from – or, at least, represent non-traditional legal communities. That document said that the Advisory Committee was to identify candidates who are representative of the diversity in Canada. However, it might be said that since the stated purpose of the new process is to continue the tradition of appointing “only the most exceptional and impressive individuals to the Court …[to] ensure that the best, most well- qualified people … are named to Canada’s top court”, this could be taken to confirm the priority of high standing in traditional legal communities.
15 Office of the Commissioner for Judicial Affairs Canada, “Terms of Reference of the Advisory Board” (2 August 2016), online:
16 Ibid at 8(f). 2017] VIRTUES AND SHORTCOMINGS 43 remembered that the application of general truths, especially perhaps the general truths expressed in a constitutional declaration of rights, are always contingent on the actual social needs and contexts that are present within the particularities of any application. It is judicial understanding of contexts, not the general precepts, that seem most at question, and when only one set of personal identities has access to that sensitive intellectual mix of fact and law, it is inevitable that uncertainty over judicial neutrality will arise. This state of awareness of representational need has clearly been established when it comes to gender; it is inconceivable, I believe, that the Supreme Court will ever have less than three members who are identified as female or less than three who are identified as male. The goal of a broader degree of representativeness expressed in the Terms of Reference clearly has resonance, even if it not likely to become a determining force in the political practice of making Supreme Court appointments.
In Canada, the questions of geographic and language representation exist at a very high level of sensitivity. Certainly, Prime Minister Trudeau discovered, following the issue of his new process for Supreme Court appointments, that federalism and the recognition of the Canadian provinces and regions in making such 2017 CanLIIDocs 175 appointments are not trifling matters. It seems unlikely that any of the four regions represented in Court composition will graciously accept the loss of a seat to make room for the appointment of, for example, an Indigenous person. Such an appointment will, one thinks, need to take place within the existing allocation of regional seats. The force of the claim of the delegitimizing effect of ignoring the existing regional and provincial representation was evident in the reaction to the Terms of Reference. However, the actual appointment of Justice Rowe from Newfoundland prevented the appointment from becoming precedent breaking.
Bilingualism has become so firmly entrenched in national governance that the requirement in the Terms of Reference that nominated persons be functionally bilingual is hardly surprising. But it is a highly significant shift in the essential conditions for appointment. The requirement is not misplaced in light of Canada’s national solidarity needs, but its effects will be dramatic. It disqualifies from appointment many, many fine lawyers and jurists who could serve with distinction and whose judicial corpus could enhance Canadian jurisprudence. It will tend to limit those who can be appointed to the Court to those Anglophones who have grown up in, or attended university in, a bilingual environment, or those who have benefited from intensive French language training either as superior court judges or as senior federal public servants. The requirement represents the fear of potential discrimination against francophone lawyers who file facta and make oral argument in French, but it does so through ordering the exclusion of a large number of Anglophone judges and lawyers, which in itself, is a possible threat to the Court’s legitimacy.
44 UNBLJ RD UN-B [VOL/TOME 68
The final feature of the 2016 appointment process relates to parliamentary engagement with the Supreme Court appointment process.17 This process has three parts. The first is that the Minister of Justice and the Advisory Board chair would appear before Parliament to review the selection process. It seems unlikely that this meeting would engage any specifics with respect to the actual list of nominees or to the person appointed. The second element involves the Minister of Justice and the Chair of the Advisory Board appearing before the House of Commons Justice committee to explain how the nominated person satisfies the criteria for appointment. The third element is a moderated question and answer session with the nominee and some parliamentarians.18 As a device to produce a process for reviewing the legitimacy of the exercise of executive power in making a Supreme Court appointment, it would be hard to imagine anything less rigorous or effective. This element was, it would seem, included to avoid abandoning the legislative hearing process instituted by Prime Minister Martin and first used in the nomination of Justice Rothstein, but, by no means, enhancing it. It seems included without conviction that this process is necessary or prudent. This is a pity.
It is clear that the Supreme Court’s visibility and reputation will be 2017 CanLIIDocs 175 increasingly formed around its decisions relating to the rights and protections included in the Charter of Rights. These decisions will raise questions of judicial method, as well as questions of the relationship between judges’ values and the disposition of cases. The public has a right to know how the judges of the Supreme Court consider the factors of decision-making – the content and nature of the constitutional text, the precedents that bear on the issue before the Court, the governmental program that, it is argued, infringes rights, the context of the application of the challenged law (or application) and the understanding of the values and purposes that undergird constitutional recognition of the right that is being claimed. These are questions that speak to the jurisprudential frame of mind of the person nominated. Answers and explanations will differ from judge to judge and there is no bright line delineating which approaches are legitimate and which are not. But there is, even without clear standards for assessing what is a reasonable jurisprudential approach, value in this conversation between persons nominated and parliamentarians. Such a process illuminates the nature of constitutional standards and the nature of constitutional judgment – an awareness that serves both the political branch and the judicial branch well. Furthermore, there may be some attitudes that if they were to guide a nominee’s decision-making would violate basic entrenched constitutional values, or would reflect notions about society and social relations that markedly deviate from ideas of inter-personal respect and the principles of liberty and fair process that stand at the heart of liberal democracy. While it is unlikely that this would often (or ever) be the case, the check of a parliamentary
17 For consideration of legislative review of persons named for appointment to the Supreme Court of Canada, see John D Whyte, “Political Accountability in Appointments to the Supreme Court of Canada” (2016) 25 Constitutional Forum 109 at 112–116.
18 Judicial Appointment Process, supra note 13: “A number of Members of Parliament and Senators – from all parties – will also have the opportunity to take part in a Q&A session with the eventual nominee, before she or he is appointed to the Supreme Court of Canada.”
2017] VIRTUES AND SHORTCOMINGS 45 hearing that will explore the fit between the values of a nominee and the core values of the Canadian legal order19 serves as a restraint on the executive power of appointment.
It could well be that the governance climate in Canada is not ready for this sort of exchange between members of the judiciary (or persons nominated to become Supreme Court judges), but it seems inevitable that the ever-evolving dynamic of political relationships will, at some point, come to recognize that it would be mature statecraft if this executive power were held accountable through a process of parliamentary review of decisions over who should sit on the Supreme Court.
The process announced by Prime Minister Trudeau was an exercise of political innovation that responded to a newly developed political sensibility with respect to the Supreme Court. It deserves respect for being responsive to that development. But the process for making Supreme Court appointments and the values that should be reflected in that process, require considerably more thought. 2017 CanLIIDocs 175
19 One example of such a conflict could be the view that the terms of the constitution must be applied from an originalist perspective, or on what is sometimes known as a textualist basis (notwithstanding that neither an original understanding nor textual clarity are appropriate theories of interpretation to apply to general concepts such as “in accordance with the principles of fundamental justice”). This non-purposive approach to interpretation and application deviates from the Canadian precept of constitutional application being based on dynamic and contextual interpretations.
THE FALSE DICHOTOMY BETWEEN REGIONAL REPRESENTATION AND OTHER FORMS OF DIVERSITY: REIMAGINING A REPRESENTATIVE COURT
A. Wayne MacKay, C.M., Q.C.
The 2016 Ivan C. Rand Memorial La Ivan C. Rand Memorial Lecture Lecture was given in the immediate 2016 fut prononcée immédiatement wake of the appointment of Justice après que le Juge Malcolm Rowe de Malcolm Rowe of Newfoundland and Terre-Neuve-et-Labrador fut nommé Labrador to the Supreme Court of à la Cour suprême du Canada. Cette Canada which allayed many fears nomination élimina plusieurs concerning the elimination of inquiétudes entourant la représen- Atlantic Region representation at the tation de l’Atlantique à la Cour Supreme Court of Canada. Professor suprême. Le Professeur Peter H. 2017 CanLIIDocs 175 Peter H. Russell spoke of the Russell a discuté du rapport entre la inception of regional representation représentation régionale et la and the transient reconstruction of reconstruction du processus de the appointment of Supreme Court nomination des juges de la Cour justices which, in the wake of the suprême du Premier Ministre new Trudeau approach, inspires Trudeau et les questions soulevées further questions of what exactly a par les changements et les Supreme Court justice should be and circonstances entourant nos concep- where regional repre-sentation fits tions du juge idéal pour siéger à la amidst the structure of the judiciary Cour suprême et la place de la whilst also working towards représentation régionale au sein de diversity. la structure judiciaire tout en travaillant vers la diversité.
The appointment of Supreme Court Justice Rowe was significant in a multitude of ways. The Ivan C. Rand Memorial Lecture of 2016, delivered by Professor Peter H. Russell, highlighted many procedural changes that have increased transparency in the process of appointing Supreme Court justices.1 The Trudeau government has made important additions to the process. This was especially apparent in the implementation of the Advisory Board for Supreme Court Appointments (“the Advisory Board”), a nominating body officially dedicated to the processing of
* Professor of Law, Schulich School of Law at Dalhousie University and Yogis and Keddy Chair in Human Rights Law. Dr. MacKay gave the 2014 Rand Lecture at UNB, entitled “Law as an Ally or Enemy in the War on Cyberbullying: Exploring the Contested Terrain of Privacy and Other Legal Concepts in the Age of Technology and Social Media” (2015) 66 UNBLJ 3. The author wishes to acknowledge the excellent research and drafting assistance of Nika Booker-Grandea and Kristen Stallard, third year students at the Schulich School of Law.
1 Peter H Russell, “Selecting Supreme Court Justices: Is Trudeau’s Sunny Way a Better Way?” (2017) 68 UNBLJ 3. 2017] REIMAGINING A REPRESENTATIVE COURT 47 nominations, as well as actively seeking nominees. The formalization of a recognizable body that has a mandate to find candidates is an improvement from the previous process, which involved components such as publicly-broadcasted meetings with the federal legislature and already chosen candidates. It was, at the very least, educational. However, more enriching opportunities have developed and the mystery surrounding the emergence of these nominees has diminished.
The selection process that resulted in the appointment of Justice Rowe evoked a newfound uncertainty within Canadian society about regional representation in the Supreme Court of Canada. There was a concern that the Atlantic provinces would have no representation. It was known that there were Atlantic Region nominees but there was a sense of anxiety that this representation would be removed to focus on other aspects of diversity. The removal of this representation in the Supreme Court would have been very contentious as the Atlantic region has its own challenges and familiarity with those challenges is important. Those who live in the Atlantic region may otherwise lose confidence that their interests are being considered at all.
2017 CanLIIDocs 175 It is fortunate that Prime Minister Trudeau, after some public uncertainty, appointed a Supreme Court justice from the Atlantic region. It is especially fortunate that Justice Rowe is well-versed in socioeconomic issues and the specific difficulties faced by Aboriginal peoples in the Atlantic Region. In his application, Justice Rowe conveyed two interesting considerations.2 His application described his experience of watching Newfoundland and Labrador become a more unified part of Canada and more prosperous – a stark contrast to the poor and fractured Newfoundland and Labrador he had seen in his earlier life. He also mentioned that through his previous experience he had become familiar with the challenges faced by the First Nations and Inuit in Labrador. These are two challenges that are unique to Newfoundland and Labrador and would certainly be an asset in the Supreme Court of Canada.
There was a strong movement to bring Aboriginal representation to the Supreme Court by appointing an Aboriginal Justice. Professor Russell stated that: “It would be very difficult to find a well-qualified, bilingual, Aboriginal jurist in Atlantic Canada.”3 Whether or not that is the case, it still brings up a significant issue: not necessarily that of being unable to find an individual with these qualities, but rather that our country needs these three qualities represented and the best compromise would be to find them in one individual. It is problematic that one must be chosen over the other if these ideal elements cannot be found in a single individual. It may seem unlikely to find a qualified candidate that can champion both regional representation and diversity existing in the Atlantic region but that difficulty makes finding a qualified candidate from the region more pressing. Although it was stated by Professor Russell that an individual who meets the stated criteria would be
2 John Paul Tasker, “Trudeau nominates Newfoundlander Malcolm Rowe to Supreme Court”, CBC (17 October 2016), online:
3 Russell, supra note 1 at 11.
48 UNBLJ RD UN-B [VOL/TOME 68 hard to find, it is necessary to at least attempt to seek those individuals and even more important to consider what they represent.
A loss of representation in the Atlantic region in the wake of an abandoned convention because of changing needs for proper representation would also bring up concerns about diversity. The region would additionally face an abandonment of the representation of the even more underrepresented minorities within the region – those who were considered unlikely to be found. The mutual exclusivity of diversity and regional representation is deleterious. To search solely for one or the other fails to address intersectionality. Regional representation is not valuable for the physical or geographical components; it is valuable to bring insight that represents the reality of Canadians.
A lingering question is how long it might take before the Supreme Court sees justices who are familiar with Aboriginal issues and experiences that also have the lived experience. It also brings up the question of whether anyone paid heed to the existence of Aboriginal candidates in the Atlantic region at all. Professor Naiomi
Metallic is an exceptional example and a strong candidate for the future. Though 2017 CanLIIDocs 175 Professor Metallic is two years short of meeting the criteria, she is not alone. There are a few impressive individuals who can challenge this perceived dichotomy between regional representation and diversity. Achieving this is especially important given the ever-changing role of the Supreme Court of Canada.
The Supreme Court of Canada continues to be an increasingly powerful institution. Many Canadians are familiar with the recent case of Carter v Canada (AG)4 on physician-assisted death. This was a policy decision affecting Canadians with no less weight than if the legislature had passed it. The difference is that the decision-makers constructed this policy through the administration of justice as opposed to adherence to political convictions. Another distinction is that Supreme Court justices hold power within the governance framework longer than the person who appoints them – that person being the Prime Minister. The leader of the country is ultimately responsible to choose these jurists, who may sit until the age of 75, and, in that sense, that legacy will outlast the Prime Minister’s governance. If there is an issue with an appointment, it still has the potential of surviving numerous governments.
If the tactics of the federal government are clearer but the system stays the same, then there is still a barrier to diversity that is very difficult to address. A concrete understanding of what is envisioned in terms of diversity in the Supreme Court and how to best achieve it must be decided. Jennifer Nedelsky’s notion of the “enlargement of the mind” lends itself very well to this challenge. An open mind is a mind that can reflect all aspects of society. Nedelsky notes: “To understand judicial impartiality we must ask who judges are, and with whom they imagine themselves to
4 Carter v Canada (AG), 2015 SCC 5, [2015] 1 SCR 331.
2017] REIMAGINING A REPRESENTATIVE COURT 49 be in conversation as they make their judgments. Whom do they imagine persuading and on whom do they make claims of agreement?”5
With Nedelsky’s words in mind, the idea becomes that diversity on the bench helps Canadians not only through a representation of their needs, but also by allowing a sense of cohesion to form from different experiences and perspectives within the judiciary. The selection process becomes even more significant because of the potential reframed sense of accountability amongst the justices that preside. With appropriate representation, those who feel represented have at least some sense of security that their general experiences or the trends that surround them will be considered amidst difficult decisions. These appointed justices signify more than one individual providing representation of different pockets of the Canadian population. They also passively inform and shape the cognitive frameworks of other jurists and give rise to an open-mindedness that expands the accommodation of Canadian interests.
Canadians who come before the justices of the Supreme Court are vulnerable and subject to the thought processes of these highly-esteemed decision- 2017 CanLIIDocs 175 makers. These decision-makers may have very different lived experiences. Following Nedelsky’s reasoning, the decision, based on the context provided, is affected by a collective understanding amongst the presiding justices. This supports impartiality by ensuring different perspectives are continuously considered. With the power bestowed upon the Supreme Court of Canada, there is a significant responsibility to the Canadian people. If the Supreme Court is the last chance for justice, then there should be the legitimate expectation that this Court will be the closest approximation to a Canadian’s best interest balanced against the larger public interest.
The candidates for appointment to the Supreme Court that come to the fore are often lauded for their achievements. They are presented to the Canadian public with the confidence of the government behind them. However, one might ask how large the pool was to begin with and under what circumstances these individuals had been noticed. The Trudeau government has made significant improvements in making these processes known, as was described in Professor Russell’s lecture. The appearance of a nominating body, the Advisory Board, has at least removed some concerns about the private nature of the search for candidates that preceded this change.
If it is accepted that the impartiality of Supreme Court justices is an asset rather than a compromise of the standard of neutrality, which has been challenged in cases such as R v S (RD)6 and Arsenault-Cameron v PEI,7 then the qualifications should be seen in light of that. The current process, as modified by the Trudeau
5 Jennifer Nedelsky, “Embodied Diversity and the Challenges to Law” (1997) 42 McGill LJ 91 at 107.
6 R v S (RD), [1997] 3 SCR 484, 161 NSR (2d) 241 [RDS].
7 Arsenault-Cameron v Prince Edward Island, [2000] 1 SCR 3, 2000 SCC 1 [Arsenault-Cameron].
50 UNBLJ RD UN-B [VOL/TOME 68 government, emphasizes bilingualism, has acknowledged the necessity for an Aboriginal jurist to be appointed to the Supreme Court and has left regional representation as a question mark. Justice Rowe may be representative of the Atlantic provinces but it remains to be seen whether the next vacancy will follow suit with regional representation, as it has no legal foundation as a convention in the traditional view8 and there are signs of a shifting landscape.
There have been significant departures from the traditional view that could change the status of regional representation within the Supreme Court of Canada. The constitutional principle of federalism, as explored Reference re Secession of Quebec,9 is worthy of attention. Another consideration is the legacy of the Reference re Supreme Court Act, ss. 5 and 610 concerning Marc Nadon, as well as the Reference re Senate Reform11 which set forth the necessity of examining historical context and purpose when interpreting the Constitution Act, 1982.12
Additionally, interpretation of section 41(d) of the Constitution Act could determine that the selection process is a part of “the composition of the Supreme 13 Court of Canada”. There are available approaches to change the status of the 2017 CanLIIDocs 175 regional representation convention and move it away from being politically vulnerable. Currently, it remains contested and unclear. There are no set rules or regulations for the appointment process of justices of the Supreme Court, except that the Prime Minister is the ultimate authority on appointments. There is also the statutory requirement that there be three justices from Quebec.14
Regional representation has been a consistent practice; however there is still an uncertainty on intersectionality. The mutual exclusivity of diversity and regional representation ignores the aspects of diversity that are encompassed within regional representation. For example, Justice Rowe spoke of experiences that expanded his insight into the challenges of poverty. Those surrounding the lives and experiences of those of lower socioeconomic status in the Atlantic region specifically. According to Statistics Canada,15 surveys in the year 2014 indicate that an average of 15.3 per cent of persons in the Atlantic provinces earned below low income indicators, after tax. This is a statistic based on households across the four Atlantic provinces. This
8 Reference Re Resolution to amend the Constitution, [1981] 1 SCR 753, 34 Nfld & PEIR 1.
9 Reference re Secession of Quebec, [1998] 2 SCR 217, 161 DLR (4th) 385.
10 Reference re Supreme Court Act, ss 5 and 6, 2014 SCC 21, [2014] 1 SCR 433.
11 Reference re Senate Reform, 2014 SCC 32, [2014] 1 SCR 704.
12 Constitution Act, 1982, being Schedule B to the Canada Act 1982 (UK), 1982, c 11.
13 Ibid at s 41(d).
14 Supreme Court Act RSC 1985, c 5-26, s 6.
15 Statistics Canada, “Low income statistics by economic family type, Canada, provinces and selected census metropolitan areas (CMAs)”, CANSIM Table 206-0042 (Ottawa: Statistics Canada, 2014), online:
2017] REIMAGINING A REPRESENTATIVE COURT 51 combined average is higher than those of the non-Atlantic provinces. This statistic was rivalled only by Manitoba, which sat at 15 per cent. If regional representation can truly allow for these specific regional difficulties to be considered, then diversity is achieved on some level.16
If diversity is a necessary consideration to the point of excluding regional representation, this raises the issue of what is most representative of the Canadian public. Achieving representation is a common goal. The importance of how the Canadian population is represented within the Supreme Court of Canada deserves a frank discussion on what is absent from the appointment process. There was a very real potential that the Supreme Court of Canada appointments process would turn the page on regional representation and seek a different way of ensuring the best- qualified and most representative judiciary. However, the concern of false dichotomies and of allocating more value towards one approach over another is something worthy of reflection.
A jurist that can relate reasonably to groups of people through their own personal experiences is a positive addition to any court. However, the question of 2017 CanLIIDocs 175 what instills the most faith within the Canadian population arises. It is the Canadian people who give rise to the validity and authority of the courts. Confidence is essential. Justice Sopinka once drew a very relevant and useful analogy between judges and banks: “Our justice system is in some respects like the banking system. It only works if people have confidence in it …”.17 If there is a loss of confidence in a bank, the system ceases to function. In the case of the judiciary, when confidence is lost, its standing as a reputable institution in society is challenged.
Reimagining the judicial appointment framework to align it with the needs of the Canadian population is a significant step towards maintaining a relationship of confidence. Diversity has been a significant and pressing consideration; however the development of an appropriate framework leads to questions of implementation. If Canada were to try and have greater representation of its population within its highest Court, then is it time we turned to the lived experience as opposed to the experience of exposure and understanding? There is a case to be made for the power of relatability and the value of shared experience. Is an individual more likely to cast their lot with someone to whom they can relate or someone who can relate to them, or both? None of this is to suggest that overall merit should be eclipsed as the main qualification but only that the above is one component of this.
Having a legal education and proficiency in the application of legal skills remains at the forefront for becoming a justice of the Supreme Court. Professor Russell does note that there is diversity within the current Supreme Court judiciary.
16 I would like to acknowledge that my fellow commentator, Professor John Whyte, raised this issue quite eloquently.
17 Graham Fraser, “Ethnic roots won’t sway him, Sopinka says at swearing-in”, The Globe and Mail (24 June 1988), ProQuest Historical Newspapers: The Globe and Mail (database accessed through the Dalhousie University Library).
52 UNBLJ RD UN-B [VOL/TOME 68
There also is movement towards making a more diversely experienced judiciary. However, the comment by Professor Russell returns: it is very difficult to find someone who is Aboriginal, bilingual and from Atlantic Canada. That speaks to societal challenges that are not irrelevant. Given the extensive diversity within Canada, all groups cannot be represented on a nine-person Supreme Court. Not every experience is the same. It is a lofty and respectable goal to appoint a more diverse judiciary to the Supreme Court of Canada. A determination of what, exactly, the vision for the justices of the Supreme Court should be is vital for a successful selection of the next justice of the Supreme Court when the next opening occurs.
Just as RDS18 and Arsenault-Cameron19 contended with how much a member of the judiciary can use their own experiences, the question must again be posed but with the qualifier of how much value there is in the experiences they are able to bring. This is clearly not quantifiable, nor should it be. As the Trudeau process begins to push towards background and representation, the consideration of how judges should judge becomes infinitely more complex. It is not a deterrent but a challenge to the constructs of what adjudication is and its relationship to where the
Canadian people stand. It is an exercise in understanding what breeds confidence. It 2017 CanLIIDocs 175 is a matter of considering relatability and trust in our institutions, including the Supreme Court of Canada.
The greatest issue faced before the appointment of Justice Rowe was the creation of an “either/or” situation for two significant populations. Whether representation can be better approximated as new empty seats arise is a matter of deciding what is needed in a judge, as well as what is required of Canadian governance. There must also be a recognition of the importance of the role of justices of the Supreme Court and how the role of the Supreme Court of Canada has expanded. The weight of the decisions made in the Supreme Court, in conjunction with the movement to a different system of appointing Supreme Court justices, calls for a reimagined approach.
As the appointment process evolves, it is important to look at the core of the institution it is impacting. Obtaining the desired results of being a more representative Court and instilling continued confidence, echoes the Edwards v Canada (AG)20 legacy of the “living tree” approach. When something is of such value to the Canadian population, adaptability and inclusivity are necessary. Meaningful consideration should be given to the criteria for appointing justices of the Supreme Court of Canada. A careful adaptation that is not simply needs-based but also demonstrates open-mindedness and accountability to Canadian society may very well breed confidence in our highest Court. These changes move towards giving Canadians the respect and consideration each Canadian deserves. It is such an effort
18 RDS, supra note 6.
19 Arsenault-Cameron, supra note 7.
20 Edwards v Canada (AG), 1929 UKPC 86, [1930] AC 124.
2017] REIMAGINING A REPRESENTATIVE COURT 53 towards more inclusive consideration which builds institutions that reflect what Canadians stand for.
2017 CanLIIDocs 175
A JUDICIARY CLEAVED: SUPERIOR COURTS, STATUTORY COURTS AND THE ILLOGIC OF DIFFERENCE
The Hon. Justice David Stratas
In one important respect, we in the Federal Courts are lucky. My Court, the Federal Court of Appeal, regularly hears applications and appeals here in New Brunswick. We regularly see the beauty of this province and the warmth of its people. We also see the interesting nature of the cases that arise and your many excellent counsel. We are constantly enriched by the insights of the great professors at the University of New Brunswick law school. Truly New Brunswick is a special place.
Here there is a great legal tradition of excellence. One need only think of legendary jurists from New Brunswick like Ivan Rand, Gérard La Forest and Michel Bastarache, intellectually at the top of their class, in every respect hard-working and 2017 CanLIIDocs 175 exceptional.
When I think of New Brunswick, though, I cannot help but recall, with deep affection, two other New Brunswickers, two wonderful judges who served in the Federal Courts system, two who recently and prematurely succumbed to cancer: Chief Justice Edmond Blanchard and Justice Carolyn Layden-Stevenson. Hard working, selfless and brilliant, these two moved from backgrounds of devoted public service—in one case an elected politician; in another, a school teacher—and offered themselves for national judicial service in the Federal Courts system. Truly, they distinguished themselves from coast to coast to coast. Their jurisprudence continues to shine brightly, and as exemplars of great character they remain in our memories as role models. Their lives were well lived.
My lecture today is very much about an issue related to these judges. They served in the Federal Courts system. That’s a system of statutory courts, much like the Provincial Courts system in New Brunswick.
In fact, across Canada, thousands of judges serve in statutory courts like these, more than those that serve on superior courts. And statutory courts decide more cases than superior courts.
Justice, Federal Court of Appeal and Courts Martial Appeal Court of Canada, LL.B. (Queen’s University, Kingston), B.C.L. (Oxon), LL.D. (honoris causa, Queen’s University, Kingston). This paper is a modified version of the 2017 Viscount Bennett Lecture which I delivered at the University of New Brunswick Faculty of Law, February 9, 2017. I acknowledge and thank Professor Emeritus David Mullan for his comments on an earlier draft, and the editorial assistance of Bohdana Tkachuk and Miriam Clouthier, law clerks at the Federal Court of Appeal. The views expressed are mine alone. Like the many sitting judges who write extra-judicial works and comment on the jurisprudence, I remain open to correction and persuasion on any point in this paper, especially in the courtroom. My views evolve and sometimes change following discussion and debate, especially in court when faced with real-life situations. 2017] A JUDICIARY CLEAVED 55
Yet today, as things stand, statutory courts are treated as the lesser cousins of superior courts. In some respects their powers are needlessly shackled, to the detriment of the people they serve.
This needn’t be so. And this shouldn’t be so. Let me explain.
In Canada, we have superior courts like the New Brunswick Court of Queen’s Bench and the New Brunswick Court of Appeal. Then there are statutory courts, the New Brunswick Provincial Court, the Federal Court and the Federal Court of Appeal. All operate in New Brunswick.
Superior courts are fully empowered. For example, superior courts have a full ability to award any constitutional remedies under section 24 and 52 of the Constitution Act, 1982.1
The power to declare legislation invalid under section 52 matters, deeply so. Let me illustrate. Suppose a statute sets out a police power of dubious constitutionality. The police have the power across the province. Suppose that it 2017 CanLIIDocs 175 likely offends the right against unreasonable search and seizure in section 8 of the Charter.2
If you go to any courtroom of New Brunswick’s superior court, the Court of Queen’s Bench, in Bathurst, Woodstock, Edmunston, Campbellton, Miramichi, Moncton or Fredericton and if the court declares the police power unconstitutional, it is invalid not just in that locality, but everywhere in New Brunswick. The ruling becomes the law all across New Brunswick. Right across the province, the police must comply right away. All New Brunswickers are protected and treated alike.
Now take the New Brunswick Provincial Court, a statutory court. Same police power. Same challenge. But the challenge is brought in Edmunston in the Provincial Court there. It reaches the same result, word-for-word the same.
But the ruling doesn’t apply across the province. It applies only to the particular case. The Provincial Court does not have the power to declare legislation unconstitutional. It can only disregard laws that are unconstitutional in the particular cases before it.
This means that the law, adjudged unconstitutional, remains on the books across the province. All across the province, the police can continue to use the unconstitutional power. Even in the particular locality where a ruling of unconstitutionality was made, that ruling does not bind a later judge in the locality. The ruling applies only in that particular case.
1 Constitution Act, 1982, being Schedule B to the Canada Act 1982 (UK), 1982, c 11.
2 Canadian Charter of Rights and Freedoms, Part I of the Constitution Act, 1982, being Schedule B to the Canada Act 1982 (UK), 1982, c 11 [Charter].
56 UNBLJ RD UN-B [VOL/TOME 68
Sounds strange? This is the law as set out in the 2016 decision of the Supreme Court in R v Lloyd.3
Lloyd was convicted in the British Columbia Provincial Court of possession of drugs for the purpose of trafficking. Because he had an earlier conviction for a similar offence, he was subject to a mandatory minimum sentence of one year imprisonment. The Provincial Court judge issued a declaration. He declared that the mandatory minimum sentencing provision offended the guarantee against cruel and unusual punishment under section 12 of the Charter and was not justified under section 1 of the Charter.
The Supreme Court agreed. It held that the mandatory minimum sentencing provision was unconstitutional. But it also held that the Provincial Court, as a statutory court, did not have the power to make a declaration of invalidity. All the statutory court could do is rule the provision unconstitutional and decline to apply it in the case before it, but nothing more.
The Supreme Court acknowledged that this left the unconstitutional 2017 CanLIIDocs 175 provision alive in other cases across the province or in future cases that arise, even before the same court. It confirmed that only a superior court could make a ruling of province-wide effect through its power to make declarations.
The effect of this is to leave an unconstitutional law in force across the province. And if the Crown does not appeal the decision, there is no risk of a province-wide declaration of unconstitutionality from a superior court. The Crown remains free to reargue the point in any Provincial Court in the province.4
As a practical matter, this means persons in one part of a province can enjoy different Charter protections and can be subject to different laws than those in another part of the province. Under the constitutional principle of the rule of law,5 laws should apply to similarly situated people in a similar way. They should not apply depending on where you live.
By far, most criminal prosecutions in this country—and thus, most issues involving the all-important criminal law protections of the Charter—take place in provincial courts. The inability of provincial courts, as statutory courts, to have their
3 2016 SCC 13, [2016] 1 SCR 13 (6:3 majority).
4 Recently, in another case, R v Sharkey, 2014 ONCJ 437, Justice Paciocco, then of the Ontario Provincial Court, considered this unjust. He wrote that because “the decision rendered by a judge after solemn consideration and full argument cannot be used as a persuasive precedent by other courts,” the Crown “is in a position to isolate a Charter challenge that it loses by not appealing that decision to a superior court” and relitigated in another provincial court all over again. In my view, he’s right. But the Lloyd case is the law. See also to similar effect R v Michael, 2014 ONCJ 360.
5 Preamble to the Constitution Act, 1982.
2017] A JUDICIARY CLEAVED 57 findings of unconstitutionality apply across the province is a significant hole in the coverage of the Charter.6
So why are provincial courts as statutory courts treated so differently from superior courts?
The difference is merely one of historical oddity. The superior courts are the heirs of the Royal Courts in England, such as the High Court of Chancery. Those English courts had inherent jurisdiction. Thus, so do the superior courts. Historically, superior courts have had the power to grant declarations. So it has been said that statutory courts do not have that power unless their statutes give them it.
I call the difference an oddity because in pith and substance there is no difference between superior courts and statutory courts. Institutionally and functionally, courts are courts. The judges who staff them are judges. Their judgments are judgments. For those convicted in either type of court, jail is jail.
Finally, and perhaps more importantly, in pith and substance both are 2017 CanLIIDocs 175 statutory courts. Both have statutes that set out what they can do and add powers that historically did not exist.7
Now let’s turn to the superior court’s inherent jurisdiction and examine it more closely. First, it’s important to get our terms right.
The question of jurisdiction, properly defined, is different from the question of the powers that can be exercised.8 Jurisdiction—literally from the Latin, speaking the law—means the ability of a court to speak on a particular subject-matter; in other words, to consider it. Once a court has the ability to consider a subject-matter, the next question is what it can do while considering that subject-matter; in other words, what powers can it exercise?
6 It would also seem to be against the notion in R v Ferguson, 2008 SCC 6, [2008] 1 SCR 96, that constitutional remedies should not be granted on a case-by-case basis and that a constitutionally invalid provision should be treated as unconstitutional across the board.
7 In New Brunswick, see Judicature Act, RSNB 1973, c J-2 (superior courts); Provincial Court Act, RSNB 1973, c P-21 (provincial courts).
8 See Watson v Clarke, [1990] 1 NZLR 715 at 720. See also Marcelo Rodriguez Ferrere, “The Inherent Jurisdiction and Its Limits” (2013) 12 Otago L Rev 107; Jessica Liang, “The Inherent Jurisdiction and Inherent Powers of International Criminal Courts and Tribunals: An Appraisal of Their Application” (2012) 15 New Criminal L Rev 375 at 379–380. The two are frequently confused, as is noted in William H Charles, “Inherent Jurisdiction and its Application by Nova Scotia Courts: Metaphysical, Historical or Pragmatic?” (2010) 33 Dal LJ 63 at 64. Martin Dockray has said it is “a difficult idea to pin down” and “[t]here is no agreement on what it is, where it came from…and what it can be used for” (“The Inherent Jurisdiction to Regulate Civil Proceedings” (1997) 113 LQR 120 at 120). Rosara Joseph has said: “The courts’ treatment of inherent jurisdiction and inherent powers has been fraught with confusion and misapplication. Many of the judgments dealing with inherent jurisdiction have conflated the distinct concepts of inherent jurisdiction and inherent power” (“Inherent Jurisdiction and Inherent Powers in New Zealand” (2005) 11 Canterbury L Rev 220 at 221).
58 UNBLJ RD UN-B [VOL/TOME 68
An influential article by Sir Jack Jacob 45 years ago—adopted uncritically in Canada9—describes “inherent jurisdiction” of a court as a “residual source of powers, which the court may draw upon as necessary whenever it is just or equitable to do so, in particular to ensure the observance of the due process of law, to prevent improper vexation or oppression, to do justice between the parties and to secure a fair trial between them.”10
But here Jacob is not really talking about jurisdiction, properly defined. Rather, he is speaking of powers that a court must have by virtue of being a court. This is nothing unique to superior courts. All courts, even statutory courts, have these powers.11
With the proper definition of jurisdiction in mind, the inherent jurisdiction of the superior courts means nothing more than a residual jurisdiction over matters that cannot be dealt with by others.
So said the Supreme Court in the 1998 decision in Liberty Net.12 There it said that superior courts have a “residual jurisdiction” and confirmed that the 2017 CanLIIDocs 175 doctrine of inherent jurisdiction means nothing more than that.13 In its words, this residual jurisdiction of the superior court “does not operate to narrowly confine a statutory grant of jurisdiction” to another court, nor does it say anything “about the proper interpretation of such a grant.”14
Liberty Net put the inherent jurisdiction of superior courts in its proper place. Superior courts do have a time-honoured power to grant declarations. But that
9 See e.g. Ontario v Criminal Lawyers’ Association of Ontario, 2013 SCC 43 at para 20, [2013] 3 SCR 3; Conseil scolaire francophone de la Columbie-Britannique v British Columbia, 2013 SCC 42 at paras 72– 74, [2013] 2 SCR 774; R v Caron, 2011 SCC 5 at paras 24–34, [2011] 1 SCR 78.
10 Jack IH Jacob, “The Court’s Inherent Jurisdiction” (1970) 23 Current Leg Probs 23 at 51.
11 All of the cases in supra note 9 and most of the cases that rely upon Jacob’s article do so in support of the idea that courts have certain plenary powers stemming from their status as courts, powers that all courts, whether superior or statutory, have. See also the discussion in the text to notes 41–48, infra.
12 Canada (Human Rights Commission) v Canadian Liberty Net, [1998] 1 SCR 626 at para 35, 157 DLR (4th) 385 [Liberty Net]: “In my view, the doctrine of inherent jurisdiction operates to ensure that, having once analysed the various statutory grants of jurisdiction, there will always be a court which has the power to vindicate a legal right independent of any statutory grant. The court which benefits from the inherent jurisdiction is the court of general jurisdiction, namely, the provincial superior court. The doctrine does not operate to narrowly confine a statutory grant of jurisdiction; indeed, it says nothing about the proper interpretation of such a grant.” The Supreme Court here is speaking of subject-matters, not powers.
13 Ibid at para 35. The “inherent jurisdiction” was also explained as a residual jurisdiction in Peacock v Bell (1667), 1 Wms Saund 73, 85 ER 84 at 87–88: “the rule for jurisdiction is that nothing shall be intended to be out of the jurisdiction of the superior court but that which specifically appears to be so.”
14 Liberty Net, supra note 12 at para 35. In so deciding, Liberty Net does not exalt the status of superior courts under s 96 of the Constitution Act, 1867 and allows the federal Parliament to create its own fully- empowered courts under s 101 of the Constitution Act, 1867. This is perfectly consistent with clear constitutional text: s 101 operates “notwithstanding anything in this Act,” including s 96.
2017] A JUDICIARY CLEAVED 59 has nothing to with their inherent subject-matter jurisdiction. Nor does it automatically foreclose the existence or the scope of that power—express, implied or necessarily incidental—in a statutory court.
So why can’t statutory courts grant constitutional declarations of invalidity?
After all, constitutional declarations of invalidity bear no relation to the sorts of declarations granted by the Royal Courts of England as part of their inherent jurisdiction, such as declarations of right under statutes or the common law. This is hardly surprising—the United Kingdom does not have a written constitution and so its courts could not have granted declarations of invalidity.
If declarations concerning the invalidity of legislation do not emanate from the Royal Courts of England, where do they come from?
They came in part from a British statute, the Colonial Laws Validity Act,15 a statute that as far as issues of constitutional validity are concerned in substance had 16 been part of the law of Canada until 1982. That statute applied not only to Canada’s 2017 CanLIIDocs 175 superior courts. It applied to all Canadian courts, including statutory courts.
Around the time of Confederation, we had a system of superior courts and certain statutory courts, such as the local provincial courts and the federal Exchequer Court. All these courts, both superior and statutory, had to act according to law, interpreting and applying the law.
Under section 2 of the Colonial Laws Validity Act, those courts had to rule “void and inoperative” any federal or provincial laws inconsistent with those of the Parliament of the United Kingdom, including the British North America Act, 1867.17
The Exchequer Court, a statutory court operating at the federal level and the predecessor to today’s Federal Courts, recognized this power and understood that in appropriate cases it would not apply legislation that conflicted with a law of the Parliament of the United Kingdom and that decision would apply to all subsequent cases in the Court.18
15 1865 (UK), 28 & 29 Vict, c 63. See discussion in Windsor (City) v Canadian Transit Co, 2015 FCA 88 at paras 55–62, [2016] 1 FCR 265 [Windsor FCA].
16 From 1931 to 1982, s 2 of the Constitution Act, 1931 (formerly the Statute of Westminster, 1931 (UK), 22 & 23 Geo V, c 4) required that Canadian laws conform with the British North America Act, 1867 (UK), 30 & 31 Vict, c 3, in substance continuing the law on issues of constitutional validity as set out in the Colonial Laws Validity Act. See the discussion in Windsor FCA, supra note 15 at para 59.
17 Now the Constitution Act, 1982. See also the discussion in Re Manitoba Language Rights, [1985] 1 SCR 721 at 746, 19 DLR (4th) 1.
18 See e.g. Algoma Central Railway Co v Canada (1901), 7 Ex CR 239 at 254–255, rev’d on other grounds (1902), 32 SCR 277, aff’d [1903] AC 478 (PC).
60 UNBLJ RD UN-B [VOL/TOME 68
For most of Canada’s early years, the idea of a private litigant seeking a formal declaration “having as its sole object” the invalidity of a statute was “not really imaginable.”19 It was only much later, well after Canada’s superior courts and statutory courts came into being, that the law developed to embrace the possibility of such declarations.20 Declarations of invalidity have no origin in the Royal Courts of England whatsoever and in no way can be considered part of the inherent jurisdiction of superior courts.21
In 1982, section 52(1) of the Constitution Act, 1982 came into force. It provides that any legislative provisions inconsistent with the Constitution of Canada are of no force or effect. That overarching principle is not restricted to any one set of courts. It applies to all.
For good measure, under section 24 of the Charter “courts of competent jurisdiction” can make any “just and appropriate” remedies to address Charter violations. Later jurisprudence has affirmed the ability of any courts, superior courts or statutory courts, to use section 24 as long as they, by structure or function, can 22 grant the remedy. Statutory courts are structurally and functionally capable of 2017 CanLIIDocs 175 granting constitutional declarations, even, if need be, declarations under section 24.
Finally, there is the purposive, “living tree” approach to constitutional interpretation.23 What purpose does it serve to cleave the courts, empowering superior courts to fully enforce the Charter but leaving statutory courts less than fully empowered? Going back to my New Brunswick hypothetical, what purpose is served by allowing police to act under an unconstitutional law in one locality of the provincial court but not in a neighbouring one?
Unless statutory courts can grant constitutional declarations, the same issue of constitutionality can be placed over and over again before statutory courts, leading to relitigation that soaks up precious resources and impedes access to justice.24 All
19 Lazar Sarna, The Law of Declaratory Judgments, 3d ed (Toronto: Thomson Canada Limited, 2007) at 122.
20 Dyson v AG (1910), [1911] 1 KB 410, [1912] 1 Ch 158 at 168 (Eng CA); Smith v Ontario (AG), [1924] SCR 331, [1924] 3 DLR 189.
21 Recently the New Zealand Court of Appeal has confirmed that declarations of invalidity bear no relation to declarations of right. They are a different remedy altogether. The Court further confirmed that the power to make declarations of invalidity is an implied one that springs from the position of all courts vis- à-vis Parliament: there is no reason to hold that declarations of invalidity are the exclusive preserve of the superior courts. See AG v. Taylor, [2017] NZCA 215 at paras 43–109.
22 See the “structural and functional” test for determining “courts of competent jurisdiction” under s 24: R v 974649 Ontario Inc, 2001 SCC 81, [2001] 2 SCR 575; R v Hynes, 2001 SCC 82, [2001] 3 SCR 623.
23 Edwards v Canada (AG), [1929] UKPC 86, [1930] AC 124.
24 Access to justice is now a constitutional principle: Trial Lawyers Association of British Columbia v British Columbia (AG), 2014 SCC 59, [2014] 3 SCR 31 [Trial Lawyers]. And in any event, principles and
2017] A JUDICIARY CLEAVED 61 because of an idealized, inaccurate understanding of the nature of the inherent jurisdiction of superior courts.
Today we have the modern, purposive approach to constitutional interpretation. In our law, it’s pervasive like a sea. And in the middle of that sea, we have an isolated island. On that island, there is the historical oddity of superior court jurisdiction, a relic of originalism not purposivism, not discarded but worshipped.25
We must leave this island. We must keep sailing upon the sea.
Early on in the life of the Charter, the Supreme Court did just that, in the seminal case of Big M Drug Mart.26 There, the Crown argued that a provincial court had no jurisdiction to exercise a “prerogative power to declare legislation invalid.”27
The Supreme Court flatly stated that the objection “must…be rejected.” With no equivocation or qualification, it added that “it has always been open to provincial courts to declare legislation invalid in criminal cases.”28 And in support of that, it explained that because of section 52 of the Constitution Act, 1982, “no one may be 2017 CanLIIDocs 175 convicted of an offence under an invalid statute.”29
The reasoning in Big M Drug Mart could not be clearer: provincial statutory courts must have the power to make declarations under section 52 in order to fulfil the purposes of the Charter. And Big M does not stand alone: for good measure in the 1997 Judges’ Reference the Supreme Court reiterated that holding.30 practices that pose barriers to access to justice are to be discouraged: Hryniak v Mauldin, 2014 SCC 7, [2014] 1 SCR 87.
25 The unrelated jurisprudence under s 96 of the Constitution Act, 1867 delineating the “core powers” of superior courts in part may be responsible for the worshipping. Under s 96, only the federal Governor General can appoint the judges of the superior courts. Provincially-appointed bodies are impermissibly “superior courts” when they exercise the “core powers” of a superior court outside of a modern institutional context: see e.g. MacMillan Bloedel Ltd v Simpson, [1995] 4 SCR 725, 130 DLR (4th) 385 [MacMillan Bloedel]; Reference re Young Offenders Act (PEI), [1991] 1 SCR 252, 89 Nfld & PEIR 91; Re Residential Tenancies Act, 1979, [1981] 1 SCR 714, 123 DLR (3d) 554 [Re Residential Tenancies]; Tomko v Labour Relations Board (Nova Scotia), [1977] 1 SCR 112, 69 DLR (3d) 250. This notion of “core powers” has been wrongly transposed to what sorts of powers statutory courts may generally exercise. It is one thing to say that provinces cannot colourably get around s 96 by creating and appointing administrative bodies that have the “core powers” of a superior court. It is quite another to say that provinces, exercising their own constitutional powers to create valid criminal and civil courts under s 92(14) of the Constitution Act, 1867 cannot arm those courts with the remedial tools they need to fully exercise their criminal and civil jurisdictions and to enforce the dictates of the Constitution.
26 R v Big M Drug Mart, [1985] 1 SCR 295, 60 AR 161 [Big M Drug Mart].
27 Ibid at 315.
28 Ibid at 316.
29 Ibid.
30 Ref re Remuneration of Judges of the Prov Court of PEI; Ref re Independence and Impartiality of Judges of the Prov Court of PEI, [1997] 3 SCR 3 at 85–86, 156 Nfld & PEIR 1 [Judges’ Reference].
62 UNBLJ RD UN-B [VOL/TOME 68
But Lloyd was decided in the way it was. And for good measure it was soon confirmed by the Supreme Court in Windsor v Canadian Transit Company.31
In Windsor, like Lloyd, the Supreme Court queried the ability of statutory courts—this time, the Federal Courts—to make declarations of invalidity.32 The Supreme Court cited the very portions of Big M Drug Mart that I just mentioned even though they go against its position.33
But in a very important respect, Windsor extends Lloyd, further diminishing Canada’s statutory courts.
Windsor concerned the Ambassador Bridge, an international bridge connecting Windsor, Ontario with Detroit, Michigan. At issue in Windsor was whether parties could seek relief in the Federal Courts system concerning issues concerning the maintaining of the bridge, the construction of a new bridge span, and the management and disposition of certain properties in Windsor, Ontario bought by the bridge owner for these bridge-oriented purposes.
2017 CanLIIDocs 175 The Federal Courts have the power in certain circumstances to make rulings concerning interprovincial works and undertakings under subsection 23(c) of the Federal Courts Act. In Windsor, the Supreme Court found that in the particular circumstances of the case, subsection 23(c) did not give the Federal Courts jurisdiction over the Ambassador Bridge.
In reaching its decision, the Supreme Court emphasized that statutory courts are not superior courts with inherent jurisdiction. It reaffirmed that the powers of statutory courts are set out in statutory provisions.34
By itself, that is not controversial and is not much of a limit on what statutory courts can do. Statutes are to be interpreted in accordance with their text, context and purpose.35 And in the case of public institutions—especially courts that have essential
31 2016 SCC 54, [2016] 2 SCR 617 (5:4 majority) [Windsor]. I was the author of the reasons in the court below, which was reversed: Windsor FCA, supra note 15.
32 Ibid at paras 70–71.
33 It also cited Douglas/Kwantlen Faculty Association v Douglas College, [1990] 3 SCR 570, 77 DLR (4th) 94 to the effect that administrative tribunals do not have the power to make declarations of no force or effect under s 52 of the Constitution Act, 1982. That sort of restriction makes sense for administrative tribunals, often staffed with lay people and fully reviewable by full-fledged judges on judicial review courts. Further, statutory courts occupy a different place in the constitutional firmament and have a constitutional status that administrative decision-makers do not: see the constitutional bases for Canada’s statutory courts set out in ss 92(14) and 101 of the Constitution Act, 1867. Put another way, “[t]he Federal Court is a superior court, not an administrative tribunal”: Charkaoui v Canada (Citizenship and Immigration), 2007 SCC 9 at para 136, [2007] 1 SCR 350. See also the fundamental constitutional differences between administrative decision-makers and courts described in Ocean Port Hotel Ltd v British Columbia (General Manager, Liquor Control and Licensing Branch), 2001 SCC 52, [2001] 2 SCR 781.
34 Windsor, supra note 31 at para 33.
2017] A JUDICIARY CLEAVED 63 purposes to fulfil and functions to discharge in the separation of powers—the consideration of purpose must play a dominant role. Finally, section 12 of the federal Interpretation Act confirms all of this. It provides that “[e]very enactment is deemed remedial, and shall be given such fair, large and liberal construction and interpretation as best ensures the attainment of its objects.”36
In the case of the Federal Courts, the objects and the purposes of the Federal Courts Act are clear: the Act establishes courts at the federal level to regulate matters that provinces alone cannot regulate and to harmonize the interpretation and application of federal laws.
To illustrate, imagine that the Federal Courts system did not exist. Suppose that federal law could be interpreted only by provincial superior courts. Much chaos would follow. A tax deduction under the federal Income Tax Act might be allowed in the western part of the town of Lloydminster, Alberta, but across the street in Lloydminster, Saskatchewan, the deduction might be disallowed. A business act might be anti-competitive in one jurisdiction, but not in another. Federal workers for a single federal company might end up having rights in one province that the workers in another 2017 CanLIIDocs 175 province don’t have.37
As for regulating matters that provinces cannot alone regulate, suppose there was a big dispute between Ontario and Quebec and so Quebec closed all the bridges to Ontario. As interprovincial works, the bridges are federally regulated. Who would consider the closure? The courts of Quebec? The courts of Ontario? Both? What if they disagree with each other? A strong, neutral Federal Courts system has a role to play.
As for other sections in the Federal Courts Act that provide some context surrounding the powers of the Federal Courts, sections 3 and 4 provide that the Courts are “additional court[s] of law, equity and admiralty in and for Canada, for the better administration of the laws of Canada and …superior court[s] of record having civil and criminal jurisdiction.”
When trying to figure out the powers of the Federal Courts, all this purpose and context is rich grist for the interpretive mill. The striking thing about Windsor, though, is that the Supreme Court did not look to any of this.38
35 Bell ExpressVu Limited Partnership v Rex, 2002 SCC 42, [2002] 2 SCR 559; Rizzo & Rizzo Shoes Ltd (Re), [1998] 1 SCR 27, 154 DLR (4th) 193. 36 RSC 1985, c I-21 at s 12.
37 On the purposes of the Federal Courts system, see the Rt Hon John N Turner, “The Origin and Mission of the Federal Court of Canada” (Paper delivered at the 20th Anniversary Symposium of the Federal Court, 26 June 1991).
38 Windsor has already been the subject of criticism for its narrow textual approach to the interpretation of the powers of a statutory court and for not having regard to the purpose of the statute: see Paul Daly, “When is a Court Not a Court? Windsor (City) v. Canadian Transit Co., 2016 SCC 54”, online:
64 UNBLJ RD UN-B [VOL/TOME 68
Instead, on its face, Windsor seems to teach us that the powers of statutory courts are limited to the express powers you see in their Acts, read literally. What you see in black-and-white is exactly what the statutory court can do, nothing more.
It’s as if the only method for interpreting the powers of a statutory court is to look only at provisions that grant visible and tangible powers. And it’s as if the only tool to assist in that interpretation is a dictionary.39
In Windsor, the Supreme Court carved out just a small exception to this. In a footnote, it mentioned that the Federal Court as a statutory court also has the powers that are “necessarily implied in the [statutory] grant of power to function as a court of law,” such as the power to control the court’s processes.40
But even this derogates from our previous understanding of the sorts of powers a statutory court has.
Until Windsor was decided, statutory courts were thought to have express powers in their statute, but also implied and necessarily incidental powers discerned 2017 CanLIIDocs 175 through the normal process of statutory interpretation.41 And this applies to all statutory powers, not just, as Windsor says, the powers of the court over its own processes.
This orthodoxy was set down in the Supreme Court’s 1992 decision in Chrysler.42 There, following a textual, contextual and purposive approach to the
Interpretation: Windsor (City) v. Canadian Transit Co.”, online:
39 Indeed, at para 47, the majority in Windsor expressly declines to examine the purposes behind the Act, finding the “explicit language” clear. Even where the language of a provision is not clear—and the argument before the Supreme Court in Windsor confirms that the language of subsection 23(c) was not clear—a court is nevertheless obligated to look at the context and purpose of the provision. See Canada Trustco Mortgage Co v Canada, 2005 SCC 54 at para 10, [2005] 2 SCR 601: “The relative effects of ordinary meaning, context and purpose on the interpretive process may vary, but in all cases the court must seek to read the provisions of an Act as a harmonious whole” [emphasis added]. A court must consider the total context of the provision to be interpreted “no matter how plain the disposition may seem upon initial reading”: ATCO Gas and Pipelines Ltd v Alberta (Energy and Utilities Board), 2006 SCC 4 at para 48, [2006] 1 SCR 140; see also R v Monney, [1999] 1 SCR 652 at para 26, 171 DLR (4th) 1, and R v Lewis, [1996] 1 SCR 921 at para 68, 133 DLR (4th) 700.
40 Windsor, supra note 31 at para 33, n 1.
41 See supra note 35.
42 Chrysler Canada Ltd v Canada (Competition Tribunal), [1992] 2 SCR 394, 92 DLR (4th) 609 [Chrysler]. See also Canada Labour Relations Board v Québecair, [1993] 3 SCR 724, 108 DLR (4th) 1. For an analysis of both, see David Stratas, “A Unique Approach to Interpreting Tribunal Powers: Justice Gonthier and the cases of Chrysler and Québecair” in Michel Morin et al, eds, Responsibility, Fraternity and Sustainability in Law: In Memory of the Honourable Charles Doherty Gonthier (Markham: LexisNexis, 2012) 123. In Chrysler, the Supreme Court was examining the powers of the Competition Tribunal, a statutory body. Strangely, as things stand today with Chrysler and Windsor on the books, the
2017] A JUDICIARY CLEAVED 65 relevant statutory provisions, the Supreme Court held that a statutory body had the power to punish parties for contempt of its orders even though that power was not in the black-and-white of its statute.
This idea was further confirmed in Liberty Net.43 There, the Supreme Court said that the Federal Courts have express, implied and necessarily-incidental powers. For example, suppose the Federal Court of Appeal, a statutory court, has an express statutory power to review decisions of tribunals. But suppose one decision of a tribunal is in conflict with another. Does the Federal Court of Appeal have the power to resolve the conflict? It does, under what the Supreme Court in one case called “inherent jurisdiction,” but what is really just an implied power stemming from its general judicial review jurisdiction.44
Liberty Net did even more. It introduced the idea that the Federal Courts, as statutory bodies, had a plenary power to regulate proceedings before it. This power was said to be akin to the general powers of the superior courts to regulate their proceedings.45
2017 CanLIIDocs 175 This isn’t something in the express words of the Federal Courts Act, or implied or necessarily incidental to them. Rather, the plenary power emanates from the Federal Courts’ status as courts. The idea is that if you are a court, you have all the powers that a court ought to have.46
statutes of administrative bodies like the Competition Tribunal may be interpreted more liberally than the statutes establishing and governing statutory courts.
43 Supra note 12.
44 British Columbia Telephone Co v Shaw Cable Systems (BC) Ltd, [1995] 2 SCR 739 at 768, 125 DLR (4th) 443.
45 Indeed, one foremost constitutional scholar, looking at the nature of the powers, the purposes and the functions of the Federal Courts has concluded that they are superior courts in a very real sense and should be treated as such: William Ralph Lederman, “The Independence of the Judiciary” (1956) 34 Can Bar Rev 1139. Another scholar has written that these days all superior courts are creatures of statute and have a number of qualities such as the ability to determine their own jurisdiction, inherent jurisdiction, immunity from suit and the ability to control contempt. Therefore, “[t]o determine whether a court has the status of a superior court … [one must] look primarily to the statute which has brought the court into being.” See Enid Campbell, “Inferior and Superior Courts and Courts of Record” (1997) 6 J Judicial Administration 249.
46 These are the sorts of powers Jacob identified as inherent powers: see supra notes 9–11 and accompanying text. For examples in the Federal Courts system, see Canada (National Revenue) v Derakhshani, 2009 FCA 190; Canada (National Revenue) v RBC Life Insurance Company, 2013 FCA 50; Canada (National Revenue) v McNally, 2015 FCA 195; Coote v Lawyers’ Professional Indemnity Company, 2013 FCA 143; Jaffal v Davidson, 2016 FCA 226; Mazhero v Fox, 2014 FCA 226, Mazhero v Fox, 2014 FCA 238; Mazhero v Fox, 2014 FCA 219; Forner v Professional Institute of the Public Service of Canada, 2016 FCA 35; Philipos v Canada (AG), 2016 FCA 79; Amgen Canada Inc v Apotex Inc, 2016 FCA 121; Olumide v Canada, 2016 FCA 287; Valeant Canada LP v Canada (Health), 2014 FCA 50; Pfizer Canada Inc v Teva Canada Limited, 2016 FCA 218; Association des Compagnies de Téléphone du Québec Inc v Canada (AG), 2012 FCA 203; Lukács v Canada (Transportation Agency), 2016 FCA 103; Apotex Inc v Allergan, Inc, 2016 FCA 15.
66 UNBLJ RD UN-B [VOL/TOME 68
And this isn’t something that is novel. It has been in place for over a century.47 In Cocker v Tempest, a decision 175 years old, Baron Alderson stated that the inherent power “of each Court over its own process is unlimited; it is a power incident to all Courts, inferior as well as superior.”48
In short, Chrysler and Liberty Net and authorities like Cocker tell us that to determine the powers of a statutory court we should look at the statute purposively, sensitively and alert to nuances. There may be more than meets the eye—there may be implied, necessarily incidental powers and plenary powers too.
But all of this—all these principles and cases—went unmentioned in Windsor.
The statutory courts and superior courts spring from similar provisions in the Constitution Act, 1867, benignly worded, seemingly narrow.49 But today, never wider has been the difference in their status and jurisdiction and the interpretation of their powers.50
2017 CanLIIDocs 175 So, in closing, where do we stand today?
We have a judiciary cleaved: at one level fully empowered and at another level less so. And for no good reason.
47 See Ferrere, supra note 8; Dockray, supra note 8 at 125–126; Shalin Sugunasiri, “The Inferior Jurisdiction of Inferior Courts (1990) 12 Adv Q 215 at 218–219.
48 Cocker v Tempest, [1841] ER 242, (1841) 7 M&W 502 at 503–504. See also R v Norwich Crown Court, [1992] 1 WLR 54 (QB); Connelly v Director of Public Prosecutions, [1964] AC 1254 at 1301 (HL).
49 See the Constitution Act, 1867, ss 92(14) (the provincial power concerning “the Constitution, Maintenance, and Organization of Provincial Courts, both of Civil and of Criminal Jurisdiction”), 96 (the Governor General’s power “to appoint the Judges of the Superior, District, and County Courts in each Province”) and 101 (Parliament’s power “notwithstanding anything in this Act” to “provide for the Constitution, Maintenance, and Organization of a General Court of Appeal for Canada, and for the Establishment of any additional Courts for the better Administration of the Laws of Canada”).
50 For example, compare the differing judicial treatment of ss 96 and 101 of the Constitution Act, 1867. Putting aside part of one recent case, s 101 of the Constitution Act, 1867—the constitutional basis for the Federal Courts—has never enjoyed the expansive, purposive, “living tree” approach to interpretation that the most of rest of the Constitution has enjoyed: see e.g. R v Thomas Fuller Construction Co (1958) Ltd, [1980] 1 SCR 695 at 707, 106 DLR (3d) 193; Quebec North Shore Paper Co v Canadian Pacific Ltd, [1977] 2 SCR 1054 at 1065–1066, 71 DLR (3d) 111; Northern Telecom Canada Ltd v Communication Workers of Canada, [1983] 1 SCR 733, 147 DLR (3d) 1; ITO-International Terminal Operators Ltd v Miida Electronics Inc, [1986] 1 SCR 752, 28 DLR (4th) 641. The recent case is the Reference re Supreme Court Act, ss 5 and 6, 2014 SCC 21, [2014] 1 SCR 433. There, the Supreme Court, a statutory court under s 101, formed at the same time as the Exchequer Court (see the Supreme and Exchequer Court Act, SC 1875, c 11), the predecessor to the Federal Courts, constitutionally entrenched itself. Contrast this with s 96 of the Constitution Act, 1867 discussed at note 25, supra. Read literally, it is just a simple power of the federal government to appoint judges of the provincial superior courts. But the Supreme Court has stretched it to protect the “core” jurisdiction of provincial superior courts (see e.g. Re Residential Tenancies, supra note 25), to restrict the power of the federal government to vest powers in its own Federal Courts (MacMillan Bloedel, supra note 25) and even to guarantee a constitutional right of access to all courts (Trial Lawyers, supra note 24).
2017] A JUDICIARY CLEAVED 67
We want our statutory courts and the judges on them—the successors of great judges like Justices Rand, La Forest, Bastarache, Blanchard and Layden- Stevenson—protecting and vindicating our rights. We don’t want them shackled by artificial distinctions that serve no purpose.
The Supreme Court in Big M Drug Mart, Chrysler and Liberty Net got it exactly right. And the majority reasons of the Court in Liberty Net were written by Justice Bastarache—of course a great New Brunswicker!
But of this there can be no doubt: Lloyd and Windsor are the law today and they are binding.51
But sometimes—not always—counsel can persuade courts that authorities are distinguishable. Sometimes they can persuade courts that earlier authorities, not overruled, still govern.52 Sometimes the Supreme Court—comprised of talented jurists deeply dedicated to the betterment of Canadian law for all—reevaluates and reverses its own authorities.53
2017 CanLIIDocs 175 But for now, to eliminate uncertainty, to prevent any deleterious effects and to serve the public well, statutory courts might well wish to seek statutory amendments to shore up their powers.
In the end, to fulfil the great role of the judicial branch under our separation of powers in this complex age, statutory courts cannot be seen as lesser, cannot be treated as limited, and cannot be left emasculated. Fully empowered they must be.
51 On the binding nature of Supreme Court authority, see R v Henry, 2005 SCC 76, [2015] 3 SCR 609.
52 As for what intermediate appellate courts can legitimately and usefully do in the face of binding higher authority, see Richard M Re, “Narrowing Supreme Court Precedent from Below” (2016) 104 Georgetown LJ 921.
53 Recent examples where the Supreme Court has overruled its own authorities include R v Jordan, 2016 SCC 27, [2016] 1 SCR 631; Carter v Canada (AG), 2016 SCC 4, [2016] 1 SCR 13; Mounted Police Association of Ontario v Canada (AG), 2015 SCC 1, [2015] 1 SCR 3; Saskatchewan Federation of Labour v Saskatchewan, 2015 SCC 4, [2015] 1 SCR 245; Nova Scotia (Workers’ Compensation Board) v Martin; Nova Scotia (Workers’ Compensation Board) v Laseur, 2003 SCC 54, [2003] 2 SCR 504; Canada v Craig, 2012 SCC 43, [2012] 2 SCR 489; Halifax (Regional Municipality) v Nova Scotia (Human Rights Commission), 2012 SCC 10, [2012] 1 SCR 364. In Teva Canada Ltd v Pfizer Canada Inc, 2012 SCC 60, [2012] 3 SCR 625, the Supreme Court invalidated Pfizer’s patent. It later discovered that invalidation was not an available remedy under the regulatory regime in issue in the case and corrected its error: post- judgment decision on motion, June 4, 2013.
THE SIGNAL AND THE NOISE IN ADMINISTRATIVE LAW
Paul Daly*
The signal is the truth. The noise is what distracts us from the truth. This is a book about the signal and the noise.
Nate Silver, The Signal and the Noise: The Art and Science of Prediction (London: Penguin, 2013).
Introduction
There has been an unfortunate trend in recent Supreme Court of Canada administrative law cases.1
While academics, practitioners and lower-court judges try to establish 2017 CanLIIDocs 175 coherent frameworks to understand the general principles of judicial review, the Court has been resolving cases one by one without, with respect, any serious attempt to explain how they fit into its existing body of administrative law jurisprudence. The institutional context in which the Court operates, explained in Part I, no doubt influences the Court’s resolution of individual cases in this way. Whatever the background institutional context, confusion has been the unfortunate result, as I explain in Part II. How should lawyers read these cases: as attempts to resolve one-off issues of substantive law (workers’ compensation law, immigration law, discrimination law and so on) or as continual refinements to an already complex body of administrative law doctrine?
I argue that it is now necessary to distinguish between signal and noise, between those cases that do modify administrative law doctrine and those cases that simply deal with a particular substantive area of law. In Part III, I set out criteria which will help Canadian administrative lawyers to distinguish signal from noise.
This confusion seems to me, however, to be unnecessary. As I suggest in Part IV, the Court could take up judicial and academic proposals for a unified
* Senior Lecturer in Public Law, University of Cambridge; Derek Bowett Fellow in Law, Queens’ College, Cambridge; Research Associate, Centre de recherche en droit public. A previous version was delivered as the Keynote Address to the Law Society of Upper Canada’s 24th Annual Immigration Law Summit, November 23, 2016 and to the Administrative Law Section, Prince Edward Island branch, of the Canadian Bar Association on December 14, 2016. Thanks, with the usual disclaimer, to participants at both events and the anonymous reviewers for their penetrating comments and insights. Some of the material in this article appeared originally on my blog, Administrative Law Matters.
1 My focus here is on the period from 2008 to the present day. I have read all of the Supreme Court of Canada’s administrative law output since Dunsmuir v New Brunswick, 2008 SCC 9, [2008] 1 SCR 190 [Dunsmuir] and a significant proportion of the post-Dunsmuir case law produced by the lower courts. I appreciate that some of my comments may seem impressionistic. The concerned reader can rest assured that I opine from a solid knowledge base. 2017] THE SIGNAL AND THE NOISE 69 reasonableness standard that would allow it to resolve individual cases without creating uncertainty about whether it has also modified administrative law doctrine. Adopting this approach would obviate the need to distinguish signal from noise, because they would both fade into the comforting hum of reasonableness review.
I. Institutional Context
The problem of distinguishing signal from noise must be understood in context.
The Court is an apex court, sitting athwart Canada’s judicial hierarchy. With the exception of certain criminal cases where there is an automatic right of appeal, the Court deals only with questions of “public importance”, important legal questions and other questions “of such a nature or significance as to warrant [its] decision”.2
Moreover, its decisions are important not only for the parties that appear before it but also for the wider community: the reasons it gives lay down important 2017 CanLIIDocs 175 guidance for lower courts, litigants and laypeople.
The Court itself is aware of this. In R v Henry,3 Binnie J explained that the traditional common law distinction between ratio decidendi – those matters essential to the decision of a case – and obiter dicta – everything else – was a “supposed dichotomy” that is an “oversimplification of how the common law develops”.4 Rather:
The issue in each case…is what did the case decide? Beyond the ratio decidendi which…is generally rooted in the facts, the legal point decided by this Court may be as narrow as the jury instruction at issue in [Sellars v The Queen, [1980] 1 SCR 527] or as broad as the Oakes test. All obiter do not have, and are not intended to have, the same weight. The weight decreases as one moves from the dispositive ratio decidendi to a wider circle of analysis which is obviously intended for guidance and which should be accepted as authoritative. Beyond that, there will be commentary, examples or exposition that are intended to be helpful and may be found to be persuasive, but are certainly not “binding” in the sense the Sellars principle in its most exaggerated form would have it. The objective of the exercise is to promote certainty in the law, not to stifle its growth and creativity. The notion that each phrase in a judgment of this Court should be treated as if enacted in a statute is not supported by the cases and is inconsistent with the basic fundamental principle that the common law develops by experience.5
2 Supreme Court Act, RSC 1985, c s-26, s 40(1).
3 R v Henry 2005 SCC 76, [2005] 3 SCR 609 [Henry].
4 Ibid at para 52.
5 Ibid at para 57.
70 UNBLJ RD UN-B [VOL/TOME 68
The result, though, is that even if the decisions of the Court are not treated as legislative pronouncements, there is a temptation to closely parse obiter statements for indications of changes in the law. The question is no longer “is this part of the ratio?” but rather “is this authoritative?”
The undoubted need to give general guidance is also a reason for enhanced collegiality on an apex court. It has been said that under the stewardship of McLachlin CJ, the Court has placed a premium on collegiality and the production of majority and even unanimous reasons.6 This can increase certainty and clarity by committing all members of the Court to the same position. But it can also undermine certainty and clarity by achieving a narrow unified position at the cost of drowning out dissenting noises that may prove too loud to ignore in later cases.7
II. Signal or Noise?
A problem that can be traced to the institutional context in which the Supreme Court of Canada operates plagues the Canadian law of judicial review of administrative 2017 CanLIIDocs 175 action. Because of the Court’s role in answering important questions of law, it is often difficult to determine whether the reasons given for deciding a judicial review case are intended to have an effect on the general principles of administrative law. Administrative law consists of general principles that have to be applied to different areas of substantive law. Decisions in environmental law, discrimination law, workers’ compensation law, immigration law and so on are, where taken by administrative decision-makers, subject to judicial review for legality, rationality and fairness.8 These general principles exist at one remove from the substantive law that provides the context in which administrative decisions are taken.
As a result, the Court may grant leave to appeal (and may eventually decide a case) for one of two reasons: it may wish to answer a question or questions relating to the general principles of judicial review; it may wish to answer a question of substantive law; some combination of the two is also possible. Although reasons for granting or denying leave are not given, my experience is that the Court is much more interested in questions of substantive law than questions relating to the general principles of judicial review.
Indeed, as the judicial body of last resort, the Court has to give authoritative guidance on matters of substantive Canadian law to other actors (individuals, politicians, lawyers and lower-court judges), a task that might be impeded by consideration of the general principles of judicial review. According deference to
6 See e.g. Emmett Macfarlane, “Consensus and Unanimity at the Supreme Court of Canada” (2010) 52 SCLR (2d) 379.
7 For a generalized critique of the demise of seriatim opinions, see Dyson Heydon, “Threats to Judicial Independence: the Enemy Within” (2013) 129 Law Q Rev 205.
8 Canada (AG) v TeleZone Inc, 2010 SCC 62 at para 24, [2010] 3 SCR 585.
2017] THE SIGNAL AND THE NOISE 71 administrative decision-makers, for instance, means favouring legal pluralism, permitting those decision-makers to put their own spin on rules of substantive and procedural law. But as court of final resort, the Court has an institutional obligation to set down clear substantive and procedural rules for courts and decision-makers across the country.
Few of the judicial review cases the Court agrees to hear provide meaningful guidance to lower courts on how to apply the general principles of administrative law. Sometimes the Court ignores the role of the administrative decision-maker entirely – Febles v Canada (Citizenship and Immigration),9 which involved an important question of immigration law touching upon the interpretation of the United Nations Convention Relating to the Status of Refugees is an example; sometimes it dresses up its authoritative exposition of the law in the guise of reasonableness review – so-called “disguised correctness review”, in which it says it is applying a reasonableness standard but in fact performs its own analysis of the law and the facts to reach an independent conclusion that it labels ‘reasonable’ or ‘unreasonable’10; and sometimes in its drive for coherence it undermines legal 11 pluralism. The techniques are not mutually exclusive, of course, and they are often 2017 CanLIIDocs 175 deployed in combination.
These techniques might permit the Court to provide authoritative guidance on important questions of substantive law, but their use raises inevitable questions about their impact on the general principles of judicial review. When the Court ignores administrative law, engages in disguised correctness review, or otherwise plays fast and loose with administrative law doctrine to enable it to give guidance to the wider community on substantive law, it risks warping the administrative law framework and creating confusion.
Consider Québec (Commission des droits de la personne et des droits de la jeunesse) v Bombardier Inc (Bombardier Aerospace Training Center),12 a case that
9 Hernandez Febles v Canada (Citizenship and Immigration), 2014 SCC 68, [2014] 3 SCR 431.
10 David J Mullan, “2015 Developments in Administrative Law Relevant to Energy Law and Regulation” (2016) 4(1) ERQ online 2015 Developments in Administrative Law Relevant to Energy Law and Regulation. See e.g. Canada (Canadian Human Rights Commission) v Canada (AG) 2011 SCC 53, [2011] 3 SCR 471 though see also the benign interpretation of that decision offered in Canada (AG) v Canadian Human Rights Commission 2013 FCA 75 at para 15, [2013] FCJ No 249 (QL), a point to which I will return below in Part IV.
11 See the discussion below of Bombardier, infra note 12. See also Mouvement laïque québécois v Saguenay (City), 2015 SCC 16, [2015] 2 SCR 3 [Saguenay]; Paul Daly, “Why Would Jurisdiction Be Concurrent? Another Thought on Mouvement laïque québécois v Saguenay (City), 2015 SCC 16” (17 April 2015) Administrative Law Matters (blog), online:
12 Quebec (Commission des droits de la personne et des droits de la jeunesse) v Bombardier Inc (Bombardier Aerospace Training Center), 2015 SCC 39, [2015] 2 SCR 789 [Bombardier].
72 UNBLJ RD UN-B [VOL/TOME 68 was resolved decisively in favour of the exposition of national rules of substantive and procedural law.
Latif is a pilot who was denied training by Bombardier in 2004. The denial was based on a national security decision of the American authorities, a decision Bombardier applied because it did not want to imperil its standing with the Federal Aviation Authority. Latif is Pakistani. The Quebec Human Rights Tribunal found that Bombardier had discriminated against him. Although there was no direct evidence of discrimination by Bombardier, the Tribunal based its decision on an expert report and circumstantial evidence about racial profiling in the United States after 9/11.
At first blush, this looks like a straightforward administrative law case that required the Tribunal to weigh evidence and come to a conclusion. Moreover, it conducted the weighing exercise in a very particular context, one in which an individual like Latif is powerless in the face of an unreviewable decision.13 In this sort of context, one can understand why the Tribunal was not especially impressed by Bombardier’s automatic application of the American decision and why the 2017 CanLIIDocs 175 Tribunal thought Bombardier should have been more proactive.14 As a large institution, it was certainly in a better position than Latif to follow up with the U.S. authorities.
Why, then, did the Court grant leave? There are two large clues in the joint reasons of Wagner and Côté JJ for a unanimous Court.
First, this was the first opportunity for the Court to consider “a form of discrimination allegedly arising out of the decision of a foreign authority”.15
Second, the Court had “never clearly enunciated the degree of proof associated with the plaintiff’s burden” of making out a prima facie case of discrimination.16 Unsurprisingly, the bulk of Wagner and Côté JJ’s reasons are devoted to giving administrative decision-makers and lower courts guidance on these inter-related issues. The guidance is that the civil standard of the burden of proof always applies.17
What about legal pluralism? Wagner and Côté JJ accept that tribunals have the authority to adapt their rules of procedure and admissibility of evidence to their
13 Ibid at para16.
14 Ibid at para101
15 Ibid at para 2.
16 Ibid at para 55.
17 Ibid at para 65, though note that strictly speaking this conclusion only applies to Quebec, which has a specific legislative provision about the burden of proof.
2017] THE SIGNAL AND THE NOISE 73 particular regulatory context.18 But not the burden of proof, “in order to maintain the uniformity, integrity and predictability of the law”.19 It is difficult see a justification here for according the burden of proof a special status different to rules of procedure and evidence. Wagner and Côté JJ say only “that the application of a given legal test must be based on the same elements and the same degree of proof in every case”,20 but this is difficult to square with the Court’s openness to allowing administrative decision-makers to mould rules of substantive law to better achieve their regulatory purposes. As Fish J put it in Nor-Man Regional Health Authority Inc. v Manitoba Association of Health Care Professionals, an administrative decision-maker “may properly develop doctrines and fashion remedies appropriate in their field, drawing inspiration from general legal principles, the objectives and purposes of the statutory scheme” and other contextual considerations.21 Hopefully future reviewing courts will focus on the language highlighting the effective reversal of the burden of proof, because this is the sort of error that superior courts can more plausibly claim they have authority to correct on judicial review.
There then follows an intrusive analysis of the Tribunal’s appreciation of the facts, which looks suspiciously like de novo review even though it is adorned by 2017 CanLIIDocs 175 the language of reasonableness.22 At one point, Wagner and Côté JJ comment that the “practical” effect of the Tribunal’s decision was to reverse the burden of proof they had previously established,23 but in their analysis they carefully pick apart the Tribunal’s reasons, using different expressions: insufficient evidence,24 evidence not “tangibly related”,25 evidence “not sufficiently related”,26 or simply “no evidence”.27 But whether the evidence is adequate or not is a matter for the administrative decision-maker. As has been said many times, it is emphatically not “the function of the reviewing court to reweigh the evidence”.28
Further, the evidence was insufficient on only one of the three grounds Latif needed to prove to make out a prima facie case of discrimination — does this mean that a reviewing court is entitled to pick apart a human rights tribunal’s decision and
18 Ibid at paras 67–68.
19 Ibid at para 69.
20 Ibid at para 69.
21 Nor-Man Regional Health Authority Inc v Manitoba Association of Health Care Professionals, 2011 SCC 59 at para 45, [2011] 3 SCR 616.
22 Bombardier, supra note 12 at para 81.
23 Ibid at para 88.
24 Ibid at para 84.
25 Ibid at para 89.
26 Ibid at para 89.
27 Ibid at para 99.
28 Canada (Citizenship and Immigration) v Khosa, 2009 SCC 12 at para 60, [2009] 1 SCR 339 [Khosa].
74 UNBLJ RD UN-B [VOL/TOME 68 examine the sufficiency of the evidence on each ground independently? We have been told, however, that judicial review is not a “line-by-line treasure hunt for error”.29
Inasmuch as there is any meaningful guidance to lower courts here about the general principles of administrative law, the unfortunate effect would be to license intrusive judicial review of tribunals’ appreciation of the facts. In my view, Bombardier should be treated primarily as a case about discrimination law. If the Court’s goal was — as I suggested — to set out general principles relating to the burden of proof in discrimination cases, Wagner and Côté JJ’s reasons surely achieved it; it was a strong signal to lower courts and administrative tribunals about how to proceed in discrimination cases. But as far as administrative law is concerned, its comments on the standard of proof applied by the decision-maker and its close review of the evidence constitute noise that future courts should tune out. To put the point in the terms the Court employed in Henry, the “wider circle of analysis” in Bombardier is authoritative only as to the substantive law of employment discrimination, not as to the general principles of administrative law.
2017 CanLIIDocs 175
III. Sorting Signal from Noise
How do the rest of us know when the Court is telling us something about judicial review principles and when it is not? Should lower courts, lawyers and litigants try to integrate all of the Court’s judicial-review jurisprudence into their analytical frameworks or should they be selective?
In general, where the Court expressly sets out to give authoritative guidance on the general principles of administrative law its decisions should be closely parsed by administrative lawyers. But the rest, with respect, is “noise” as far as administrative law is concerned. These cases are characterized by purely pro forma references to correctness and reasonableness, an absence of detailed discussion of the general principles of standard of review and lengthy explanations of substantive law designed to guide lower courts.
This last characteristic is important. I am not saying that these cases should be ignored, but that they should be treated as authoritative only in respect of the particular area of substantive law they address. What is “noise” to someone interested in the general principles of judicial review may be a very strong “signal” to someone interested in, say, access-to-information law, or human-rights law. From the perspective of the general administrative lawyer, the distinction between “signal” and “noise” will be viewed differently and many cases of interest to others will have to be discarded.
29 Communications, Energy and Paperworkers Union of Canada, Local 30 v Irving Pulp & Paper, Ltd, 2013 SCC 34 at para 54, [2013] 2 SCR 458.
2017] THE SIGNAL AND THE NOISE 75
There are several well-known examples of cases in which the Court has explicitly sent signals about administrative law: CUPE v New Brunswick Liquor Corporation;30 Pushpanathan v Canada (Minister of Citizenship and Immigration);31 and, most recently, Dunsmuir v New Brunswick.32 These examples are straightforward, however. What is necessary is a set of criteria for identifying other cases that send important signals.
Extent of Treatment of an Issue
First, the Court may give greater or lesser treatment in its reasons to a particular issue.
An easy example is Newfoundland and Labrador Nurses' Union v Newfoundland and Labrador (Treasury Board).33 This was an unexceptional case about the calculation of vacation benefits by a labour arbitrator.
But Abella J began the judgment of the Court by referencing the 2017 CanLIIDocs 175 “transformative” Dunsmuir decision, in which the Court had said that the “purpose of reasons, when they are required, is to demonstrate “justification, transparency and intelligibility””.34 She then set out the issues for resolution: “whether the arbitrator’s reasons in this case satisfied these criteria and whether the reasons engaged procedural fairness”.35 Most of her reasons focused not on the facts of the case but on the general principles of administrative law.
She began by laying out “the context for understanding what the Court meant in Dunsmuir when it called for “justification, transparency and intelligibility””.36 She explained that Dunsmuir was not authority “for the proposition that the “adequacy” of reasons is a stand-alone basis for quashing a decision, or as advocating that a reviewing court undertake two discrete analyses — one for the reasons and a separate one for the result”.37 She also rejected the suggestion that “alleged deficiencies or flaws in the reasons fall under the category of a breach of the duty of procedural fairness and that they are subject to a
30 CUPE v New Brunswick Liquor Corporation, [1979] 2 SCR 227, 25 NBR (2d) 237.
31 Pushpanathan v Canada (Minister of Citizenship and Immigration), [1998] 1 SCR 982, 160 DLR (4th) 193 [Pushpanathan].
32 Dunsmuir, supra note 1.
33 Newfoundland and Labrador Nurses' Union v Newfoundland and Labrador (Treasury Board), 2011 SCC 62, [2011] 3 SCR 708 [Newfoundland Nurses].
34 Ibid at para 1, citing Dunsmuir, supra note 1 at para 47.
35 Ibid at para 1.
36 Ibid at para 13.
37 Ibid at para 14.
76 UNBLJ RD UN-B [VOL/TOME 68 correctness review”.38 And she laid down a general rule: “if the reasons allow the reviewing court to understand why the tribunal made its decision and permit it to determine whether the conclusion is within the range of acceptable outcomes, the Dunsmuir criteria are met”.39
Newfoundland Nurses thus sent an important signal to lower courts (something that was especially important given that some provincial appellate courts had seen the references to “justification, intelligibility and transparency” as inviting close scrutiny of the reasons given for administrative decisions).40
I should not be understood as saying that Newfoundland Nurses must therefore be read like a statute. To do so would be contrary to the Court’s express admonition in Henry. In addition, Newfoundland Nurses has to be read with other important decisions of the Court. In particular, in Alberta (Information and Privacy Commissioner) v Alberta Teachers' Association,41 decided the previous day, Rothstein J in his majority reasons had specified that the direction that courts are to give respectful attention to the reasons “which could be offered in support of a 42 decision” is not a “carte blanche to reformulate a tribunal’s decision in a way that 2017 CanLIIDocs 175 casts aside an unreasonable chain of analysis in favour of the court’s own rationale for the result”.43 It will be necessary in some situations to engage in a classical common law analysis which seeks to reconcile these two decisions.44 But there can surely be little doubt that Newfoundland Nurses sent a signal.
Conversely, cursory treatment of an issue may indicate that the Court’s views on a particular point should be considered to be noise. In Agraira v Canada (Public Safety and Emergency Preparedness)45 the underlying issue for decision – the meaning to be given to “national interest” in the context of one of the Minister’s discretionary powers – had been certified as a general question of law by the Federal Court for resolution by the Federal Court of Appeal. One might reasonably think that
38 Ibid at para 21.
39 Ibid at para 16.
40 See e.g. Clifford v Ontario Municipal Employees Retirement System, 2009 ONCA 670, (2009) 98 OR (3d) 210.
41 Alberta (Information and Privacy Commissioner) v Alberta Teachers' Association, 2011 SCC 61, [2011] 3 SCR 654 [Alberta Teachers].
42 Dunsmuir, supra note 1 at para 48, citing David Dyzenhaus, “The Politics of Deference: Judicial Review and Democracy”, in Michael Taggart, ed, The Province of Administrative Law (Oxford: Hart Publishing, 1997) 279 at 286.
43 Alberta Teachers, supra note 41 at para 54, citing Petro-Canada v Workers’ Compensation Board (BC), 2009 BCCA 396, 276 BCAC 135, at paras 53 and 56.
44 See Lemus v Canada (Citizenship and Immigration), 2014 FCA 114, 372 DLR (4th) 567.
45 Agraira v Canada (Public Safety and Emergency Preparedness), 2013 SCC 36, [2013] 2 SCR 559 [Agraira].
2017] THE SIGNAL AND THE NOISE 77 this would have been a strong indication that the correctness standard should apply.46 But LeBel J simply said, “the standard of review applicable in the case at bar has been satisfactorily determined in past decisions to be reasonableness”.47
In a subsequent case, the Federal Court of Appeal refused to follow Agraira on this point, because it appeared “to depart inexplicably from earlier Supreme Court of Canada jurisprudence”, essentially treating it as noise.48 Instead, the Federal Court of Appeal would continue “its practice of providing the definitive answer to a certified question on a point of statutory interpretation”.49 As a result, the issue had to be resolved expressly by the Court in favour of the Agraira approach. In Kanthasamy v Canada (Citizenship and Immigration), the Court gave explicit reasons that constituted a clear signal as to the Court’s view of the general principles of judicial review, viz. that the presumption of deferential review of a decision- maker’s interpretations of its home statute is not rebutted by the presence of an appeal clause.50
Sometimes, a decision of the Court can contain some signal and some noise. 51 Consider, in this respect, McLean v British Columbia (Securities Commission). 2017 CanLIIDocs 175 On the one hand, Moldaver J gave a lengthy exposition of the meaning of the other operative part of paragraph 47 of Dunsmuir, namely the injunction that an administrative decision must fall “within a range of possible, acceptable outcomes which are defensible in respect of the facts and law”.
Moldaver J explained the implications of this injunction for the review of administrative interpretations of statutory provisions: “because legislatures do not always speak clearly and because the tools of statutory interpretation do not always guarantee a single clear answer, legislative provisions will on occasion be susceptible to multiple reasonable interpretations”,52 in which case “the resolution of unclear language in an administrative decision maker’s home statute is usually best left to the decision maker…because the choice between multiple reasonable interpretations will often involve policy considerations that we presume the legislature desired the administrative decision maker — not the courts — to make”.53 In some cases, however, “[w]here the ordinary tools of statutory interpretation lead
46 See Pushpanathan, supra note 31, though see also Baker v Canada (Minister of Citizenship and Immigration), [1999] 2 SCR 817, 174 DLR (4th) 193.
47 Agraira, supra note 45 at para 49.
48 Kanthasamy v Canada (Citizenship and Immigration), 2014 FCA 113 at para 30, [2014] 1 FCR 335 [Kanthasamy].
49 Ibid at para 35.
50 Agraira, supra note 45 at paras 42–44.
51 McLean v British Columbia (Securities Commission), 2013 SCC 67, [2013] 3 SCR 895 [McLean].
52 Ibid at para 32.
53 Ibid at para 33 [emphasis in original].
78 UNBLJ RD UN-B [VOL/TOME 68 to a single reasonable interpretation”, the range of reasonable outcomes “will necessarily be limited to a single reasonable interpretation — and the administrative decision maker must adopt it”.54 This was a clear signal about how to address questions of statutory interpretation under paragraph 47 of Dunsmuir.
On the other hand, Moldaver J dealt summarily with the question of whether deference could be accorded to the Commission’s choice between the competing possible reasonable interpretations of the statutory provision that was at issue. The British Columbia Court of Appeal considered it “impossible” to review the interpretation,55 but Moldaver J was satisfied that “a basis for the Commission’s interpretation is apparent from the arguments advanced by the respondent”.56 Does this mean that the reasons to which a reviewing court must pay “respectful attention”57 include those advanced in argument by counsel for an administrative decision-maker? This is a troubling proposition,58 the potential wider impact of which Moldaver J did not seem to consider. It would be better, then, for lower courts to consider this aspect of McLean to constitute noise.
2017 CanLIIDocs 175 Concurring and Dissenting Reasons
Second, the presence of concurring and dissenting reasons as to a majority judgment’s treatment of the general principles of administrative law will suggest that a decision should be treated as signal rather than noise. Concurring and dissenting reasons tend to concentrate attention on particular points of dispute that might otherwise be obfuscated by a bland set of majority reasons.59
For instance, the presence of a lengthy and detailed dissent by Fish J in Canada (Citizenship and Immigration) v Khosa, in which he would have quashed the decision of the Immigration Appeal Division for giving too much weight to Mr. Khosa’s lack of remorse for engaging in street racing gives additional heft to the holding of the majority that reviewing courts should not reweigh the evidence considered by an administrative decision-maker.60 That the point was raised by the dissent and decided without equivocation by the majority made it a clear signal about the general principles of administrative law.
54 Ibid at para 38.
55 British Columbia (Securities Commission) v McLean, 2011 BCCA 455 at para 30, 343 DLR (4th) 432.
56 McLean, supra note 51 at para 72.
57 Dunsmuir, supra note 1 at para 48, citing Dyzenhaus, supra note 42 at 286.
58 See further Paul Daly, “The Scope and Meaning of Reasonableness Review” (2015) 52 Alta L Rev 799 at 815–817.
59 See generally James Lee, “A Defence of Concurring Speeches” (2009) Pub L 305.
60 Khosa, supra note 28 at para 59.
2017] THE SIGNAL AND THE NOISE 79
This was reinforced by Alberta Teachers’, in which the majority expressly rejected Binnie J’s suggestion in his concurring reasons that there ought to be variable degrees of deference within the reasonableness standard set out in paragraph 47 of Dunsmuir. Similarly, Cromwell J’s reluctance in his concurring reasons in Alberta Teachers’ to countenance the abolition of “true jurisdictional questions” as a category of question attracting correctness review61 underscored that the majority’s insistence that this category has to be extremely narrowly construed was a clear signal to lower courts. Finally, the refusal of the majority in Edmonton East (Capilano) Shopping Centres v Edmonton (City) to revise the administrative law principles concerning the application of the presumption of reasonableness review even to a statutory framework containing an appeal clause was, in view of the strident dissent on this point, a clear signal of the scope of deference.62
Consideration of Lower Court Decisions
Third, detailed consideration of lower court decisions will indicate that a decision contains important signals about administrative law. 2017 CanLIIDocs 175
Directly at issue in Mouvement laïque québécois v Saguenay (City)63 was a decision of Quebec’s Human Rights Tribunal about whether the recitation of a prayer before municipal meetings (allied to the presence of religious symbols) amounted to discriminatory treatment of an atheist who attended the meetings. Indirectly at issue was whether a statutory appeal clause could pre-empt the general principles of administrative law; does the standard of review analysis apply in all cases, even if a statute creates a very broad right of appeal?
Decisions of the Tribunal are appealable, with leave, directly to the Quebec Court of Appeal.64 The relevant statute also provides that the general rules governing appeals are to apply in this context. The Quebec Court of Appeal had split previously on the proper interpretation of its role on appeal from the Tribunal: some judges applied judicial review criteria (following the well-established rule that appeal clauses do not eliminate deference to specialized tribunals) but some applied appellate criteria based on the apparently plain language of the statute and the evident goal of giving the Quebec Court of Appeal a gatekeeping power by limiting appeals to those raising questions of general law.65
61 See especially Alberta Teachers, supra note 41 at paras 102–103 and see also Lauren Wihak, “Whither the Correctness Standard of Review: Dunsmuir Six Years Later” (2014) 27 Can J Admin L & Prac 174.
62 Edmonton East (Capilano) Shopping Centres v Edmonton (City), 2016 SCC 47, [2016] 2 SCR 293.
63 Saguenay, supra note 11.
64 Charter of Human Rights and Freedoms, CQLR c C-12, ss 132–133.
65 Saguenay, supra note 11 at paras 31–37.
80 UNBLJ RD UN-B [VOL/TOME 68
Having set out the conflicting approaches at some length, Gascon J came to a firm conclusion based on a comprehensive review of the authorities: “Where a court reviews a decision of a specialized administrative tribunal, the standard of review must be determined on the basis of administrative law principles…regardless of whether the review is conducted in the context of an application for judicial review or of a statutory appeal”.66
He acknowledged that “the scope of a right to appeal and the absence of exclusive jurisdiction may sometimes affect the deference to be shown to decisions of a specialized administrative tribunal” but held nonetheless that these features of a regulatory scheme would “not justify replacing the standards of review applicable to judicial review with the appellate standard”.67
Given the lengthy treatment of the conflicting approaches in the court below and the consideration of relevant authority, Saguenay sends a very clear signal about the relationship between appeal clauses and the standard of review framework, viz. that language creating a statutory appeal never pre-empts administrative law 68 principles (a point recently underscored again by Edmonton East). 2017 CanLIIDocs 175
Consistency of Treatment
Fourth, a decision which is out of line with other decisions on the general principles of administrative law is more likely to be noise than signal. Bombardier is an example, especially on the question whether reviewing courts can reweigh the evidence considered by an administrative decision-maker: one would scour the post- Dunsmuir Supreme Court Reports in vain for any indication that reweighing is a permissible activity for reviewing courts.
Another example might be the uncertainty created by Katz Group Canada Inc v Ontario (Health and Long‑Term Care)69 about the scope of the Dunsmuir framework. At issue in Katz was the validity of a set of regulations imposed by a provincial cabinet on the sale of generic medication. Abella J made no attempt to situate judicial review of regulations in the Dunsmuir framework, preferring instead to rely on 1980s authority on judicial review of regulations.70 This was an odd
66 Ibid at para 38.
67 Ibid at para 43.
68 See also Kanthasamy, supra note 48. Interestingly, the only other post-Dunsmuir case in which the standard of review analysis has been applied by the Court to a statute containing a leave provision was Bell Canada v Bell Aliant Regional Communications, 2009 SCC 40, [2009] 2 SCR 764. Readers will not be surprised to learn that the Court paid no attention to the existence of an appeal clause containing a leave provision.
69 Katz Group Canada Inc v Ontario (Health and Long‑Term Care), 2013 SCC 64, [2013] 3 SCR 810.
70 Thorne’s Hardware Ltd v The Queen, [1983] 1 SCR 106, 143 DLR (3d) 577.
2017] THE SIGNAL AND THE NOISE 81 outcome, because the Court had invariably in the post-Dunsmuir era applied the Dunsmuir framework in administrative law cases.
Subsequently, in Canadian National Railway Co v Canada (AG), Rothstein J denied that Katz cast any doubt on the general applicability of the Dunsmuir framework, characterizing it is a case limited to a challenge to the vires of regulations issued by a body (and, one might add, an elected body) acting in a “legislative capacity”.71 Similarly, in Green v Law Society of Manitoba,72 the standard of review framework was applied to the question whether rules imposing a mandatory continuing professional development requirement were within the Law Society’s statutory mandate;73 Katz was treated as an application of the framework to regulations.74 Given that Katz is inconsistent with the rest of the Court’s post- Dunsmuir jurisprudence, it must be considered to have been a noise – not a signal – about the general principles of administrative law. The better view is that Dunsmuir provides the framework for judicial review of all administrative decision-makers.75
Conclusion 2017 CanLIIDocs 175
One way of addressing the problem of distinguishing the signal from the noise in the Court’s administrative law cases is to develop criteria along the lines of those I have laid out above. My list of criteria is not exhaustive; the distinction between signal and noise will not always be an easy one to make and will typically require lawyerly judgement, just as sorting ratio from obiter has never been a purely mechanical exercise.
To anticipate potential objections, I do not think that my distinction between “signal” and “noise” is an incitement to illegitimate judicial disobedience to binding commands issued by the Court. To reiterate the Court’s own statements in Henry, to think of a “a strict and tidy demarcation” between ratio and obiter is an “oversimplification”.76 For the administrative lawyer, the “noise” cases are limited to their particular facts, but the “signal” cases involve commentary that is part of “a wider circle of analysis which is obviously intended for guidance and which should be accepted as authoritative”.77
71 Canadian National Railway Co v Canada (AG), 2014 SCC 40 at para 51, [2014] 2 SCR 135.
72 Green v Law Society of Manitoba, 2017 SCC 20.
73 Ibid at para 19.
74 Ibid at paras 20 and 67.
75 Though it is worth noting that the British Columbia Court of Appeal has applied Katz rather than Dunsmuir to a regulation adopted by an unelected regulatory body: Sobeys West Inc v College of Pharmacists of British Columbia, 2016 BCCA 41, 97 Admin L Rev (5th) 35.
76 Henry, supra note 3 at para 52.
77 Ibid at para 57.
82 UNBLJ RD UN-B [VOL/TOME 68
If the Court’s decisions continue to sow confusion by resolving individual cases in ways that are inconsistent with the general principles of judicial review, administrative lawyers will have to develop analytical tools that permit them to distinguish the important cases from the unimportant cases, the signal from the noise.
IV. Reasonableness and the Court’s Institutional Role
However, the Court’s decisions need not sow confusion. There is an alternative approach, which would permit the Court to set out general guidance about substantive areas of law in a way consistent with its institutional role at the apex of the Canadian legal system without causing confusion about the operation of the principles of administrative law.
The alternative approach is for the Court to abolish the standard of correctness and subject administrative decisions to reasonableness review in all cases. There is academic78 and extra-judicial79 support for this approach, which has recently received the cautious imprimatur of Abella J. 2017 CanLIIDocs 175
In her partially concurring reasons in Tervita Corp v Canada (Commissioner of Competition), she confessed to finding it “increasingly difficult to discern the demarcations between a reasonableness and correctness analysis” and even floated the possibility that the demarcations could be “completely erased” at some point in the future.80
What was implicit in Tervita recently became explicit in Wilson v Atomic Energy of Canada Ltd.81 Abella J’s reasons were notable for her forthright suggestion that the Court ought to consider collapsing correctness and reasonableness into a single reasonableness standard, the “most obvious and frequently proposed reform of the current system”.82 She set up the question as follows: “whether we need two different names for our approaches to judicial review, or whether both approaches can live comfortably under a more broadly conceived understanding of reasonableness”.83 She also highlighted the key advantage of a general reasonableness standard, that it is flexible enough to allow “a wider range for those kinds of issues and decision-makers traditionally given a
78 See e.g. Paul Daly, “Struggling Towards Coherence in Canadian Administrative Law? Recent Cases on Standard of Review and Reasonableness” (2016) 62 McGill LJ (forthcoming) [Daly, “Struggling Towards Coherence”].
79 See The Honourable Justice David Stratas, “The Canadian Law of Judicial Review: A Plea for Doctrinal Coherence and Consistency” (2016) 42 Queen’s LJ 27.
80 Tervita Corp v Canada (Commissioner of Competition), 2015 SCC 3 at para 171, [2015] 1 SCR 161.
81 Wilson v Atomic Energy of Canada Ltd, 2016 SCC 29, [2016] 1 SCR 770 [Wilson].
82 Ibid at para 28.
83 Ibid at para 24.
2017] THE SIGNAL AND THE NOISE 83 measure of deference, and a narrow one of only one ‘defensible’ outcome for those which formerly attracted a correctness review”.84
It is this flexibility that makes the reasonableness standard of such utility in responding to the signal and noise problem created by the Court’s recent judgments in the area of administrative law.
Dunsmuir’s reasonableness standard, with its twin references to “justification, intelligibility and transparency” and a “range of possible, acceptable outcomes”,85 has been refined over the years. It now seems clear that most of the analytical burden has been assumed by the concept of a “range of reasonable outcomes”,86 with the “justification, intelligibility and transparency” requirement satisfied by a decision that is “clearly understand[able]” to a reviewing court.87
Reasonableness “takes its colour from its context”.88 The “range will necessarily vary”.89 It “must be assessed in the context of the particular type of decision making involved and all relevant factors”.90
2017 CanLIIDocs 175 Of particular interest in the present context is the role of legal principles in narrowing the range of reasonable outcomes.
In Catalyst Paper, McLachlin CJ noted that range of reasonable outcomes may be “circumscribed” by reference to “the rationale of the statutory regime”.91
Similarly, in Canada (Minister of Transport, Infrastructure and Communities) v Jagjit Singh Farwaha Stratas JA explained that, in some cases “Parliament may have constrained the decision-maker’s discretion by specifying a recipe of factors to be considered – all things being equal, this narrows the range of options the decision-maker legitimately has”.92 An excellent illustration of this principle is the Court’s acknowledgement in McLean that, sometimes, the range of reasonable outcomes will be so narrow as to admit of only one outcome.
84 Ibid at para 33.
85 Ibid at para 47.
86 Khosa, supra note 28 at para 67.
87 Agraira, supra note 45 at para 89.
88 Khosa, supra note 28 at para 59.
89 Wilson, supra note 81 at para 22.
90 Catalyst Paper Corp v North Cowichan (District) 2012 SCC 2 at para 18, [2012] 1 SCR 5 [Catalyst Paper].
91 Ibid at para 25.
92 Canada (Minister of Transport, Infrastructure and Communities) v Jagjit Singh Farwaha, 2014 FCA 56 at para 91, [2014] 2 FCR 1006.
84 UNBLJ RD UN-B [VOL/TOME 68
In addition, prior judicial decisions on matters subsequently considered by an administrative tribunal will also tend to narrow the range of reasonable outcomes.93
And in Canada (AG) v Igloo Vikski Inc, albeit in dissent, Côté J took the view that the range of reasonable interpretations of a tariff schedule was constrained by the need to provide an answer consistent with Canada’s international obligations in respect of tariff harmonization.94
It has thus been said that “[l]egal matters, as opposed to factual or policy matters, admit of fewer possible, acceptable outcomes”.95 In most cases, this is likely to be true. The insight that the more legal in nature a question is the narrower the range of reasonable outcomes will be is significant because it allows us to appreciate how the Court might send signals about individual cases for the benefit of the environmental, immigration and workers’ compensation lawyers without creating too much noise for the administrative lawyers.
2017 CanLIIDocs 175 Very simply put, by demonstrating that the range of reasonable outcomes is constrained by statutory language, pre-existing jurisprudence and so on, the Court can provide a significant degree of structure to areas of substantive law. Without necessarily substituting judgment as it would by applying correctness (or “disguised correctness”) review, it can indicate that administrative decision-makers have, relatively speaking, a narrower margin of interpretation in some areas than in others. Although I think that the move to a unified reasonableness standard is the most rational next step in Canadian administrative law, I have significant reservations about a one-size-fits-all reasonableness standard. For one thing, applying the concept of a range of reasonable outcomes risks reintroducing distinctions between questions of law and questions of policy, fact and discretion.96 For another thing, permitting reviewing courts to define the range of reasonable outcomes in respect of a particular factual and legal matrix will allow judges to confine administrative decision-makers within strict limits.97
It would be better in my view for any “range” to be established by reference to the statute as a whole rather than to a particular statutory provision; departure from a line of previous cases, judicial authority or the natural meaning of a statute
93 Canada (AG) v Canadian Human Rights Commission, 2013 FCA 75 at paras 16 and 18, [2013] FCJ No 249 (QL).
94 Canada (AG) v Igloo Vikski Inc, 2016 SCC 38 at para 58, [2016] 2 SCR 80.
95 Canada (AG) v Abraham, 2012 FCA 266 at para 45, [2012] FCJ No 1324 (QL).
96 See Paul Daly, “Unreasonable Interpretations of Law” (2014) 66 SCLR (2d) 233.
97 See Paul Daly, “The Struggle for Deference in Canada” in Hanna Wilberg & Mark Elliott, eds, The Scope and Intensity of Substantive Review: Traversing Taggart’s Rainbow (Oxford: Hart Publishing, 2015).
2017] THE SIGNAL AND THE NOISE 85 would constitute indicia of unreasonableness that may justify judicial intervention.98 Nonetheless, adopting the range of reasonable outcomes concept is the most rational next step in the development of Canadian administrative law, though judges ought to apply it with due regard to the decisional autonomy accorded by the legislature to the administrative decision-maker under review.99
As Iacobucci J explained in Law Society of New Brunswick v Ryan, the analytical structure of reasonableness review provides some safeguards against judicial intrusion on administrative decision-makers’ autonomy: “Applying the standard of reasonableness gives effect to the legislative intention that a specialized body will have the primary responsibility of deciding the issue according to its own process and for its own reasons”.100 At no point should the reviewing court “ask itself what the correct decision would have been”,101 because even “if there could be, notionally, a single best answer, it is not the court’s role to seek this out when deciding if the decision was unreasonable”.102
It is thus inappropriate for a reviewing court to set up a benchmark based on its independent view of the legal and factual matrix, because “any departure from the 2017 CanLIIDocs 175 reviewing court’s hypothetical decision is bound to appear unreasonable”.103 Rather, a reviewing court should start from the decision and work outwards, identifying “badges of unreasonableness” that cannot be cogently explained by the decision- maker and, as a result, bring the decision outside the range of reasonable outcomes.104
Applying this methodology to Bombardier, Wagner and Côté JJ could have begun with the Tribunal’s decision and gone on to lay out the conventional burden of proof in discrimination cases, as well as the undisputed elements of discriminatory treatment. This would have permitted them to give general guidance to the legal community on the appropriate approach administrative decision-makers should follow in discrimination cases. A departure from the undisputed framework for discrimination cases would have been a badge of unreasonableness.
However, Wagner and Côté JJ could then have emphasized that the Tribunal, in the unique factual circumstances before it, had provided cogent reasons for its modification of the burden of proof (or reasons that were insufficiently cogent, as the case may be). In this way, Wagner and Côté JJ could have given general
98 See Paul Daly, “Unreasonable Interpretations of Law” (2014) 66 SCLR (2d) 233.
99 See Daly, “Struggling Towards Coherence”, supra note 78.
100 Law Society of New Brunswick v Ryan, 2003 SCC 20 at para 50, [2003] 1 SCR 247.
101 Ibid.
102 Ibid at para 51.
103 Ottawa Police Services v Diafwila, 2016 ONCA 627 at para 66, 270 ACWS (3d) 205.
104 Delios v Canada (AG), 2015 FCA 117 at para 27, [2015] FCJ No 549 (QL).
86 UNBLJ RD UN-B [VOL/TOME 68 guidance on discrimination law without causing confusion about the operation of the general principles of administrative law.
Conclusion
The institutional context in which the Court operates puts pressure on it to provide general guidance on issues of substantive law. Where such issues are first resolved by administrative decision-makers and addressed by the Court only on judicial review, however, the general principles of administrative law may be perceived as inhibiting the Court’s ability to provide general guidance. Unfortunately, the Court has in recent decisions prioritized the giving of general guidance over the sound operation of the principles of administrative law.
Accordingly, administrative lawyers need to be able to distinguish between the signal and the noise, between those decisions of the Court that are designed to structure the administrative law framework (“signal”) and those that are designed to resolve pressing issues of substantive law (“noise” – at least as far as administrative 2017 CanLIIDocs 175 lawyers are concerned).
Alternatively, however, the Court could take up academic, extra-judicial and now judicial suggestions to adopt a unified reasonableness standard of review. By setting the “range” of reasonable outcomes and rigorously following the well- established analytical structure of reasonableness review, the Court could send signals about substantive areas of law without creating unnecessary noise in the operation of the principles of judicial review. Noise and signal would merge in the comforting hum of reasonableness review.
THE TIME HAS COME: STANDARD OF REVIEW IN CANADIAN ADMINISTRATIVE LAW
Jonathan M. Coady1
“The time has come,” the Walrus said, “To talk of many things: Of shoes – and ships – and sealing-wax – Of cabbages – and kings – And why the sea is boiling hot – And whether pigs have wings.”2
The Walrus was right. The time has come in Canadian administrative law to revisit, once again, the issue that has bedevilled – and sometimes bewildered – lawyers, judges, and academics alike: standard of review. Pleas for coherence have been 2017 CanLIIDocs 175 issued.3 Calls for submissions have been made.4 And with the complexion of our highest court now almost completely different since the last revision,5 it appears that the next chapter in this story may be a deceptively simple one. A single standard of review for reasonableness would not only bring consistency to the judicial review of administrative decisions, but also strike a sound doctrinal balance between legislative supremacy and the rule of law. But, before doing so, we must first – as the Walrus said – talk of many things.
1 BSc (UPEI), LLB (Dal), LLM (Cantab). Partner, Stewart McKelvey, Charlottetown, Prince Edward Island. Thank you to the editors of the University of New Brunswick Law Journal and two anonymous peer reviewers for their insightful comments on earlier versions of this article. All errors are my own.
2 Lewis Carroll, Through the Looking-Glass and What Alice Found There (London: MacMillan and Co, 1872) at 75.
3 Honourable Justice David Stratas, “The Canadian Law of Judicial Review: A Plea for Doctrinal Coherence and Consistency” (2016) 42:1 Queen’s LJ 27. See also David Mullan, “Unresolved Issues on Standard of Review in Canadian Judicial Review of Administrative Action – The Top Fifteen!” (2013) 42 Adv Q 1; and Paul Daly, “Struggling Towards Coherence in Canadian Administrative Law: Reasonableness, the Rule of Law and Democracy” McGill LJ [forthcoming in 2017], online:
4 Wilson v Atomic Energy of Canada Ltd, 2016 SCC 29 at para 19, [2016] 1 SCR 770 [Wilson].
5 Dunsmuir v New Brunswick, 2008 SCC 9, [2008] 1 SCR 190 [Dunsmuir]. As of this article, only two judges remain from this groundbreaking panel: Chief Justice McLachlin and Justice Abella.
88 UNBLJ RD UN-B [VOL/TOME 68
Introduction
This article will not be a definitive account of the way forward. Its author is neither “the dean”6 of Canadian administrative law nor “a rising member”7 of the academy. In fact, he is not even in that company. Rather, this article will represent an attempt to contribute the perspective of a lawyer – just a plain old lawyer from a small town who is trying to help clients navigate this labyrinth of fundamental principles and basic practicalities. It will be one more answer to the call; nothing more and nothing less.
Part I will review the organizing principles distilled by the Supreme Court of Canada in Dunsmuir. There is soundness in these principles, and they remain useful in the judicial review process. Part II will examine some of the cases decided after Dunsmuir that have plagued and perplexed practitioners in this field. Common threads will be drawn from this sample of work by our highest court. Part III will consider how our understanding of legislative supremacy and the rule of law – the seemingly omnipresent source of tension in this area of law – has matured over time.
Each now recognizes a legitimate role for both administrative decision-makers and 2017 CanLIIDocs 175 courts. And finally, Part IV will explore how a contextual standard of review for reasonableness could operate in a principled yet practical way.
Part I: The Basic Soundness of Dunsmuir
Administrative law has been the great Canadian re-write. As Daly has unfortunately noted, “major recalibrations” have occurred every ten years or so.8 Cases like CUPE, Bibeault, Southam, and Dunsmuir will all echo in the ears of lawyers, judges, and academics working in this area. And with no real restatement of the law since 2008, it appears that we are due. Recent cases suggest that even the Supreme Court of Canada thinks there is still work to be done.9 An epilogue to Dunsmuir seems to be inevitable.
6 David Mullan. A title deservedly bestowed by Justice Stratas. See Stratas, supra note 3 at 28.
7 Paul Daly. Again, a well-earned compliment from Justice Stratas. See Stratas, supra note 3 at 28. However, one may compellingly argue that Professor Daly is not just a rising member of the academy, but a rising star. See e.g. Wilson, supra note 4 at para 27, and Edmonton (City) v Edmonton East (Capilano) Shopping Centres Ltd, 2016 SCC 47 at paras 72 and 89, [2016] 2 SCR 293 [Edmonton East].
8 Paul Daly, “The Scope and Meaning of Reasonableness Review” (2015) 52:4 Alta L Rev 799 at 827 [Daly, “Meaning of Reasonableness”]. In support of this thesis, Professor Daly points to the following cases: Canadian Union of Public Employees, Local 963 v New Brunswick Liquor Corporation, [1979] 2 SCR 227, 97 DLR (3d) 417 [CUPE]; UES, Local 298 v Bibeault, [1988] 2 SCR 1048, 95 NR 161 [Bibeault]; Canada (Director of Investigation and Research) v Southam Inc, [1997] 1 SCR 748, 144 DLR (4th) 1 [Southam]; and Dunsmuir, supra note 5.
9 See Tervita Corp v Canada (Commissioner of Competition), 2015 SCC 3 at para 170, [2015] 1 SCR 161 [Tervita]. See also Wilson, supra note 4 at paras 19–38.
2017] THE TIME HAS COME 89
When that time comes, our highest court will be tasked, once again, with making the judicial review process even “simpler” and even “more workable.”10 The question for this new panel must be whether the law of standard of review – once described as a juggling act with three seemingly transparent objects11 – requires only revision or “fundamental re-thinking.”12 Now, for this practitioner at least, the decision in Dunsmuir provides a solid foundation for any future “recalibration.”13
A. Principles of Legislative Supremacy and the Rule of Law
The Supreme Court of Canada in Dunsmuir helpfully identified and described the basic legal principles that animate judicial review: legislative supremacy and the rule of law.14 It is important to emphasize, however, that these principles do not just explain the purpose of judicial review. They also guide “its function and operation.”15 These principles provide the doctrinal bases for two other operational rules, namely the deference extended to administrative decision-makers operating at first instance and the supervisory function assigned to courts conducting independent review. And while courts will have “the last word” on some questions of general 2017 CanLIIDocs 175 law, they no longer “have a monopoly on deciding all questions of law.”16 Standard of review must balance both of these foundational principles.
B. Principle of Deference
In Dunsmuir, the Supreme Court of Canada also embraced the principle of deference in substantive review.17 But deference is not just an attitude that must be assumed by the court. It is also “a requirement of the law of judicial review.”18 In its attitudinal sense, deference is unhelpfully described by what it is not. It does not require a court
10 Dunsmuir, supra note 5 at paras 32, 43.
11 Miller v Newfoundland (Workers’ Compensation Commission) (1997), 154 Nfld & PEIR 52 at para 27, 2 Admin LR (3d) 178 (SC (TD)).
12 Wilson, supra note 4 at para 72.
13 Daly, “Meaning of Reasonableness”, supra note 8 at 827. Interestingly, this language of “recalibration” from Professor Daly found its way into the reasons of Justice Karakatsanis in Edmonton East, supra note 7 at para 20.
14 Dunsmuir, supra note 5 at paras 31–37.
15 Ibid at para 27.
16 Ibid at para 30.
17 Ibid at paras 48–50. This is not to say that the principle of deference was new. Its historical roots may be traced to CUPE, supra note 8 at 236, where Justice Dickson (as he then was), writing on behalf of the Supreme Court of Canada, emphasized the need for “judicial restraint” when considering interpretive questions falling within “the specialized jurisdiction” of an administrative decision-maker.
18 Ibid at para 48.
90 UNBLJ RD UN-B [VOL/TOME 68 to be “subservient” or to show “blind reverence.”19 And it is neither “lip service” nor “submission.”20 Rather, it is said to be “respectful attention” for the reasons supporting an administrative decision.21 The legal requirement of deference, however, is often obscured by this type of descriptive language.22 The obligation arises from the expression of legislative choice; that is, from the “governmental decisions to create administrative bodies with delegated powers.”23 At its core, deference is respect for that exercise of legislative authority. It is “not a gift conferred by the court.”24 Standard of review must therefore recognize deference as a legal obligation and not simply a mindset.
As noted in Dunsmuir, the role of the court is a supervisory one.25 The “triumph”26 of reasonableness “[did] not pave the way for a more intrusive review.”27 Rather, judges were directed to inquire into the reasons offered – and the outcome reached – by the decision-maker under review.28 Even in the apparent absence of deference, a reviewing court was still told to ask “whether the tribunal’s decision was correct” and to “decide whether it agree[d] with the determination of the decision-maker.”29 In other words, while the last word on certain legal questions was reserved for the court, Dunsmuir emphasized that the court no longer had the 2017 CanLIIDocs 175 only word. Standard of review must always allow for judicial scrutiny. But the priority of the administrative decision-maker must now be acknowledged.
19 Ibid.
20 Ibid.
21 Ibid, citing David Dyzenhaus, “The Politics of Deference: Judicial Review and Democracy” in Michael Taggart, ed, The Province of Administrative Law (Oxford: Hart Publishing, 1997) 278 at 286 [Dyzenhaus, “Politics of Deference”]. See also Ryan v Law Society (New Brunswick), 2003 SCC 20 at para 49, [2003] 1 SCR 247 [Ryan].
22 See generally Paul Daly, “The Language of Administrative Law” Can Bar Rev [forthcoming] at 21, online:
23 Dunsmuir, supra note 5 at para 48, citing Canada (AG) v Mossop, [1993] 1 SCR 554 at 596, 100 DLR (4th) 658.
24 Right Honourable Chief Justice Beverley McLachlin, “‘Administrative Law Is Not for Sissies’: Finding a Path through the Thicket” (2016) 29 Can J Admin L & Prac 127 at 133 [McLachlin, “Finding a Path”].
25 Dunsmuir, supra note 5 at para 28.
26 Honourable Justice John Evans, “Triumph of Reasonableness: But How Much Does It Really Matter?” (2014) 27 Can J Admin L & Prac 101.
27 Dunsmuir, supra note 5 at para 48.
28 Ibid at para 47.
29 Ibid at para 50 [emphasis added].
2017] THE TIME HAS COME 91
C. Principle of Contextual Review
Finally, the Supreme Court of Canada recognized in Dunsmuir that judicial review is a “contextual” exercise.30 It was said that context “always” informs the interpretation of the law.31 For that reason, an administrative decision has to be considered not only in light of the “legal context” in which the decision-maker is operating, but also “the context of the legislative wording.”32 In short, the relevant context will vary “with the relevant circumstances.”33 However, this contextual exercise is also intended to yield “a range of possible, acceptable outcomes which are defensible in respect of the facts and law.”34 Intervention by the court is to be limited to situations “where justice requires it, but not otherwise.”35 Standard of review must therefore not just have the capacity to operate in a variety of administrative environments. It must also have the ability to reveal justifiable outcomes within each of those fields.
There is a basic soundness in these principles from Dunsmuir. And they provide a solid foundation for any future revision. By focusing on them, instead of the categories created in Dunsmuir for sorting administrative decisions, substantive 2017 CanLIIDocs 175 review has the potential to be simplified even further.
Part II: The Story Since Dunsmuir
In the cases since Dunsmuir, it has become, as Justice Abella noted in Tervita, “increasingly difficult to discern the demarcations between a reasonableness and correctness analysis.”36 This difficulty arises, in part, from the practical reality that, if a reviewing court wishes to intervene, it is capable of finding a way to do so – regardless of the standard of review. Even the Supreme Court of Canada itself has struggled with coherence. For lawyers and litigants, it has felt at times that the
30 Ibid at para 64. While perhaps most overt in Dunsmuir, context has long been relied upon by the Supreme Court of Canada during the substantive review process. See e.g. Bibeault, supra note 8 at paras 120, 141, 161, and 185, where Justice Beetz embraced the view that “context” was a necessary consideration in the judicial review of administrative action. See also CUPE, supra note 8 at 240.
31 Dunsmuir, supra note 5 at para 74.
32 Ibid at paras 74, 76.
33 Ibid at para 150.
34 Ibid at para 47.
35 Ibid at para 43.
36 Tervita, supra note 9 at para 170. See e.g. Commission scolaire de Laval v Syndicat de l’enseignement de la région de Laval, 2016 SCC 8, [2016] 1 SCR 29 [Laval], where the Supreme Court of Canada, sitting as a panel of seven judges, unanimously agreed to dismiss the appeal but divided sharply on the applicable standard of review. Writing on behalf of the three judges who preferred the correctness standard, Justice Côté ultimately conceded at paragraph 86 that “the result is the same regardless of whether the applicable standard is correctness or reasonableness.”
92 UNBLJ RD UN-B [VOL/TOME 68 juggling act has continued. And the only thing that has changed is the number of objects in the air.
On the particular subject of standard of review, the cases decided by the Supreme Court of Canada after Dunsmuir have given rise to at least three practical complaints. First, the Court has sometimes failed to mention37 or even decide38 the applicable standard of review. Second, the Court, having chosen one standard of review, has appeared to apply another.39 Third, having directed parties to focus on the merits, the Court itself has become deeply divided on the preliminary question of standard of review.40 The result has been needless confusion for litigants, lawyers, and reviewing courts. What was intended to be a “more coherent and workable”41 framework for substantive review has become a “labyrinth.”42 But, before proposing to simplify the entry to judicial review, it is important to understand what we must try to avoid. With that objective in mind, a sample of this “imperfect”43 work from our highest court is examined below.
In Bombardier,44 the Supreme Court of Canada considered a decision by the
Quebec Human Rights Tribunal which found that the aerospace company had 2017 CanLIIDocs 175 discriminated against a pilot by refusing to allow him to participate in a flight training program. That refusal was rooted in an earlier decision by American authorities on grounds of national security. After investigation, the complaint proceeded before the Tribunal and damages were ordered. An appeal to the Quebec Court of Appeal was successful, and the decision by the Tribunal was set aside. At the Supreme Court of Canada, the appeal was dismissed. In a judgment delivered on behalf of the Court, however, Justices Wagner and Côté provided no reasons on the subjects of deference or standard of review. Instead, the Court embarked on its own review of the evidence in the record and, ultimately, it concluded that the decision was unsupported and therefore unreasonable. These types of omissions have done nothing to advance the predictability or clarity of the substantive review process.
37 See e.g. Quebec (Commission des droits de la personne et des droits de la jeunesse) v Bombardier Inc (Bombardier Aerospace Training Center), 2015 SCC 39, [2015] 2 SCR 789 [Bombardier]. See also Febles v Canada (Citizenship and Immigration), 2014 SCC 68, [2014] 3 SCR 431.
38 See e.g. B010 v Canada (Citizenship and Immigration), 2015 SCC 58, [2015] 3 SCR 704 [B010].
39 See e.g. Canada (Canadian Human Rights Commission) v Canada (AG), 2011 SCC 53, [2011] 3 SCR 471 [Mowat]. See also Martin v Alberta (Workers’ Compensation Board), 2014 SCC 25, [2014] 1 SCR 546.
40 See e.g. Edmonton East, supra note 7. See also Laval, supra note 36.
41 Dunsmuir, supra note 5 at para 32.
42 Wilson, supra note 4 at para 19.
43 To borrow a word from Jocelyn Stacey & Alice Woolley, “Can Pragmatism Function in Administrative Law?” (2016) 74 SCLR (2d) 211 at 2, online:
44 Supra note 37.
2017] THE TIME HAS COME 93
In Mowat,45 the Supreme Court of Canada considered whether the Canadian Human Rights Tribunal had the authority to award legal costs. The Tribunal itself had concluded that it did and granted an award. That award was upheld by the Federal Court but set aside by the Federal Court of Appeal, which found that the Tribunal had no such authority. At the Supreme Court of Canada, the reasonableness standard of review was found to be applicable and deference due. However, immediately after making those findings, Justices LeBel and Cromwell, writing on behalf of the Court, proceeded to interpret the enabling legislation, including its history and surrounding context, without any mention of the reasons why the Tribunal reached the outcome that it did. This seeming disregard for the justification offered by the delegate with “primary responsibility”46 for making the decision under review has only contributed to confusion about how the reasonableness standard is to be applied in practice.
Finally, in Edmonton East,47 the Supreme Court of Canada considered whether a local assessment review board had the statutory ability to increase – and not just lower or confirm – a tax assessment under review. Following a statutory appeal, the Alberta Court of Queen’s Bench set aside the decision of the local board. 2017 CanLIIDocs 175 That judgment was later affirmed by the Alberta Court of Appeal. At the Supreme Court of Canada, the appeal was allowed and the decision of the local board reinstated. However, the Court was a fractured one. Justice Karakatsanis, writing on behalf of five judges, concluded that the proper standard of review was reasonableness. Not less than twenty-one paragraphs were devoted to that preliminary issue. Justices Côté and Brown, on the other hand, concluded on behalf of four judges that the standard of review was correctness and explained the grounds for that position over the course of twenty-six paragraphs. Such division on a subject other than the merits of the administrative decision has only served to reinforce – unhelpfully – for litigants and lawyers that “[t]he disposition of the case may [still] well turn on the choice of standard of review.”48
In summary, there is work to do. While there is no doubt that some of these decisions can be usefully distinguished as “noise,”49 there must be a simpler way forward. However, in addition to being principled, any new framework must also be practical. It must be capable of being stated and understood quickly, it must avoid unnecessary discussion about preliminary matters that are secondary to the merits, and it must focus litigants, lawyers, and judges on explaining why a particular result
45 Supra note 39.
46 Ryan, supra note 21 at para 50.
47 Supra note 7.
48 Dunsmuir, supra note 5 at para 133.
49 Paul Daly, “The Signal and the Noise in Administrative Law” (Paper delivered at the Law Society of Upper Canada’s Annual Immigration Law Summit, 23 November 2016), online:
94 UNBLJ RD UN-B [VOL/TOME 68 is justified or not. The order is a tall one. But, as the late Justice Scalia once said, “[a]dministrative law is not for sissies.”50
Part III: Revisiting Our Understanding of Legislative Supremacy and the Rule of Law
Moving forward, the temptation to juxtapose the rule of law and legislative supremacy must be resisted. Both foundational principles recognize a legitimate function for courts and administrative decision-makers in our system of justice. Any tension that may exist – in theory or in practice – can be resolved by a standard of review that insists upon respect for the reasons offered by administrative decision- makers and upon justifications from reviewing courts when they depart from them. But this insistence upon “justifiability”51 from both courts and administrative decision-makers would not just strike a sound balance between the rule of law and legislative supremacy. It would also recognize that the task of interpreting and applying the law is now a shared one.
2017 CanLIIDocs 175
A. Rule of Law
The rule of law is no longer the monopoly of courts. The “war” between administrative decision-makers and courts has ended.52 And the rule of law – once thought to be the very “opposite”53 of administrative law – has matured to recognize that judging is a pluralist exercise in a modern state like Canada. While this is not to say that no tension remains, it is clear that the court has moved from being “a brute guardian” of the rule of law to “a partner” in its construction and protection.54 It is now recognized that “administrative tribunals have an integral role in the maintenance of our legal order.”55 Whether this is the result of more sophistication,
50 Honourable Justice Antonin Scalia, “Judicial Deference to Administrative Interpretations of Law” (1989) 3 Duke LJ 511 at 511. For a recent discussion of this pronouncement in the context of Canadian administrative law, see McLachlin, “Finding a Path,” supra note 24 at 127–134.
51 Honourable Justice Louis LeBel, “Some Properly Deferential Thoughts on Deference” (2008) 21 Can J Admin L & Prac 1 at 18. David Dyzenhaus has also described justification as a requirement of deference. See David Dyzenhaus, “Dignity in Administrative Law: Judicial Deference in a Culture of Justification” (2012) 17 Rev Const Stud 87 at 109–114 [Dyzenhaus, “Culture of Justification”].
52 Robert Reid, “Hot Buttons: An Overview of Recent Developments in Administrative Law” in Philip Anisman and Robert Reid, eds, Administrative Law: Issues and Practice (Toronto: Carswell, 1995) 1 at 8.
53 Right Honourable Lord Hewart, The New Despotism (London: Ernest Benn Ltd, 1929) at 37.
54 Right Honourable Chief Justice Beverley McLachlin, “The Roles of Administrative Tribunals and Courts in Maintaining the Rule of Law” (1998-1999) 12 Can J Admin L & Prac 171 at 175 [McLachlin, “Rule of Law”]. See also Mary Liston, “Governments in Miniature: The Rule of Law in the Administrative State” in Colleen M Flood & Lorne Sossin, eds, Administrative Law in Context, 2nd ed (Toronto: Emond Montgomery, 2013) 39 at 82, where Professor Liston has observed that “[a]ll parts of … the state participate in the creation and maintenance of the rule of law.”
55 Ibid at 173.
2017] THE TIME HAS COME 95 additional context, or just plain necessity is not known,56 but it is certain that both – courts and administrative decision-makers – are here to stay.
What courts and administrative decision-makers do share is a commitment to reasoned justifications for the exercise of their respective powers. In a modern democratic society like ours, any legitimate exercise of public authority must be capable of justification. This expectation, according to Dyzenhaus, is the sign of a mature rule of law.57 It is also expected that arbitrary or irrational decisions will be subject to independent scrutiny.58 These common threads run through all institutions operating under the rule of law and, over time, they have been stitched together to create a culture or “ethos of justification.”59 In short, the rule of law “can speak in several voices.”60 And for most individuals, it does not matter whether that voice is an administrative or judicial one. The outcome is the same.
Standard of review must reflect this theoretical evolution of the rule of law as well as its practical reality. There is a role for both courts and administrative decision-makers.61 Each is also grounded in the same foundational principle: there must be a reasoned justification for any exercise of their legal authority. And while 2017 CanLIIDocs 175 the source of that authority is different, it is no less legitimate or credible.62
56 For a recent account of the origins of the Canadian administrative state, see Colleen M Flood & Jennifer Dolling, “An Introduction to Administrative Law: Some History and a Few Signposts for a Twisted Path” in Colleen M Flood & Lorne Sossin, eds, Administrative Law in Context, 2nd ed (Toronto: Emond Montgomery, 2013) 1 at 3–23.
57 Dyzenhaus, “Politics of Deference”, supra note 21 at 278–307.
58 H Wade MacLauchlan, “Reconciling Curial Deference with a Functional Approach in Substantive and Procedural Judicial Review” (1993) 7 Can J Admin L & Prac 1 at 4–6.
59 McLachlin, “Rule of Law”, supra note 54 at 174. See also Stacey & Woolley, supra note 43 at 11, where Professors Stacey and Woolley note that “public decisions gain their democratic and legal authority through a process of public justification in which all public decision-makers offer reasons that justify their decisions.”
60 Ibid at 175.
61 Professors Stacey and Woolley have described the roles of courts and administrative decision-makers as “complimentary and, to a significant extent, co-extensive.” See Stacey & Woolley, supra note 43 at 9. See also LeBel, supra note 51 at 18 and 20, where Justice LeBel observed that courts and administrative decision-makers “share a responsibility to maintain the rule of law” and that “both … have roles to play in preserving the rule of law.”
62 See e.g. Rasanen v Rosemount Instruments Ltd (1994), 17 OR (3d) 267 at 279–280, 112 DLR (4th) 683 (CA), where Justice Abella recognized that administrative decision-makers were designed to “resolve disputes in their area of specialization more expeditiously and more accessibly, but no less effectively or credibly.”
96 UNBLJ RD UN-B [VOL/TOME 68
B. Legislative Supremacy
Legislative supremacy is hardly supreme. An administrative decision-maker exercising delegated authority is limited by its enabling statute, its common law duty of fairness, and its constitutional boundaries. In our modern state, no authority – legislative, administrative, or judicial – is absolute.63 Only the Constitution is “the supreme law of Canada.”64 And anchored within that “fundamental law”65 – to borrow a dated phrase from Dicey – is not just protection for, but a guarantee of, judicial review.66 When considered in this contemporary light, it is clear that legislative authority is not boundless and there is a necessary, albeit supervisory, role for courts to review the work of legislated delegates.
The democratic principle has, in a word, matured. Neither Parliament nor the provincial legislatures intend their delegates to exercise authority in a manner that usurps the role of legislators themselves. They also do not intend to vest those administrative decision-makers with authority to infringe the rights of citizens or to act outside the boundaries of their delegated powers.67 Such conduct is objectionable and, if alleged, “courts have no choice but to hear ... and decide whether the 2017 CanLIIDocs 175 administrative board or tribunal has in fact exceeded the powers granted to it by its constating statute.”68 Judges are therefore recognized as having an independent function. As Justice Rand observed in Roncarelli, “there is always a perspective within which a statute is intended to operate.”69 That perspective is tempered, however, with respect for the choice made by the legislator to designate someone other than the court as the “primary” decision-maker.70
Standard of review must reflect not just the authority of legislators to delegate the task of decision-making to administrative actors, but also the role of
63 See generally Dyzenhaus, “Culture of Justification”, supra note 51 at 105–106. Professor Liston has described the relationship between the court and other branches of government as “a joint effort in governance.” See Liston, supra note 54 at 66.
64 The Constitution Act, 1982, being Schedule B to the Canada Act 1982 (UK), 1982, c 11, s 52(1). As our highest court so often reminds, the system of government in Canada is one of “constitutional supremacy.” See e.g. Reference re Supreme Court Act, ss 5 and 6, 2014 SCC 21 at para 89, [2014] 1 SCR 433.
65 AV Dicey, Introduction to the Study of the Law of the Constitution, 3rd ed (London: MacMillan, 1889) at 4.
66 Dunsmuir, supra note 5 at para 31.
67 See Dunsmuir, supra note 5 at para 131.
68 McLachlin, “Rule of Law”, supra note 54 at 178.
69 Roncarelli v Duplessis, [1959] SCR 121 at 140, 16 DLR (2d) 689 [Roncarelli].
70 Ryan, supra note 21 at para 50. As David Mullan has observed, the role of the court is not “micromanaging.” Its function is one of general oversight. See David Mullan, “Section 7 and Administrative Law Deference: No Room at the Inn?” (2006) 34 SCLR (2d) 227 at 236. See also LeBel, supra note 51 at 16–17.
2017] THE TIME HAS COME 97 courts to supervise those decisions. Each has a constitutional function, and neither one is supreme or absolute. Both, however, are legitimate.
Part IV: A Single Standard of Review for Reasonableness
A single standard of review for reasonableness would bring new predictability and clarity to the substantive review of administrative decisions. Our highest court could easily trace its doctrinal roots to Dunsmuir and strike a defensible balance between legislative supremacy and the rule of law. But most importantly for the individuals who are actually impacted by these decisions, the “obstacle course”71 that is standard of review would be replaced with a clear “runway”72 to the merits of them. Unproductive “lawyer’s talk” would be significantly reduced.73
A. Meaning of Reasonableness
Reasonableness, as conceived in Dunsmuir, examines “whether the decision falls 2017 CanLIIDocs 175 within a range of possible, acceptable outcomes which are defensible in respect of the facts and the law.”74 In short, it reviews the reasons provided by a decision- maker for a justifiable explanation of the result. In Newfoundland Nurses, the Supreme Court of Canada helpfully clarified that this process does not require “two discrete analyses.”75 Rather, the exercise is an “organic” one where “the reasons must be read together with the outcome and serve the purpose of showing whether the result falls within a range of possible outcomes.”76 Embracing this outcome- oriented conception of reasonableness would, as Justice Abella observed in Wilson, allow our highest court to capture “the animating principles of both former categories of judicial review.”77
Notwithstanding the rare and exceptional nature of questions said to require a “correct” answer, there will be concern that a single standard of review for reasonableness could prevent a reviewing court from properly safeguarding the rule of law or – even worse – result in the court abdicating its constitutional duty of judicial review.78 However, Dunsmuir itself recognized that this duty only requires
71 Wilson, supra note 4 at para 20.
72 Ibid at para 25.
73 Dunsmuir, supra note 5 at para 133.
74 Ibid at para 47.
75 Newfoundland and Labrador Nurses' Union v Newfoundland and Labrador (Treasury Board), 2011 SCC 62 at para 14, [2011] 3 SCR 708 [Newfoundland Nurses].
76 Ibid.
77 Wilson, supra note 4 at para 33.
98 UNBLJ RD UN-B [VOL/TOME 68 the court to have “the last word” on the legal boundaries of administrative decision- making.79 It does not require the court to have the only word. In the years since Dunsmuir, the Supreme Court of Canada has refined the reasonableness standard to recognize that there will be occasions when only one defensible outcome exists.80 It is this capacity to recognize a single result that now answers any constitutional concern for the rule of law.
It must also be recognized that the task of interpreting and applying the law is a shared one.81 In this current culture of justification, it is not clear why the reasoning of an administrative decision-maker would be ignored when a court is reviewing the answers to certain questions – but not others.82 Regardless of how they are labelled on judicial review, all of the questions were before the “primary” decision-maker for consideration.83 As Justice Abella has observed, nothing in Dunsmuir “precludes the adoption of a single standard of review, so long as it accommodates the ability to continue to protect both deference and the possibility of a single answer where the rule of law requires it.”84 In other words, a contextualized review for reasonableness will still provide a final, judicial answer to “the four 85 categories [of questions] singled out for correctness review in Dunsmuir.” And if 2017 CanLIIDocs 175 deference is truly mutual respect, then it means “jettisoning the correctness standard”86 for even these types of questions where the administrative decision- maker has provided a reasoned justification at first instance. As the judges of our highest court have come to find, notwithstanding their disagreement about the applicable standard of review, the outcome at the end of the day is often the same.87
78 See e.g. Lauren J Wihak, “Whither the Correctness Standard of Review? Dunsmuir, Six Years Later” (2014) 27 Can J Admin L & Prac 173.
79 Dunsmuir, supra note 5 at para 30.
80 See Mowat, supra note 39. See also McLean v British Columbia (Securities Commission), 2013 SCC 67 at para 38, [2013] 3 SCR 895 [McLean], where Justice Moldaver, writing for a majority of six judges, reasoned that there will be cases where the range of reasonable outcomes “will necessarily be limited to a single reasonable interpretation – and the administrative decision maker must adopt it.”
81 See McLachlin, “Rule of Law”, supra note 54 at 185–189. See also LeBel, supra note 51 at 18–19. This legal pluralism was also recognized in Dunsmuir, supra note 5 at para 30, where the Supreme Court of Canada cautioned against a “court-centric conception of the rule of law” and acknowledged that “courts do not have a monopoly on deciding all questions of law.”
82 Sheila Wildeman has made a similar query. See Sheila Wildeman, “Pas de Deux: Deference and Non- Deference in Action” in Colleen M Flood & Lorne Sossin, eds, Administrative Law in Context, 2nd ed (Toronto: Emond Montgomery, 2013) 323 at 334.
83 Ryan, supra note 21 at para 50.
84 Wilson, supra note 4 at para 31 [emphasis in original].
85 Ibid.
86 See Dyzenhaus, “Culture of Justification”, supra note 51 at 109. See also David Dyzenhaus, “David Mullan’s Theory of the Rule of (Common Law)” in Grant Huscroft & Michael Taggart, eds, Inside and Outside Canadian Administrative Law: Essays in Honour of David Mullan (Toronto: University of Toronto Press, 2006) 448 at 462, 475.
2017] THE TIME HAS COME 99
This practical consequence must not be lost in a theoretical debate about the approach to judicial review.
Admittedly, substantive review on a reasonableness standard is a deceptively simple innovation in the direction of judicial review. As Justice Binnie colourfully cautioned in the last revision, this type of traffic engineering may do nothing more than “shift rush hour congestion from one road intersection to another without any overall saving to motorists in time or expense.”88 And while there is a benefit to ending the “terminological battles”89 and “rhetorical debates,”90 the real prize for litigants, lawyers, and judges continues to be that stated in Dunsmuir itself: “a principled framework that is more coherent and workable.”91 With that objective in mind, this article proposes a single, contextual standard of review. If properly anchored, a single standard of review for reasonableness has the potential to operate in a principled and practical way that is focused on the merits of the administrative decision under review. While the exact number of those anchors is a matter for decision by our highest court, it seems to this practitioner that the following could provide “real guidance”92 and “get the parties ... back to arguing 93 about the substantive merits of their case”: (i) deference; (ii) reasons; and (iii) 2017 CanLIIDocs 175 context.
B. Requirement for Deference
Deference must be a requirement of substantive review. Litigants, lawyers, and judges would all begin their analyses by recognizing that the outcome reached by the administrative decision-maker (and the reasons for it) are entitled to – and not just deserving of – respect. While much emphasis has been placed on the proper “attitude” to be adopted by reviewing courts, returning focus to the reality that deference is “a requirement of the law of judicial review”94 would serve to reinforce
87 See e.g. Laval, supra note 36, where the Supreme Court of Canada divided sharply on the applicable standard of review, but ultimately agreed on the outcome. See also Alberta (Information and Privacy Commissioner) v University of Calgary, 2016 SCC 53, [2016] 2 SCR 555, where a majority of the Supreme Court of Canada applied the correctness standard. Justice Abella, writing partially concurring reasons, reached the same outcome as the majority, but applied the reasonableness standard of review. For additional support, see Tervita, supra note 9, where Justice Abella concurred in the final result, but did so using a different standard of review.
88 Dunsmuir, supra note 5 at para 139.
89 Wilson, supra note 4 at para 25.
90 Ibid at para 24.
91 Dunsmuir, supra note 5 at para 32.
92 Ibid at para 1.
93 Ibid at para 145.
94 Ibid at para 48.
100 UNBLJ RD UN-B [VOL/TOME 68 that a single standard of review is not “carte blanche”95 for intervention by the court. A reviewing judge would be required to examine whether there is a justifiable explanation for the outcome reached by the administrative decision-maker and, if so, respect it.96 As the Supreme Court of Canada has found, reasonableness review will generate a single outcome in exceptional cases when context requires it.97 But deference, which arises from the governmental choice to create the administrative decision-maker, must be part of the entire review process. It is an obligation – not a gift.98
Moving to a single standard for the judicial review of administrative decisions will no doubt give rise to some temptation to relegate deference in the analysis. After all, the standard would now have to guard against “incorrect” answers to questions in a number of law-laden areas.99 However, as was the case in Dunsmuir, our highest court must remain committed to the view that reasonableness “does not pave the way for a more intrusive review by courts”100 and “should not be seen by potential litigants as a lowering of the bar to judicial intervention.”101 Regardless of how the question was previously sorted for the purpose of substantive review, that question was always one assigned to a decision-maker other than the 2017 CanLIIDocs 175 court. And that legislative choice remains entitled to deference or, in the words of Justice Fichaud, “a dose of judicial humility.”102
Now, given the spectrum of administrative decision-makers, there has always been some question whether there must also be varying degrees of deference. The Federal Court of Appeal thought so.103 And even in Dunsmuir, Justice Binnie suggested that, by collapsing deferential review into a single reasonableness standard, a reviewing court would sometimes be required to act “more deferentially”
95 Catalyst Paper Corp v North Cowichan (District), 2012 SCC 2 at para 24, [2012] 1 SCR 5 [Catalyst Paper].
96 Dyzenhaus, “Culture of Justification”, supra note 51 at 113. See also Stacey & Woolley, supra note 43 at 13.
97 See Mowat, supra note 39. See also McLean, supra note 80 at para 38.
98 McLachlin, “Finding a Path”, supra note 24 at 133.
99 See Dunsmuir, supra note 5 at paras 58–61. See also Smith v Alliance Pipeline Ltd, 2011 SCC 7 at para 26, [2011] 1 SCR 160. However, now that the Supreme Court of Canada has refined the reasonableness analysis so that it is capable of generating a single defensible outcome, it is worth asking, as Justice Abella did in Wilson, supra note 4 at para 24, whether the historical label of “correct” has any real meaning: “Are we not saying essentially the same thing when we conclude that there is only a single ‘reasonable’ answer available and when we say it is ‘correct’?”
100 Dunsmuir, supra note 5 at para 48.
101 Ibid at para 155.
102 Honourable Justice Joel Fichaud, “Between a Rock and a Hard Place: Deference, Consistency and Transparency in Administrative Decision-making” (Remarks delivered at the Canadian Bar Association’s National Administrative Law, Labour and Employment Law Conference, 18 November 2016).
103 See e.g. Wilson, supra note 4 at para 18.
2017] THE TIME HAS COME 101 and, in other cases, “less deferentially.”104 Most recently, however, the Supreme Court of Canada has endeavoured to draw a distinction between deference and the context within which an administrative decision is made. Some members of the Court have even gone so far as to expressly reject the potential for an “indeterminate number of varying degrees of deference.”105 For litigants and their lawyers, cementing this position would be a welcome development by our highest court.
Deference as a singular obligation, absent any conception of a spectrum of degrees or a sliding scale, would rightly direct parties to the merits of the administrative decision being reviewed. That, after all, is the very purpose of substantive review. Deference would be grounded in respect for the legislative choice that has been made rather than the nature or expertise of any particular decision-maker.106 This is not to say that the latter factors are irrelevant. But, in the interest of creating a framework that is simpler and more workable for parties, those factors would be better considered as part of the context surrounding the outcome under judicial review. As discussed below, those factors and others could assist lawyers, litigants, and reviewing courts in discerning the range of defensible results.
Adopting this contextual approach would also be, to borrow the words of Justice 2017 CanLIIDocs 175 Abella in Wilson, “a principled way to simplify the path to reviewing the merits.”107 One more potential obstacle would be removed from the course.
C. Insistence Upon Reasons
Reasons must be the starting point for – and remain the focus of – substantive review. The analyses of litigants, lawyers, and judges would all examine what was said by the decision-maker who was delegated the authority to make the decision in the first place. As Dyzenhaus has noted, deference requires not just that an administrative decision-maker justify its conclusion.108 It also requires a reviewing court to examine that justification109 and resist the temptation to “undertake its own analysis of the question.”110 This commitment to the primacy of reasons not only serves to reinforce the legislative choice that has been made, but it also contributes to the rule of law by forcing the court to justify any departure from the reasons offered
104 Dunsmuir, supra note 5 at para 152.
105 Wilson, supra note 4 at paras 18, 73.
106 As Professor Liston has observed, “courts are conscious of the separation of powers and … are themselves under rule-of-law constraints to respect legislative and executive branches.” See Liston, supra note 54 at 65. This constitutional principle alone is a sound doctrinal basis for deference.
107 Wilson, supra note 4 at para 20.
108 Dyzenhaus, “Culture of Justification”, supra note 51 at 113.
109 Ibid.
110 Dunsmuir, supra note 5 at para 50.
102 UNBLJ RD UN-B [VOL/TOME 68 by the administrative decision-maker.111 In short, it answers the question that matters most to parties: why?
This attention on reasons is not new. In Southam, an unreasonable decision was described as “one that … is not supported by any reasons that can stand up to a somewhat probing examination.”112 Later, in Ryan, an outcome was said to be unreasonable when “there are no lines of reasoning supporting the decision which could reasonably lead [the] tribunal to reach the decision it did.”113 In Dunsmuir, the Supreme Court of Canada went on to explain that reasonableness is “concerned mostly with the existence of justification, transparency and intelligibility within the decision-making process.”114 Finally, in Newfoundland Nurses, reviewing courts were instructed to examine the outcome “in the context of the evidence, the parties’ submissions and the process.”115 In summary, any administrative decision must be reviewed in light of the whole record before the decision-maker and any intervention by the court explained from within that record.
Much of the confusion that has arisen since Dunsmuir has resulted from cases where the Supreme Court of Canada has commenced102 substantive review 2017 CanLIIDocs 175 with its own analysis and not that of the administrative decision-maker. Respectful attention sometimes looks and feels like careless disregard. By insisting that litigants, lawyers, and judges examine the work done by the delegate under review, all would be required to provide explanations for departing from the line of reasons chosen by the decision-maker who was actually granted the authority to make the decision. The justification for any departure by a reviewing court would therefore come from within the administrative decision – not from without. This requirement would not just be consistent with the ethic of justification that animates the rule of law. It would also reflect the proper role of courts, who are charged with supervising only the “outer boundaries” of legislative supremacy.116
111 See generally Liston, supra note 54 at 76. Professor Liston has noted that reasons have “the potential to advance both restraint and respect.” Reasons from an administrative decision-maker provide an opportunity to illustrate competence and expertise. Judicial recognition of those reasons then constrains the ability of a reviewing court to re-weigh the original factors, but still allows the court to confirm specific instances of reasonable decision-making.
112 Southam, supra note 8 at para 56.
113 Ryan, supra note 21 at para 53.
114 Dunsmuir, supra note 5 at para 47.
115 Newfoundland Nurses, supra note 75 para 18. See also Construction Labour Relations v Driver Iron Inc, 2012 SCC 65 at para 3, [2012] 3 SCR 405, where the Supreme Court of Canada directed reviewing courts to ask “whether the decision, viewed as a whole in the context of the record, is reasonable.”
116 Dunsmuir, supra note 5 at para 141.
2017] THE TIME HAS COME 103
D. Context in Operation
Context must become the essence of substantive review. Having been directed immediately to the nodes contained in the administrative reasoning process, all parties would be called upon to answer – and quickly – the question at the heart of judicial review: whether that reasoning leads to a result that is defensible in fact and law. Or, in other words, an outcome that is reasonable. As Justice Binnie observed in Dunsmuir, “[a] driving speed that is ‘reasonable’ when motoring along a four-lane interprovincial highway is not ‘reasonable’ when driving along an inner city street.”117 Context therefore “always matters.”118 And the range of what is reasonable “will necessarily vary.”119 In short, it is here that the heavy lifting will have to be done by litigants, lawyers, and reviewing courts.
In Dunsmuir, the Supreme Court of Canada recognized that “the law does not operate in a vacuum”120 and, when called upon to consider the underlying decision “as a whole,”121 it concluded that the interpretation offered by the arbitrator was “unreasonable in the context of the legislative wording and the larger labour 122 context in which it [was] embedded.” In a later decision, Chief Justice McLachlin 2017 CanLIIDocs 175 explained that reasonableness “must be assessed in the context of the particular type of decision making involved and all relevant factors.”123 Reasonableness, it was said, “takes its colour from the context.”124 Context can therefore be used by parties to demonstrate the number of defensible outcomes available and, in some exceptional cases, yield the only reasonable result.
Of course, determining the content of that context will be the most difficult passage in any future revision by our highest court.125 Yet another “threshold test”126 would do little, however, to refocus parties on the substantive result under review. Daly, for his part, has helpfully suggested that the range of defensible outcomes “be determined by reference to contextual factors drawn from the rule of law and democratic principles.”127 While those factors, which would expand or contract the
117 Ibid at para 150.
118 Edmonton East, supra note 7 at para 73, Côté and Brown JJ, dissenting.
119 Wilson, supra note 4 at para 22.
120 Dunsmuir, supra note 5 at para 74.
121 Ibid at para 72.
122 Ibid at para 76.
123 Catalyst Paper, supra note 95 at para 18.
124 Ibid, citing Canada (Citizenship and Immigration) v Khosa, 2009 SCC 12 at para 59, [2009] 1 SCR 339.
125 Edmonton East, supra note 7 at para 20.
126 Dunsmuir, supra note 5 at para 154.
127 Daly, “Struggling Towards Coherence”, supra note 3 at 29.
104 UNBLJ RD UN-B [VOL/TOME 68 range of outcomes in each case,128 will obviously have to vary from decision to decision, Dunsmuir and subsequent cases already appear to provide some workable factors that are known to – and understood by – even lawyers:
i. The nature of the administrative decision-maker may be relevant.129 A delegate applying guidelines issued by the Minister does not enjoy the same range of outcomes as the Minister herself, who is charged with providing general direction on public policy. As the Supreme Court of Canada has historically said, “[t]he very nature of the body must be taken into account in assessing the technique of review.”130
ii. The type of question or issue may be relevant.131 An administrative decision-maker exercising discretion in the “national interest”132 has a wider choice of defensible outcomes than one deciding a matter of constitutional law. As Justice Binnie observed in Dunsmuir, the issue to be decided “helps to define the
range of reasonable outcomes within which the administrator is 2017 CanLIIDocs 175 authorized to choose.”133
iii. The content of the statutory scheme may be relevant.134 This could include the purpose or rationale of the statute, its text and legislative history, and whether the statute includes a privative clause or right of appeal. Notwithstanding the breadth of any grant of statutory authority, “legislators do not intend results that depart from reasonable standards.”135 And the range of defensible outcomes does not include a result that “fundamentally contradicts” the object or purpose of an enabling statute.136
128 Ibid.
129 See e.g. Catalyst Paper, supra note 95 at para 23.
130 Inuit Tapirisat of Canada v Canada (AG), [1980] 2 SCR 735 at 753, 115 DLR (3d) 1, cited in Dunsmuir, supra note 5 at para 136.
131 See e.g. Catalyst Paper, supra note 95 at paras 19, 32.
132 Agraira v Canada (Public Safety and Emergency Preparedness), 2013 SCC 36, [2013] 2 SCR 559.
133 Dunsmuir, supra note 5 at para 138.
134 See e.g. Wilson v British Columbia (Superintendent of Motor Vehicles), 2015 SCC 47 at paras 26–41, [2015] 3 SCR 300. See also Catalyst Paper, supra note 95 at para 25; McLean, supra note 80 at paras 42– 50; and Mowat, supra note 39 at paras 43–52.
135 Dunsmuir, supra note 5 at para 131.
136 Halifax (Regional Municipality) v Canada (Public Works and Government Services), 2012 SCC 29 at para 54, [2012] 2 SCR 108.
2017] THE TIME HAS COME 105
iv. The relative expertise of the decision-maker may be relevant.137 In Dunsmuir, the Supreme Court of Canada cautiously guarded certain legal questions falling outside an administrative decision- maker’s “specialized area of expertise.”138 This was contrasted with cases where the decision-maker had “particular familiarity” with the legal questions at issue.139 Stated in other terms, the number of defensible outcomes will be increased in areas of administrative speciality and decreased in areas falling within the traditional expertise of the court.
Obviously, one cannot exhaustively define context for the purpose of substantive review.140 In fact, we must resist any temptation to do so. If the vice of formalism is to be avoided, there must be room for flexibility in future cases “about whose composition we remain ignorant.”141
Deference and context, however, ought to be conceptually distinct.142 Deference is aimed at ensuring that the required attitude, discipline, or humility is exercised by litigants, lawyers, and reviewing courts. It engenders respect. Context, 2017 CanLIIDocs 175 on the other hand, is aimed at the decision itself and the field within which the administrative decision-maker operates. It defines a range. Collapsing the two concepts has tended to result in “needless complexity”143 about varying degrees of judicial intrusiveness. Reasonableness must be a single, contextual standard of review. And while complaints about the uncertainty of context will remain,144 the reality, as Daly has pointed out, is that “context simply cannot be eliminated from
137 See e.g. Laval, supra note 36 at para 38. See also McLean, supra note 80 at paras 30–31.
138 Dunsmuir, supra note 5 at paras 55, 60.
139 Ibid at para 54.
140 For example, in some cases, context may include consideration of constitutional values. See e.g. Loyola High School v Quebec (AG), 2015 SCC 12, [2015] 1 SCR 613. In other cases, it may involve consideration of prior interpretations or decisions. See e.g. Mowat, supra note 39 at paras 53–56. In still other cases, it may require consideration of international obligations. See e.g. B010, supra note 38 at paras 47–66.
141 HLA Hart, The Concept of Law, 3rd ed (Oxford: Oxford University Press, 2012) at 129–130, cited in Edmonton East, supra note 6 at para 70, Côté and Brown JJ, dissenting. For a classic discussion of formalism in the administrative context, see H Wade MacLauchlan, “Judicial Review of Administrative Interpretations of Law: How Much Formalism Can We Reasonably Bear?” (1986) 36:4 UTLJ 343.
142 A distinction drawn by Justice Deschamps in Dunsmuir, supra note 5 at para 167.
143 Dunsmuir, supra note 5 at para 128.
144 This would not, however, be the first time that a contextual approach was used by the Supreme Court of Canada to arrive at an objective outcome. See e.g. Baker v Canada (Minister of Citizenship and Immigration), [1999] 2 SCR 817, 174 DLR (4th) 193. The content of procedural fairness varies with the surrounding context, but it does not appear to suffer from the same uncertainty that presently surrounds substantive review. See generally Matthew Lewans, “Deference and Reasonableness Since Dunsmuir” (2012) 38:1 Queen’s LJ 59.
106 UNBLJ RD UN-B [VOL/TOME 68 judicial review.”145 If nothing else, these contextual factors – already known to lawyers and judges – give us a place to start.
Conclusion
As noted at the outset, this article is, by no means, a complete answer. It is open to criticism. For example, one may reasonably complain that the proposed framework simply moves all of the “real” work to the end – as opposed to the beginning – of the judicial review process. But, one may also reasonably ask whether that is a bad thing? By doing so, the proposal gives effect to the objective resting at the heart of Dunsmuir: it “get[s] the parties ... back to arguing about the substantive merits of their case.”146 Even after almost a decade, “we still find the merits waiting in the wings for their chance to be seen and reviewed.”147 The time has come to simplify the entry to judicial review.
Admittedly, a single standard of review for reasonableness will not end debate. As our highest court has found, even within that standard, there is room for 2017 CanLIIDocs 175 disagreement.148 However, instead of those debates focusing on the preliminary issue of what standard of review to apply to the administrative decision, litigants, lawyers and judges will all be focused on the question of why that decision is reasonable or unreasonable. This new emphasis on the outcome has always been the goal of substantive review and, for those affected by the decisions in question, the only thing that really matters.
By removing an artificial barrier to judicial review, cases will be decided on the reasonableness of their result and not on the choice of standard of review. With just one option available, lawyers will now be able to predict “with confidence” what standard will be applied.149 Litigants will have an accessible “runway”150 to independent review of the merits. And, over time, the case law will provide “real guidance”151 as to what defects are commonly found in unreasonable decisions and what markers are usually found in reasonable ones. Lawyers may even become bold,
145 Daly, “Struggling Towards Coherence”, supra note 3 at 16.
146 Dunsmuir, supra note 5 at para 145.
147 Wilson, supra note 4 at para 25.
148 See e.g. MM v United States of America, 2015 SCC 62, [2015] 3 SCR 973; Kanthasamy v Canada (Citizenship and Immigration), 2015 SCC 61, [2015] 3 SCR 909; Ontario (Energy Board) v Ontario Power Generation Inc, 2015 SCC 44, [2015] 3 SCR 147; and Canada (AG) v Igloo Vikski Inc, 2016 SCC 38, [2016] 2 SCR 80.
149 See Dunsmuir, supra 5 at para 133, where Justice Binnie made the practical observation that “[l]itigants understandably hesitate to go to court to seek redress for a perceived administrative injustice if their lawyers cannot predict with confidence even what standard of review will be applied.”
150 Wilson, supra note 4 at para 25.
151 Dunsmuir, supra note 5 at para 1.
2017] THE TIME HAS COME 107 predicting not just the standard of review but also the likelihood of a reasonable outcome. And all of this would be done without creating unnecessary “noise”152 about which standard of review should have applied or whether one standard of review was actually applied under the guise of another.153 It would be a “simpler test.”154
Now this article is not to say that the next chapter written by our highest court will be free of complexity. Complexity is inherent in any legal process that must accommodate a range of actors and a variety of issues.155 But there is comfort for litigants, lawyers, and reviewing courts. First, our judges are a capable group. As Justice Binnie observed in Dunsmuir, they are often called upon to “juggle a number of variables that are necessarily to be considered together.”156 Second, our case law continues to be useful. The Supreme Court of Canada has already provided – and will continue to provide – examples “to show when it is (or it is not) appropriate for a court to intervene in the outcome of an administrative decision.”157 Third, reasonableness is a legal concept that we commonly use and generally understand. As Justice Deschamps observed in Dunsmuir, “neither the concept of reasonableness nor that of deference is particular to the field of administrative 2017 CanLIIDocs 175 law.”158 In summary, a single standard of review for reasonableness, which is anchored in the context surrounding the administrative decision-maker itself, has the potential to operate in a principled yet practical way.
The Walrus and the Carpenter Were walking close at hand; They wept like anything to see Such quantities of sand: “If this were only cleared away,” They said, “it would be grand!”159
The Carpenter, of course, expressed his doubt whether the sand could ever be cleared away – even with seven mops.160 Lewis Carroll does not, however, tell us what two more could do. The Supreme Court of Canada, with its new group of nine, appears to be waiting for the right opportunity to revisit the standard of review in
152 Daly, “Signal and Noise”, supra note 49.
153 See generally Mullan, supra note 3 at 76–81.
154 To borrow the phrase used by the Supreme Court of Canada in Dunsmuir, supra 5 at para 43, where it concluded matter-of-factly: “A simpler test is needed.”
155 Dunsmuir, supra note 5 at para 132. See also Dunsmuir, supra note 5 at para 167.
156 Ibid at para 153.
157 Ibid at para 154.
158 Ibid at para 167.
159 Carroll, supra note 2 at 73.
160 Ibid.
108 UNBLJ RD UN-B [VOL/TOME 68 administrative law. The day may yet come when it is possible to advise a client or argue a case without a lengthy discussion about the standard of review.161 And that, said the lawyer, would be grand indeed. 2017 CanLIIDocs 175
161 Borrowing from the colourful reasons of Justice Slatter who, writing for the Alberta Court of Appeal, remarked: “The day may come when it is possible to write a judgment like this without a lengthy discussion of the standard of review. Today is not that day.” See Edmonton (City) v Edmonton East (Capilano) Shopping Centres Ltd, 2015 ABCA 85 at para 11, [2015] 5 WWR 547.
RENOVATING JUDICIAL REVIEW
Matthew Lewans
“Our administrative law is a never-ending construction site where one crew builds structures and then a later crew tears them down to build anew, seemingly without an overall plan.”1
I. Introduction
Like a hundred-year-old heritage home, the law governing judicial review of administrative action attracts conflicting opinions. Everyone agrees that it has a sound constitutional foundation, which guarantees2 that administrative decisions remain subject to independent scrutiny in order to ensure they are rendered fairly3 and comply substantively with the rule of law.4 But almost no one is satisfied with 2017 CanLIIDocs 175 the outdated aspects of its infrastructure, particularly the arcane “law office metaphysics”5 lawyers and judges employ in order to determine the standard of review independently of “the who, what, why and wherefor of the litigant’s complaint on the merits.”6
So every year administrative lawyers, judges, and law professors offer suggestions for reconstructing the law of judicial review by penning thoughtful
Associate Professor, University of Alberta. This is a revised version of a paper presented on 18 November 2016 at the Canadian Bar Association Annual Administrative Law, Labour and Employment Law Conference. Thanks to Ashton Menuz for her excellent research assistance and to David Mullan and the anonymous reviewers for their constructive criticism. Comments are welcome at [email protected].
1 David Stratas, “The Canadian Law of Judicial Review: A Plea for Doctrinal Coherence and Consistency” (February 17, 2016) at 1, online:
2 Crevier v Québec (AG), [1981] 2 SCR 220 at 236, 127 DLR (3d) 1 [Crevier].
3 David Mullan, “Fairness: The New Natural Justice?” (1975) 25 UTLJ 281; Nicholson v Haldimand- Norfolk Regional Police Commissioners, [1979] 1 SCR 311, 88 DLR (3d) 671 [Nicholson]; Knight v Indian Head School Division No 19 [1990] 1 SCR 653, 69 DLR (4th) 489 [Knight]; Baker v Canada (Minister of Citizenship and Immigration), [1999] 2 SCR 817, 174 DLR (4th) 193 [Baker]; Geneviève Cartier, “Procedural Fairness in Legislative Functions: The End of Judicial Abstinence?” (2003) 53 UTLJ 217.
4 Roncarelli v Duplessis, [1959] SCR 121, 16 DLR (2d) 689; Baker, supra note 3; David Dyzenhaus & Evan Fox-Decent, “Rethinking the Process/Substance Distinction: Baker v Canada” (2001) 51 UTLJ 193 at 240; David Dyzenhaus (ed), The Unity of Public Law (Oxford: Hart Publishing, 2004).
5 New Brunswick (Board of Management) v Dunsmuir, 2008 SCC 9 at para 122, [2008] 1 SCR 190 [Dunsmuir].
6 Ibid at para 154.
110 UNBLJ RD UN-B [VOL/TOME 68 opinions,7 extra-judicial essays,8 and academic articles.9 The common theme running through many of these contributions is that the Dunsmuir framework is flawed, but there remains deep disagreement about how to repair it. Some argue we should avoid sharp distinctions between different standards of review, because the intensity of judicial review is irreducibly variable and therefore must be calibrated to suit a particular administrative decision;10 some propose to demolish the categories reserved for correctness scrutiny,11 so as to facilitate reasonableness review across the board;12 and some propose to expand Dunsmuir’s categories for correctness scrutiny in order to impose some semblance of judicial order on an otherwise heterogeneous corpus of administrative law.13 And then there are those who think we
7 See e.g. Edmonton (City) v Edmonton East (Capilano) Shopping Centres Ltd, 2016 SCC 47, 402 DLR (4th) 236 [Edmonton East] reversing Edmonton East (Capilano) Shopping Centres Ltd v Edmonton (City), 2015 ABCA 85, 12 Alta LR (6th) 236; Wilson v Atomic Energy of Canada Ltd, 2016 SCC 29, [2016] 1 SCR 770 [Wilson]; Commission scolaire de Laval c Syndicat de l’enseignement de la region de Laval, 2016 SCC 8, [2016] 1 SCR 29 [Commission scolaire]; Canadian Broadcasting Corp v SODRAC 20013 Inc, 2015 SCC 57, [2015] 3 SCR 615 [Canadian Broadcasting Corp]; Kanthasamy v Canada (Citizenship and Immigration), 2015 SCC 61, [2015] 3 SCR 909 [Kanthasamy]; Canada (Human Rights Commission)
v Canada (AG), 2016 FCA 200, 402 DLR (4th) 160 [Canada (Human Rights Commission)]; Gitxaala 2017 CanLIIDocs 175 Nation v R, 2016 FCA 187, 485 NR 258; Kabul Farms Inc v R, 2016 FCA 143, 13 Admin LR (6th) 11; Canada (Minister of Citizenship and Immigration) v Huruglica, 2016 FCA 93, 396 DLR (4th) 527 [Huruglica]; Trinity Western University v Law Society of Upper Canada, 2016 ONCA 518, 131 OR (3d) 113 leave to appeal to SCC granted [2016] SCCA No 418; Intact Insurance Co v Allstate Insurance Co of Canada, 2016 ONCA 609, 131 OR (3d) 625; Québec v ED, 2016 QCCA 536, 265 ACWS (3d) 797 leave to appeal to SCC granted [2016] SCCA No 223; Kenyon v British Columbia (Superintendent of Motor Vehicles), 2015 BCCA 485, 82 BCLR (5th) 266; Trinity Western University v Law Society of British Columbia, 2016 BCCA 423, 92 BCLR (5th) 42 leave to appeal to SCC granted 2017 CarswellBC 504 (WL Can); Stewart v Elk Valley Coal Corporation, 2015 ABCA 225, 19 Alta LR (6th) 219 [Stewart] leave to appeal to SCC granted [2015] SCCA No 389; Nova Scotia Barristers’ Society v Trinity Western University, 2016 NSCA 59, 376 NSR (2d) 1; New Brunswick (Minister of Education) v Kennedy 2015 NBCA 58, 444 NBR (2d) 92.
8 Stratas, supra note 1; Simon Ruel, “What is the Standard of Review to be Applied to Issues of Procedural Fairness?” (2016) 29 Can J Admin L & Prac 259; Beverley McLachlin, “Administrative Law is not for Sissies: Finding a Path through the Thicket” (2016) 29 Can J Admin L & Prac 127.
9 David Jones, “Administrative Law in 2016: Update on Caselaw, Recent Trends and Related Developments” (2016) online:
10 Stratas, supra note 1; Hilary Cameron, “Substantial Deference and Tribunal Expertise post-Dunsmuir: A New Approach to Reasonableness Review” (2014) 27 Can J Admin L & Prac 1.
11 Dunsmuir, supra note 5 at paras 57–61.
12 Wilson, supra note 7 at paras 28–38. See also Falzon, supra note 9; David Dyzenhaus, “Dignity in Administrative Law: Judicial Deference in a Culture of Justification” (2012) 17 Rev Const Stud 87.
13 Ibid at para 89, per Côté and Brown JJ dissenting. See also Commission scolaire, supra note 7 at para 79, per Côté, Wagner and Brown JJ dissenting; Wilson v Atomic Energy of Canada Ltd, 2015 FCA 17, 467 NR 201; Huruglica, supra note 7 at para 52; Edmonton East, supra note 7; Stewart, supra note 7;
2017] RENOVATING JUDICIAL REVIEW 111 should forgo renovations altogether – at least for the time being – because they would destabilize an already wobbly analytical framework.14
In this paper, I want to highlight some of the more remarkable contributions to the debate regarding the standard of review over the past year. But before proceeding, I want to revisit the history to that debate, because if we lose track of important points of reference we might forget some hard-won lessons and thereby lose the ability to critically assess contemporary proposals for reform.
In Part II, I will briefly review the history of judicial review in order to identify two conflicting, but enduring, intellectual frameworks that have shaped that enterprise. The first, formalist intellectual framework relies upon a set of abstract analytical distinctions from which judges purport to deduce the parameters of judicial review without encroaching upon the merits of an administrative decision. The second intellectual framework conceives of judicial review as an evaluative exercise whereby judges scrutinize administrative decision-making processes and reasons in order to ensure that they are consistent with fundamental values which underpin the legitimacy of administrative law. The formalist approach is typified by 2017 CanLIIDocs 175 elaborate attempts to define and deduce the scope of judicial review as a threshold matter without considering the attributes of a particular administrative decision, whereas the evaluative approach proceeds more directly to assess whether a particular administrative decision has been produced and explained in a fair and reasonable manner which warrants the respect of individuals and other legal officials. I will argue that the current controversy regarding the standard of review analysis stems from the fact that the Supreme Court of Canada in Dunsmuir erected a more elaborate formalist apparatus for determining the standard of review, which has distracted attention from the fundamental values which underwrite the legitimacy of administrative law and the moral purposes of judicial review. Thus, instead of simplifying the law of judicial review, the Dunsmuir framework has produced an esoteric and unproductive debate about how to define and delineate Dunsmuir’s formal concepts and categories instead of elucidating how the legitimacy of administrative decisions are tied to their legality in a procedural and substantive sense.
In Part III, I will examine and critique four proposals for reforming the law of judicial review that surfaced in Wilson v Atomic Energy of Canada Ltd – a case which served as a lightning rod for the standard of review debate in 2016. I will argue that each of these proposals are unlikely to produce a more principled approach to judicial review, because they seek to resolve confusion about the standard of review by either buttressing Dunsmuir’s formalist apparatus or adding
Lauren Wihak, “Whither the correctness standard of review? Dunsmuir, six years later’ (2014) 27 Can J Admin L & Prac 173.
14 Wilson, supra note 7 at paras 70–73 per McLachlin CJ, Karakatsanis, Wagner, Gascon, and Cromwell JJ, concurring.
112 UNBLJ RD UN-B [VOL/TOME 68 new refinements to it instead of honing an evaluative methodology for assessing the legality of administrative decisions across a broad range of regulatory contexts.
Finally, in Part IV I will conclude by offering a tentative proposal about how we might move beyond the perennial fascination with the standard of review analysis by explaining how judicial review can sustain the procedural and substantive legality of administrative decisions without resorting to categorical or conceptual claims regarding the scope of judicial review. I will argue that such an approach can be gleaned by revisiting the normative underpinnings of L’Heureux-Dubé J’s landmark opinion in Baker v Canada in order to better understand why administrative decisions which are fair and reasonable merit judicial respect.
II. A Brief History of Judicial Review
Complaints about the erratic nature of judicial review are hardly new. In fact, the common criticism that the law of judicial review is radically incoherent has not 15 changed much since DM Gordon’s scathing critique of jurisdictional error in 1929. 2017 CanLIIDocs 175 In undertaking what he likened to a Herculean labour to “ventilate one of the worst corners of the Augean stable,” Gordon remarked that “[a]nything like a serious examination at large of the case law on jurisdiction must convince an open-minded inquirer that there is virtually no proposition so preposterous that some show of authority to support it cannot be found.”16 In this respect, Gordon strikes the same note as David Stratas, Canada’s leading administrative law judge, who noted in 2016 that “[d]octrinal incoherence and inconsistency plague the Canadian law of judicial review. This must stop.”17
The doctrinal incoherence that Gordon, Stratas, and many others have criticized has a long pedigree,18 but the basic problem is that judges struggle to articulate a legal framework for judicial review which avoids both judicial quiescence and judicial overreach, and is equally applicable to decisions rendered by arbitrators,19 labour adjudicators,20 immigration officials,21 copyright boards,22
15 DM Gordon, “The Relation of Facts to Jurisdiction” (1929) 45 Law Q Review 459.
16 Ibid at 459–460.
17 Stratas, supra note 1 at 1.
18 See e.g. John Willis, “Three Approaches to Administrative Law: The Judicial, the Conceptual, and the Functional” (1935) 1 UTLJ 53; Bora Laskin, “Certiorari to Labour Boards: The Apparent Futility of Privative Clauses” (1952) 30 Can Bar Rev 986; Paul Weiler, “The ‘Slippery Slope’ of Judicial Intervention: The Supreme Court and Canadian Labour Relations 1950-1970” (1971) 9 Osgoode Hall LJ 1; David Mullan, “The Re-emergence of Jurisdictional Error” (1985) 14 Admin LR 326; H Wade McLaughlin, “Judicial Review of Administrative Interpretations of Law: How Much Formalism Can We Reasonably Bear?” (1986) 36 UTLJ 343; David Mullan, “The Supreme Court of Canada and Jurisdictional Error: Compromising New Brunswick Liquor” (1987) 1 Can J Admin L & Prac 71; David Mullan, “Jurisdictional Error Yet Again: The Imprecise Limits of the Jurisdiction-Limiting Canada (Attorney General) v PSAC’ (1993) 11 Admin LR (2d) 117.
19 Commission scolaire de Laval, supra note 7.
2017] RENOVATING JUDICIAL REVIEW 113 human rights tribunals,23 law societies,24 and countless other administrative agencies.25 The history of how judges have wrestled with this problem is worth recounting briefly, because it helps explain how we got into this standard of review mess in the first place, and to gauge whether we are any closer to cleaning it up.
The doctrine of jurisdictional error, which continues to lurk within the Dunsmuir framework, grounded the dominant approach to judicial review until 1978. Its underlying constitutional premise was that the different branches of government perform analytically distinct roles: legislatures have the exclusive power to create law, the judiciary has exclusive power to interpret law, and administrative officials wield residual discretionary power to render findings of fact and implement laws created by the legislature and interpreted by the judiciary.26 The important point is that this conception of the separation of powers reserved questions of law exclusively to superior courts, while administrative officials merely retained a discretionary power to apply the law to a particular set of facts. When defined in this way, the doctrine of jurisdictional error seemed to be based upon an apolitical (and therefore constitutionally acceptable) account of judicial review, whereby judges determine “jurisdictional” questions of law, but nevertheless allow administrative 2017 CanLIIDocs 175 officials to determine “non-jurisdictional” questions, usually relating to findings of fact and public policy questions left undetermined by statute.27 But despite projecting a sense of certainty and predictability regarding the conduct of judicial review, the doctrine was formalistic, reductive, controversial, and radically incoherent in practice.
It was formalistic, because it was designed to articulate and maintain abstract conceptual distinctions instead of ensuring that governmental decisions (regardless of their abstract classification) were consistent with fundamental legal values.28 Thus, instead of explaining how judicial review might vindicate the dignity
20 Wilson, supra note 7.
21 Kanthasamy, supra note 7; Huruglica, supra note 7.
22 Canadian Broadcasting Corp, supra note 7.
23 Canada (Human Rights Commission), supra note 7; Stewart, supra note 7.
24 Trinity Western University v Law Society of Upper Canada, supra note 7; Trinity Western University v Law Society of British Columbia, supra note 7; Nova Scotia Barristers’ Society v Trinity Western University, supra note 7.
25 Sossin, supra note 9.
26 Martin Loughlin, “Procedural Fairness: A Study of the Crisis in Administrative Law Theory” (1978) 28 UTLJ 215; David Dyzenhaus, “Formalism’s Hollow Victory” (2002) NZLR 525 [Dyzenhaus, “Formalism’s Hollow Victory”]. For an historical account of how a formalist conception of the separation of powers informed the doctrine of jurisdictional error, see Matthew Lewans, Administrative Law and Judicial Deference (Oxford, Hart Publishing, 2016) chapters 3 and 5.
27 See, e.g. Rex v Nat Bell Liquors, Ltd, [1922] 2 AC 128 (JCPC).
28 Loughlin, supra note 26; Dyzenhaus, “Formalism’s Hollow Victory” supra note 26 at 527; David Dyzenhaus, “Constituting the Rule of Law: Fundamental Values in Administrative Law” (2002) 27
114 UNBLJ RD UN-B [VOL/TOME 68 of individuals by ensuring that administrative decisions are rendered fairly and provide a public, intelligible, and rationally acceptable legal justification, legal analysis was preoccupied with the relatively esoteric exercise of slotting administrative action into conceptual categories that had no intrinsic connection to principles of political morality. If, for example, a reviewing court deemed that an administrative decision involved a “jurisdictional” question or the exercise of “judicial” or “quasi-judicial” power, the decision would be quashed if a trial-type process had not been observed even in the face of privative clause.29 Conversely, if a judge deemed that an administrative decision involved the exercise of “administrative” power, the decision would be upheld regardless of whether there had been a hearing.30 The upshot was an all-or-nothing approach to judicial review, which hinged on an arbitrary, formal categorization of administrative action rather than ensuring that administrative decisions reflected core values which underpin the legitimacy of administrative law.31
It was reductive, because it distorted the reality that legislatures regularly delegate authority to administrative officials in order to decide questions of law through a variety of rule-making, adjudicative, and policy development powers in 2017 CanLIIDocs 175 order to realize broadly-worded statutory objectives. While a legislative assembly might be capable of forging sufficient political consensus to enact open-textured objectives and enabling provisions as general guides for administrative action, these utterances rarely provide sufficiently detailed edicts to determine the array of interpretive disputes that typically give rise to judicial review proceedings. For example, legislative assemblies frequently recognize that they cannot foresee subsequent socio-economic or technological developments, and conclude that administrative decision-making processes provide a more accessible, fair, and efficient forum for individuals to tender evidence and legal arguments about how the law ought to be interpreted and applied to a specific set of facts. These institutional reasons explain one of the most important constitutional phenomena of the twentieth century: the rapid and inexorable development of the modern administrative state
Queen’s LJ 445 at 450; Paul Daly, “The Unfortunate Triumph of Form over Substance in Canadian Administrative Law” (2012) 50 Osgoode Hall LJ 317 at 320 [Daly, “Unfortunate Triumph”].
29 See, e.g. Toronto Newspaper Guild v Globe Printing, [1953] 2 SCR 18, [1953] 3 DLR 561; Alliance des Professeurs Cotholiques de Montréal v Quebec Labour Relations Board, [1953] 2 SCR 140, [1953] 4 DLR 161; Saltfleet (Township) Board of Health v Knapman, [1956] SCR 877, 6 DLR (2d) 81; Jarvis v Associated Medical Services Inc, [1964] SCR 497, 44 DLR (2d) 407.
30 See e.g. R v Legislative Committee of the Church Assembly, [1928] 1 KB 411; British Columbia (Labour Relations Board) v Traders Services Ltd, [1958] SCR 672, 15 DLR (2d) 305; Calgary Power v Copithorne, [1959] SCR 24, 16 DLR (2d) 241; Kinnaird v British Columbia (Workmen’s Compensation Board), [1963] SCR 239, 38 DLR (2d) 245; Québec (Commission des Relations Ouvrières) v Burlington Mills Hosiery Co of Canada, [1964] SCR 342, 45 DLR (2d) 730; Galloway Lumber Co v British Columbia (Labour Relations Board), [1965] SCR 222, 48 DLR (2d) 587; Bakery and Confectionary Workers International Union of America Local No 468 v White Lunch Ltd, [1966] SCR 282, 56 DLR (2d) 193; Quebec (Commission des Relations de Travail) v Canadian Ingersoll-Rand Co Ltd, [1968] SCR 695, 1 DLR (2d) 417; Noranda Mines Ltd v Saskatchewan, [1969] SCR 898, 7 DLR (3d) 1.
31 Dyzenhaus & Fox-Decent, supra note 4; Matthew Lewans, Administrative Law and Judicial Deference (Oxford, Hart Publishing, 2016) ch 6.
2017] RENOVATING JUDICIAL REVIEW 115 throughout the western industrialized world, which confounded reasoning premised upon a formalistic conception of the separation of powers.
Consequently, the doctrine also proved to be controversial, because by assuming that the judiciary has exclusive power to determine jurisdictional questions of law, reviewing courts frequently reversed considered decisions rendered by administrative officials who had been empowered through the democratic process to decide those legal questions.32 For example, despite granting broad authority to administrative officials to interpret and apply labour and human rights codes, judges often ran roughshod over the decisions of labour boards33 and human rights tribunals34 even to the point of flouting statutory privative clauses. This pattern of judicial overreach raised serious concerns regarding the constitutional legitimacy of judicial review and the relative competency of the judiciary to supervise increasingly specialized regulatory regimes. One of Canada’s leading constitutional law scholars in the 20th century, Professor Bora Laskin, questioned the manner in which judges frequently circumvented privative clauses by arguing that “judicial persistence in exercising a reviewing power involves an arrogation of authority on the basis of constitutional principle (and there is no such principle) or on the basis of some ‘elite’ 2017 CanLIIDocs 175 theory of knowing what is best for all concerned.”35 His point was that by gainsaying administrative decisions, judges frequently undermined the operation of institutions that play a vital role in modern democratic governance.
Finally as Gordon, Laskin, John Willis, and many others were keen to point out, the doctrine was radically incoherent, because judges vacillated between a narrow conception of jurisdictional error (which focused on whether the express wording of the statute authorized an official to embark on an inquiry) and a broader conception (which interpreted enabling legislation against a background of common law principles like the principles of natural justice).36 However, much of this confusion was put to rest in cases like Anisminic Ltd v Foreign Compensation
32 Willis, supra note 18; Harry Arthurs, “Protection Against Judicial Review” (1983) 43 R du B 277; Jeremy Waldron, “Authority for Officials” in Lukas Meyer, Stanley Paulson and Thomas Pogge (eds), Rights, Culture, and the Law: Themes from the Legal Philosophy of Joseph Raz (Oxford, Oxford University Press, 2003) 45.
33 See, e.g. In re Ontario Labour Relations Board: Toronto Newspaper Guild, Local 87 v Globe Printing Company, [1953] 2 SCR 18, [1953] 3 DLR 561. For academic critique of this case see Laskin, supra note 18; Weiler, supra note 18.
34 See, e.g. Bell v Ontario Human Rights Commission, [1971] SCR 756, 18 DLR (3d) 1; Gay Alliance Toward Equality v Vancouver Sun, [1979] 2 SCR 435, 10 BCLR 257; Canada (AG) v Mossop, [1993] 1 SCR 554, 100 DLR (4th) 658. For academic critique of correctness review in the human rights context, see Allison Harvison Young, “Keeping the Courts at Bay: The Canadian Human Rights Commission and its Counterparts in Britain and Northern Ireland: Some Comparative Lessons” (1993) 43 UTLJ 65; Allison Harvison Young, “Human Rights Tribunals and the Supreme Court of Canada: Reformulating Deference” (1993) 13 Admin LR (2d) 206.
35 Laskin, supra note 18 at 991.
36 See, e.g. the conflicting approaches outlined in the cases cited supra note 29 and note 30, and Weiler, supra note 18.
116 UNBLJ RD UN-B [VOL/TOME 68
Commission, a case in which Lord Reid cast a very broad net for jurisdictional error,37 prompting HWR Wade to conclude that the House of Lords had effectively stripped administrative officials of their statutory authority to decide questions of law.38 Nevertheless, the Supreme Court of Canada quickly followed suit, collapsed the distinction between jurisdictional and non-jurisdictional errors in Metropolitan Life Insurance Co v International Union of Operating Engineers, Local 796, and held that judges should generally review questions of law on a correctness basis.39
Beginning in 1978, the Supreme Court of Canada began constructing an alternative, evaluative framework for judicial review, which rejected the premise that only judges were entitled to determine questions of law. It recognized that administrative officials share responsibility for interpreting the law, but sought to preserve administrative fidelity to the rule of law by bolstering and extending administrative officials’ common law duty of procedural fairness and demonstrating how that duty entails substantive, but deferential, constraints on administrative action. That alternative approach was built upon three interlocking propositions. The first proposition is that administrative officials owe a general duty of fairness at common law that is triggered and shaped by contextual considerations (e.g. the 2017 CanLIIDocs 175 practical impact of the decision on an individual’s rights, interests, property, privileges, or liberties) as opposed to being contingent strictly upon facts or judicial suppositions about legislative intent.40 This general duty of procedural fairness ensures that individuals affected by an administrative decision are entitled to participate meaningfully in administrative decision-making processes, and to be heard by an impartial decision-maker even when these procedural entitlements are not spelled out explicitly in enabling legislation. Nevertheless, the duty of fairness does not entail that administrative processes must conform to trial-type procedures; it requires fair treatment, but recognizes that an administrative decision-maker “is free, within reason, to determine its own procedures, which will vary with the nature of the inquiry and the circumstances of the case.”41
The second proposition is that judges should refrain from casting a broad net for jurisdictional errors, and instead defer to administrative interpretations of law provided they are “rationally supported by the relevant legislation.”42 Because administrative officials are empowered through the democratic process to decide
37 Anisminic Ltd v Foreign Compensation Commission, [1969] 2 AC 147 at 170 (HL) [Anisminic].
38 HWR Wade, Administrative Law, 3rd ed (Oxford: Clarendon Press, 1971) at 97.
39 Metropolitan Life Insurance Co v International Union of Operating Engineers, Local 796, [1970] SCR 425, 11 DLR (3d) 336.
40 Nicholson, supra note 3 at 324–28; Martineau v Matsqui Disciplinary Board, [1980] 1 SCR 602 at 622– 23, 106 DLR (3d) 385.
41 Kane v Board of Governors of the University of British Columbia, [1980] 1 SCR 1105 at 1112, 18 BCLR 124.
42 Canadian Union of Public Employees, Local 963 v New Brunswick Liquor Corporation, [1979] 2 SCR 227 at 237, 25 NBR (2d) 237 [CUPE].
2017] RENOVATING JUDICIAL REVIEW 117 certain legal questions, and are more proficient in answering those questions than generalist judges, courts should eschew the Anisminic proposition that judges are entitled to review all questions of law on a correctness standard. Nevertheless, this did not mean that judges should defer blindly to administrative decisions; rather, it meant that judges engaged in a form of substantive quality control by verifying that an administrative decision-maker responds to the evidence, the parties’ submissions, and provides a conclusion which is rationally defensible in light of relevant legal principles.
Third, while legislatures are constitutionally entitled to delegate authority to decide questions of law to administrative officials, they cannot immunize administrative decisions from public scrutiny because individuals have a constitutional right to judicial review.43 Thus, the parties who are bound by an administrative decision are entitled to challenge the legality of that decision before an independent judiciary to ensure that it has been rendered in a fair and legally justified manner.
When viewed as a package, this evaluative framework rejects the premise 2017 CanLIIDocs 175 that administrative officials are mere ciphers who implement laws which are created by legislatures and whose content are determined by judges. In its place, it recognizes that administrative officials are responsible for developing a body of administrative law that sits alongside legislation and the common law in a democratic legal order, and which merits the respect of both citizens and superior courts. Nevertheless, it seeks to uphold the rule of law by ensuring that administrative processes and decisions are articulated fairly (i.e. in a manner which responds to the evidence and arguments tendered by the parties) and reasonably (i.e. in a manner which is rationally justifiable in light of relevant legal principles drawn from legislation, regulations, the common law, etc.).44 Reduced to its essence, this understanding of judicial review holds that if an administrative decision is rendered in a fair and legally justified manner, other legal officials (including judges) should respect that decision instead of substituting their own determination of law in its place.
However, instead of making a clean break with formalistic reasoning, the Supreme Court pursued a more subtle transition, which discouraged judges from casting a broad net for jurisdictional errors and instead emphasized contextual factors like the practical impact of the decision on the individual, the underlying purposes or objectives of the regulatory scheme, the democratic legitimacy of administrative institutions, and the relative expertise of administrative decision-makers.45 As a result, the complexity and confusion associated with judicial review proliferated,
43 Crevier, supra note 2 at 236.
44 For a similar, theoretical account of the rule of law, see Lon Fuller, “The Forms and Limits of Adjudication” (1978) 92 Harvard L Rev 353.
45 See, eg UES, Local 298 v Bibeault, [1988] 2 SCR 1048 at 1088, 35 Admin LR 153; Knight, supra note 3 at 669–77.
118 UNBLJ RD UN-B [VOL/TOME 68 because even while paying lip-service to the idea that judges ought to review administrative decisions according to a deferential standard, they clung to formalist concepts and categories by continuing to review administrative decisions on a correctness standard if they deemed the issue in dispute to be “jurisdictional” or took the absence of an express privative clause as a green light for de novo review. And even when the Court did apply the reasonableness standard, it often resembled what David Mullan calls “disguised correctness review,”46 because the Court would first determine how it would have decided the matter (all things considered) and then proceed to consider whether an administrative decision was sufficiently proximate to the court’s to warrant judicial restraint.47
This state of affairs often produced conflicted or confused judicial opinions. In decisions like Pezim v British Columbia (Superintendent of Brokers)48 and Canada (Director of Investigation and Research) v Southam Inc,49 the Court held that judges should review administrative decisions according to a deferential standard of reasonableness even when the enabling legislation provides a statutory right of appeal, because administrative officials possess theoretical, experiential, or procedural forms of expertise. In this respect, the Court seemed to be extending the 2017 CanLIIDocs 175 CUPE rationale for judicial deference regarding administrative interpretations of law.
However, instead of simply stating that the justification for judicial deference persists in cases where an administrative decision is subject to a statutory right of appeal, the Court went on to say there was a “spectrum” of different degrees of judicial deference that incorporates the problematic distinctions between jurisdictional/non-jurisdictional issues and questions of law/fact which are hallmarks of formalistic analysis. In Pezim, Iacobucci J suggested that the highest degree of deference should be reserved for “cases where a tribunal protected by a true privative clause, is deciding a matter within its jurisdiction and where there is no statutory right of appeal,” and the lowest degree of deference would apply in cases concerning “the interpretation of a provision limiting the tribunal’s jurisdiction (jurisdictional error) or where there is a statutory right of appeal which allows the reviewing court to substitute its opinion for that of the tribunal and where the tribunal has no greater expertise on the issue in question, as for example in the area of human rights.”50 In Southam, he returned to this point in an attempt to clarify the analytical distinction between questions of law, fact, and mixed law and fact.51 While he acknowledged
46 David Mullan, “Unresolved Issues on Standard of Review in Canadian Judicial Review of Administrative Action—The Top Fifteen!” (2013) 42 Adv Q 1 at 76 [Mullan, “Unresolved Issues”].
47 See, e.g. CAIMAW v Paccar of Canada Ltd, [1989] 2 SCR 983 at 1017-20, 40 BCLR (2d) 1 per Sopinka J, concurring [Paccar].
48 Pezim v British Columbia (Superintendent of Brokers), [1994] 2 SCR 557, 92 BCLR (2d) 145 [Pezim].
49 Canada (Director of Investigation and Research) v Southam Inc, [1997] 1 SCR 748, 144 DLR (4th) 1 [Southam].
50 Pezim, supra note 48 at 590.
2017] RENOVATING JUDICIAL REVIEW 119 that this distinction might be problematic,52 his opinion provided fuel for the formalist notion that judges retain exclusive authority to determine general questions of law while the authority of administrative decisions was confined to findings of fact and matters of law application. And when the Court decided Dr Q v College of Physicians and Surgeons of British Columbia, McLachlin CJ declared that administrative decisions concerning “pure” questions of fact “will militate in favour of showing more deference towards the tribunal’s decision,” while decisions involving “pure” questions of law would attract “a more searching review” particularly in cases “where the decision will be one of general importance or great precedential value.”53
Thus, even while acknowledging that judges should defer to administrative officials’ interpretation of their enabling statute, the Court suggested that judges should continue to review certain categories of administrative decisions on a correctness basis (e.g. “jurisdictional” questions, general questions of law, questions of law outside the expertise of the decision-maker). This discourse culminated in decisions like Pushpanathan v Canada (Minister of Employment & Immigration), in which Bastarache J outlined a multi-factor, pragmatic and functional framework for 2017 CanLIIDocs 175 identifying the appropriate standard of review separate and apart from arguments relating to the process or substantive reasoning employed by the decision-maker.54 That decision instructed judges to consider an array of contextual factors – the presence or absence of a privative clause, the expertise of the decision-maker, the purpose of the legislation as a whole as well as any provision in particular, and the nature of the issue in dispute. But in the process, Bastarache J conceived the pragmatic and functional framework primarily as an attempt to categorize the decision under review – with particular emphasis on determining whether an administrative decision involved questions of law the Court deemed to fall outside the tribunal’s expertise. At least in this respect, Pushpanathan seems to facilitate analysis quite similar to the doctrine of jurisdictional error – the difference is that “the language an approach of the ‘preliminary,’ ‘collateral’ or ‘jurisdictional’ question has been replaced by this pragmatic and functional approach.”55
By contrast, L’Heureux-Dubé J’s opinion in Baker v Canada (Minister of Citizenship) adopts an evaluative approach to judicial review, which focuses on whether a particular administrative process and decision comports with fundamental values instead of attempting to categorize the nature of the administrative decision in
51 Southam, supra note 49 at para 36.
52 Ibid at paras 35, 37.
53 Dr Q v College of Physicians and Surgeons of British Columbia, 2003 SCC 19 at para 34, [2003] 1 SCR 226.
54 Pushpanathan v Canada (Minister of Employment & Immigration), [1998] 1 SCR 982 at paras 29–38, 160 DLR (4th) 193 [Pushpanathan].
55 Ibid at para 28.
120 UNBLJ RD UN-B [VOL/TOME 68 the abstract.56 Thus, she declared that it was “inaccurate to speak of a rigid dichotomy of ‘discretionary’ or ‘non-discretionary’ decisions” which entail different standards of review.57 Instead, she focused on whether the particular decision being reviewed (regardless of its categorization) was fair and reasonably justified.58 Much like Laskin CJ’s opinion in Nicholson, L’Heureux-Dubé J articulated a contextualized duty of fairness, whose purpose was “to ensure that administrative decisions are made using a fair and open procedure, appropriate to the decision being made and its statutory, institutional, and social context, with an opportunity for those affected by the decision to put forward their views and evidence fully and have them considered by the decision-maker.”59 However, she expanded upon that idea by explaining how the duty of fairness impacts the legal substance of an administrative decision. In addition to ensuring that an individual had been informed of the case against him and heard by an impartial decision-maker, the duty of fairness required administrative officials to provide reasons “where the decision has important significance for the individual.”60 Furthermore, she noted that those reasons must disclose a reasonable basis for the decision – one which demonstrates that the decision-maker was “alert, alive and sensitive” to relevant legal principles, including
“the boundaries imposed in the statute, the principles of the rule of law, the 2017 CanLIIDocs 175 principles of administrative law, the fundamental values of Canadian society, and the principles of the Charter.”61
But instead of following L’Heureux-Dubé J’s lead in subsequent years, the Supreme Court focused its attention instead on defining different categories of administrative decisions so as to enable judges to identify the standard of review more efficiently.62 In his concurring opinion in Chamberlain v Surrey School District No 36, LeBel J made the insightful observation that a “formulaic” recitation of the pragmatic and functional framework frequently diverts attention “from the real issue of legality to an unnecessary exploration of tangential questions.”63 However, instead of employing the evaluative approach to judicial review set out in Baker, LeBel J sought rehabilitate the Pushpanathan framework by leaning more heavily upon formalistic reasoning. Thus, the Court soon began developing shortcuts around the pragmatic and functional approach by identifying abstract categories of issues that
56 Baker, supra note 3 at para 56.
57 Ibid at para 54.
58 Ibid at paras 51–56.
59 Ibid at para 22.
60 Ibid at para 43.
61 Ibid at paras 56, 75.
62 Interestingly, the Supreme Court has not tinkered with the multi-factor Baker framework for procedural fairness. Nevertheless, some have advocated for a unified framework for judicial review that would encompass both procedural and substantive grounds for review. See Paul Daly, “Canada’s Bi-Polar Administrative Law: Time for Fusion” (2014) 40 Queen’s LJ 213.
63 Chamberlain v Surrey School District No 36, 2002 SCC 86 at para 202, [2002] 4 SCR 710.
2017] RENOVATING JUDICIAL REVIEW 121 would presumptively attract correctness review: constitutional questions,64 “true” jurisdictional issues,65 and general questions of law which are of central importance to the legal system.66 This trend of heaping additional conceptual distinctions and categories onto the doctrinal structure of judicial review concludes with Dunsmuir v New Brunswick, a case in which the author of the Pushpanathan framework (Bastarache J) teamed up with its main critic (LeBel J) to coauthor an opinion with the aim of developing “a principled framework that is more coherent and workable.”67 That revised framework can be summarized as follows:
(1) the Court abolished the patent unreasonableness standard of review (leaving only the reasonableness and correctness review);68 (2) the Court held that it is unnecessary for a reviewing court to apply the contextual standard of review analysis when the standard of review has been settled by precedent;69 (3) in cases where the standard of review is not settled by precedent, the Court held that reviewing courts should apply the standard of review analysis by
considering contextual factors such as the presence or absence of a privative 2017 CanLIIDocs 175 clause, the purpose of the tribunal as determined by reference to the enabling legislation, the nature of the question at issue, and administrative expertise;70 (4) the Court attempted to simplify the process for identifying the appropriate standard of review by categorizing different types of issues which will presumptively attract either correctness or reasonableness review. The reasonableness standard applies presumptively to decisions protected by a privative clause, findings of fact, interpretations of an administrator’s enabling statute, the exercise of discretionary power, questions of public policy, and mixed questions of law and fact;71 whereas the correctness standard applies presumptively to constitutional questions, “true” jurisdictional issues, questions of law which are of general importance to
64 Nova Scotia (Workers’ Compensation Board) v Martin; Nova Scotia (Workers’ Compensation Board) v Laseur, 2003 SCC 54 at para 31, [2003] 2 SCR 504; Multani v Commission scolaire Marguerite- Bourgeoys, 2006 SCC 6 at para 20, [2006] 1 SCR 256.
65 United Taxi Drivers’s Fellowship of Southern Alberta v Calgary (City), 2004 SCC 19 at para 5, [2004] 1 SCR 485; ATCO Gas & Pipelines Ltd v Alberta (Energy Utilities Board), 2006 SCC 4 at para 87, [2006] 1 SCR 140.
66 Toronto (City) v CUPE, Local 79, 2003 SCC 63 at para 62, [2003] 3 SCR 77.
67 Dunsmuir, supra note 5 at para 32.
68 Ibid at paras 41–2.
69 Ibid at paras 57, 62.
70 Ibid at paras 55, 64.
71 Ibid at paras 52–4.
122 UNBLJ RD UN-B [VOL/TOME 68
the Canadian legal system and outside the expertise of the decision-maker, and jurisdictional boundaries between competing administrative agencies;72 (5) finally, the Court attempted to define and distinguish the reasonableness and correctness standards of review.73 Reasonableness review is “concerned mostly with the existence of justification, transparency and intelligibility within the decision-making process” so as to ensure that “the decision falls within a range of possible, acceptable outcomes which are defensible in respect of the facts and law.”74 By contrast, correctness review entitles a court to “undertake its own analysis of the question” without regard for the rationale offered by an administrative decision-maker.75
While the elimination of the patent unreasonableness standard has simplified judicial review to a point, it’s fair to say that the other reforms have not panned out as expected.76 There remains considerable doubt as to whether reliance on precedent will simplify the standard of review analysis because pre-Dunsmuir authorities have lost most of their luster, and any judge or counsel worth his or her salt can distinguish the nuances of a present case from past precedent.77
2017 CanLIIDocs 175 But the main problem with the Dunsmuir framework derives from its attempt to simplify judicial review by setting out presumptive categories for reasonableness and correctness review which overshadow the normative purposes of judicial review. The problem is not just that the boundaries of these categories are unclear, under-inclusive, or over-inclusive – it’s that they are coincident as opposed to being mutually exclusive.78 As the history of jurisdictional error amply demonstrates, judges are frequently drawn to correctness review or a disguised form of it when an administrative decision-maker is interpreting her home statute.79 But
72 Ibid at paras 57–61.
73 Ibid at paras 44–50.
74 Ibid at para 47.
75 Ibid.
76 Given the parameters of this paper, I cannot offer an extended critique of Dunsmuir here. For more thorough discussion of the Dunsmuir case, see Daly, “Unfortunate Triumph”, supra note 28; Paul Daly, “Dunsmuir’s Flaws Exposed: Recent Decisions on Standard of Review” (2012) 58 McGill LJ 483 [Daly, “Dunsmuir’s Flaws Exposed”]; Mullan, “Unresolved Issues”, supra note 46.
77 Mullan, “Unresolved Issues”, supra note 46 at 4–6; Stratas, supra note 1 at 3–4.
78 Daly, “Unfortunate Triumph”, supra note 28 at 330; Daly, “Dunsmuir’s Flaws Exposed”, supra note 76 at 488.
79 See, e.g. Cromwell J’s protestation in Alberta (Information and Privacy Commissioner) v Alberta Teachers’ Association, 2011 SCC 61 at para 98 [2011] 3 SCR 654 [Alberta Teachers’ Assoiation]: that “there are legal questions in ‘home’ statutes whose resolution the legislature did not intend to leave to the tribunal; indeed, it is hard to imagine where else the limits of an tribunal’s delegated power are more likely to be set out.” Some recent cases which demonstrate this line of thinking in action include Edmonton East, supra note 7; Stewart, supra note 7; Nova Scotia Barristers’ Society v Trinity Western University, supra note 7; Wilson, supra note 7 per Côté and Brown JJ dissenting; and Edmonton East, supra note 7 per Côté and Brown JJ dissenting.
2017] RENOVATING JUDICIAL REVIEW 123 similar observations can be made with respect to the other categories. For example, the resolution of constitutional questions often hinges upon findings of fact, mixed questions of law and fact, a deep understanding of the enabling legislation, or the broader regulatory context.80 Put simply, if one were to construct a Venn diagram of Dunsmuir’s presumptive categories there would be substantial areas of overlap, which explains why judges frequently disagree about whether a given case involves a general question of law outside an administrator’s expertise which entitles the court to determine the answer or a question concerning the interpretation of an administrator’s home statute which warrants judicial deference. The end product, unsurprisingly, is protracted litigation and judicial handwringing over whether a case can be slotted into a particular category in order to determine the appropriate standard of review. But more importantly, that protracted hand-wringing detracts from the more basic, evaluative task of verifying whether an administrative decision is fair and reasonably justified.
This might explain why the Supreme Court has continued to tinker with the Dunsmuir standard of review analysis. While the Court has restrained itself from rewriting the standard of review analysis, it initially signaled a clear preference for 2017 CanLIIDocs 175 reasonableness review in all cases concerning the interpretation of an administrator’s home statute,81 and extended the more deferential standard of review to cases where an administrative decision involves Charter values82 and general questions of law.83 Moreover, while the Court did not abolish the concept of jurisdictional error, it has repeatedly gone out of its way to say that it may be time “to reconsider whether, for the purposes of judicial review, the category of true questions of jurisdiction exists and is necessary to identifying the standard of review.”84 This line of post-Dunsmuir cases has led at least one administrative lawyer to worry that the rule of law might be compromised insofar as judges might defer to administrative decisions on jurisdictional issues, constitutional issues, and general questions of law.85 But if one subscribes to a protestant conception of the rule of law which recognizes the
80 See, e.g. Doré v Barreau du Québec, 2012 SCC 12, [2012] 1 SCR 395 [Doré]; Loyola High School v Quebec (AG), 2015 SCC 12, [2015] 1 SCR 613 [Loyola High School]. See also Geneviève Cartier, “The Baker Effect: A New Interface Between the Canadian Charter of Rights and Freedoms and Administrative Law—The Case of Discretion” in David Dyzenhaus, ed, The Unity of Public Law (Oxford: Hart Publishing, 2004) 61.
81 See, e.g. Alberta Teachers Association, supra note 79; Canadian National Railway v Canada (AG), 2014 SCC 40, [2014] 2 SCR 135; Tervita Corp v Canada (Commissioner of Competition), 2015 SCC 3, [2015] 1 SCR 161; Ontario Energy Board v Ontario Power Generation Inc, 2015 SCC 44, [2015] 3 SCR 147; Movement laïque Québécois v Saguenay (City), 2015 SCC 16, [2015] 2 SCR 3.
82 See, e.g. Doré, supra note 80 at para 45; Loyola High School, supra note 80 at paras 37–40.
83 Nor-Man Regional Health Authority Inc v Manitoba Association of Health Care Professionals, 2011 SCC 59 at para 4, [2011] 3 SCR 616.
84 Alberta Teachers Association, supra note 79 at para 34. See also ATCO Gas and Pipelines Ltd v Alberta (Utilities Commission), 2015 SCC 45 at para 27, [2015] 3 SCR 219.
85 Lauren Wihak, “Wither the Correctness Standard of Review? Dunsmuir, Six Years Later” (2014) 27 Can J Admin L & Prac 173 at 175.
124 UNBLJ RD UN-B [VOL/TOME 68 potential legitimacy of administrative law (like the one articulated in the Baker paradigm) this concern appears to conflate the rule of law with maintaining the judiciary’s interpretive monopoly over questions of law.
III. Recent Proposals to Renovate the Standard of Review
If the cases from the past year are any indication, discontent with the Dunsmuir framework has reached another tipping point, when the Supreme Court seems primed to change the standard of review analysis yet again.86 The most telling piece of evidence is Wilson v Atomic Energy of Canada Ltd, a case which generated no less than four different proposals to change the standard of review analysis. But unfortunately, many of these proposals do not turn to L’Heureux-Dubé J’s opinion in Baker for inspiration; instead, they double-down on developing a set of more intricate conceptual definitions to resolve the current malaise.
Like Dunsmuir, the Wilson case concerns an adjudicator’s decision to inquire about an employer’s reasons for terminating a non-unionized public sector 2017 CanLIIDocs 175 employee. Besides rewriting the standard of review analysis in Dunsmuir, the Supreme Court held that the adjudicator’s decision to reinstate David Dunsmuir was unreasonable because the New Brunswick Department of Justice was legally entitled to dismiss him without holding a performance review. In doing so, the Court overturned thirty years of authority stretching back to Nicholson and Knight v Indian Head School Division No 19 by holding that “in the specific context of dismissal from public employment, disputes should be viewed through the lens of contract law rather than public law.”87 When subsequent litigants invited the Court to retrench the general duty of fairness in other regulatory contexts the Court demurred, stating that the Dunsmuir exception to the duty of fairness was “rather narrow.”88 Nevertheless, it raised the possibility that the private law of contract might further erode or exclude public law principles of fair process and reasoned justification in other cases involving the dismissal of non-unionized public employees.
Therefore, when Atomic Energy of Canada Ltd (‘AECL’) terminated Joseph Wilson one year after the Supreme Court issued its decision in Dunsmuir, it set out to explore how far the common law of contract might limit or exclude administrative law protections against unjust dismissal under the Canada Labour Code.89 Wilson had been employed by AECL for approximately four and a half years – initially as a Senior Buyer/Order Administrator and later as a Procurement Supervisor. Despite having a clean disciplinary record, Wilson was dismissed summarily on 19 November 2009, at which time AECL offered him a severance
86 Stratas, supra note 1 at 2.
87 Dunsmuir, supra note 5 at para 82.
88 Canada (AG) v Mavi, 2011 SCC 30 at para 51, [2011] 2 SCR 504.
89 Canada Labour Code, RSC 1985, c L-2 [Labour Code].
2017] RENOVATING JUDICIAL REVIEW 125 package equivalent to 6 months’ salary. When Wilson filed a complaint for unjust dismissal under s. 240 of the Labour Code, it triggered the employer’s duty under s. 242 to explain the reasons for its decision.90 Throughout the proceedings, Wilson maintained that his dismissal was, in fact, a form of reprisal to punish him for reporting corrupt procurement practices within his department.91 However, instead of disputing Wilson’s factual assertions or explaining its decision, AECL responded by letter stating simply that Wilson had been “terminated on a non-cause basis and was provided a generous severance package that exceeded the statutory requirements” under ss. 230 and 235 of the Labour Code.92 Unsatisfied with his employer’s response, Wilson requested the appointment of an adjudicator to determine whether his dismissal was “unjust” within the meaning of s 242.
At the outset of the proceedings, the parties posed two questions for Stanley Schiff, a retired University of Toronto law professor with over forty years’ experience as an adjudicator under the Canada Labour Code.93 The first question was whether AECL could lawfully terminate Wilson’s employment on a without cause basis; the second was whether a generous severance package might constitute a just dismissal despite the absence of reasons. After considering the first question, 2017 CanLIIDocs 175 Adjudicator Schiff held that AECL could not lawfully terminate Wilson on a without cause basis,94 because the Federal Court had declared in Redlon Agencies Ltd v Norgren that an employer “cannot avoid the operation of the unjust dismissal provisions by resorting to the severance payment provisions” in the Code.95 Accordingly, he adjourned the proceeding with an invitation to the parties to negotiate a settlement. However, instead of pursuing negotiations with Wilson’s lawyer, AECL sought judicial review of his interlocutory ruling. Thus began Wilson’s four-year odyssey to the Supreme Court, which ultimately upheld Professor Schiff’s decision and sent the matter back so that another adjudicator could hear the parties’ submissions regarding the appropriate remedy.
90 Wilson, supra note 7 at paras 8–9.
91 Ibid at para 10.
92 Ibid at para 9. Section 230 of the Code provides a notice period or pay in lieu of notice of at least two weeks for employees who have completed three consecutive months of employment; and section 235 states that employees who have completed twelve consecutive months of employment are entitled to severance pay equivalent to the greater amount of two days wages for each year of continuous employment or five days wages.
93 Stanley Schiff is Professor Emeritus of Law, University of Toronto and author of “Labour Arbitration Procedures”, a Draft Study on Judicial Review of Labour Arbitration in Canada (Task Force on Labour Relations, Office of the Privy Council 1968). Professor Schiff began serving as an adjudicator under the Canada Labour Code in 1970, and Wilson would be the final decision he delivered in that capacity. His earliest reported decision is Re United Steelworkers and TMX Watches of Canada Ltd, [1970] CLAD No 73.
94 Wilson, supra note 7 at para 13.
95 Redlon Agencies Ltd v Norgren, 2005 FC 804 at para 39, 139 ACWS (3d) 1018 [Redlon].
126 UNBLJ RD UN-B [VOL/TOME 68
The parties raised two issues on judicial review. The first was whether AECL’s application for judicial review of the adjudicator’s decision was premature; the second was whether the adjudicator’s decision was unreasonable.96 At least at this stage in the proceedings, there was no dispute that the proper standard of review was reasonableness, because both parties had agreed that an adjudicator’s decision regarding the interpretation of the enabling legislation in general (and the unjust dismissal provisions in particular) warranted a degree of judicial deference.
Despite acknowledging that “courts discourage the breaking up of proceedings into discrete parcels and encourage the parties to finish their business in the tribunal below before coming to court,”97 O’Reilly J held that it was nevertheless permissible for AECL to seek judicial review because the adjudicator “had made a final determination on the substance of the matter before him.”98 However, he neglected to mention that s. 243 of the Code contains privative clause, which prohibits courts from entertaining an application “to question, review, prohibit or restrain an adjudicator in any proceedings of the adjudicator under section 242.”99 While a privative clause might not be a conclusive factor when determining the standard of review, it clearly directed the Federal Court to refrain from intervening 2017 CanLIIDocs 175 until the adjudicator had rendered a decision on the appropriate remedy.
On the second issue, both parties agreed that reasonableness was the appropriate standard of review, and O’Reilly J concluded that Professor Schiff’s decision was unreasonable.100 In O’Reilly J’s view, the adjudicator had “unreasonably relied on Redlon for the proposition that employers governed by the CLC must show just cause for all dismissals.”101 Given the Federal Court’s explicit statement in Redlon that an employer “cannot avoid the operation of the unjust dismissal provisions by resorting to the severance payment provisions,” one might have been tempted to conclude the adjudicator’s decision was rationally defensible in light of the reasoning expressed in that case.102 Nevertheless, O’Reilly J asserted that “[a]n employer can dismiss an employee without cause so long as it gives notice or
96 Wilson and Atomic Energy of Canada Ltd, Re, 2013 FC 733 at paras 10, 18, 230 ACWS (3d) 6 [Re Wilson and Atomic Energy].
97 Ibid at para 12.
98 Ibid at para 15.
99 Section 243 of the Labour Code, supra note 89, states: 243(1) Every order of an adjudicator appointed under subsection 242(1) is final and shall not be questioned or reviewed in any court. (2) No order shall be made, process entered or proceeding taken in any court, whether by way of injunction, certiorari, prohibition, quo warranto or otherwise, to question, review, prohibit or restrain an adjudicator in any proceedings of the adjudicator under section 242.
100 Re Wilson and Atomic Energy, supra note 96 at para 40.
101 Ibid at para 26.
102 Redlon, supra note 95 at para 39.
2017] RENOVATING JUDICIAL REVIEW 127 severance pay.”103 While he recognized that an employee retained the right to complain that his termination was unjust under s. 240, he held that a “without cause” dismissal was not ipso facto unjust because such an inference “would fail to take account of the clear remedies provided in ss. 230 and 235 (i.e. notice and severance) for persons dismissed without cause.”104 The upshot of O’Reilly J’s reasoning is that as long as an employer provides sufficient monetary severance accompanied by pro forma “without cause” statement which is devoid of malice, discriminatory animus, or reprisal, they might avoid an unjust dismissal inquiry under s. 242 of the Labour Code. Practically speaking, O’Reilly J’s interpretation of the Labour Code would enable public employers to dismiss employees arbitrarily – even for no reason at all – so long as they provided adequate notice and severance.
A month after the decision, AECL’s lawyer praised O’Reilly J’s decision in the Lawyers Weekly, saying that it would have a “monumental impact on federally regulated employers and their non-unionized staff,” because “arbitrators have uniformly held that the phrase ‘unjust dismissal’ means an employer can only terminate an employee for just cause, such as for reasons of misconduct, 105 incompetence or permanent incapacity.” In the same column, Professor Michael 2017 CanLIIDocs 175 Lynk from the University of Western Ontario (a former labour lawyer and CLC arbitrator), stated that “[t]his is a break with very consistent arbitral and judicial case law that has provided non-unionized workers in the federal sector with broad rights protections and remedies akin to what they would find in a collective agreement.”106 When cast in this light, O’Reilly J’s ruling that the adjudicator’s decision was unreasonable seems ironic: it asserts that the rule of law entitled the court to resolve a stalemate between conflicting lines of administrative law when, in fact, experts in the field of employment law generally agreed that the Labour Code required employers to explain their decisions to dismiss non-unionized employees instead of terminating them summarily.
Despite the fact that both parties and a Federal Court justice all agreed that the Adjudicator’s decision should be reviewed on a reasonableness standard, the case quickly devolved into a protracted debate about the appropriate standard of review. In subsequent legal proceedings, twelve different justices of the Federal Court of Appeal and Supreme Court of Canada went on at length about different proposals to renovate the standard of review analysis, but expended relatively little effort on understanding the Adjudicator’s decision on its own terms. In what follows, I will attempt to separate the different camps in this ongoing debate, and examine the different proposals for reforming the standard of review analysis.
103 Re Wilson and Atomic Energy, supra note 96 at para 35.
104 Ibid at para 37.
105 Christopher Guly, “Court makes ‘game-changing’ decision on employees: ‘No basis’ for contention Labour Code demands dismissals with cause, judge rules” (16 August 2013) 33(14) The Lawyers Weekly.
106 Ibid. See also Reagan Ruslim, “Unjust Dismissal the Canada Labour Code: New Law, Old Statute” (2014) 5 UWO J Legal Stud 3.
128 UNBLJ RD UN-B [VOL/TOME 68
(a) Judicial Review is a Matter of Degree
The Federal Court of Appeal dismissed Wilson’s appeal on both grounds. Writing for the Court, Stratas JA conceded that litigants should only seek judicial review after exhausting the administrative process, but noted that this principle was attenuated in “rare cases where the effect of an interlocutory decision on the applicant is so immediate and drastic that the Court’s concern about the rule of law is aroused.”107 Like O’Reilly J, Stratas JA did not mention that the adjudicator’s decision and continuing jurisdiction regarding the appropriate remedy was protected by a privative clause. Later in his reasons, Stratas JA suggested that the adjudicator had bifurcated the proceedings intentionally so as to enable the Federal Court to settle “once and for all” the legal question about whether an employer could lawfully terminate on a without cause basis.108 But because the adjudicator expressly rested his decision on Redlon, he did not appear to be casting about for another Federal Court ruling on the issue; if anything, he seemed to be encouraging the parties to negotiate a settlement, instead of having him impose one that might not be to either’s liking. In this respect, the adjudicator was likely drawing upon 40 years of experience serving as a labour adjudicator when he adjourned the hearing. 2017 CanLIIDocs 175
After dispensing with Wilson’s first ground of appeal, Stratas JA considered the standard of review, and held (surprisingly) that O’Reilly J had erred in applying the reasonableness standard. After noting that the Court was not bound by the parties’ agreement, Stratas JA called it an “unusual case”109 in which “the current state of adjudicator’s jurisprudence is one of persistent discord.”110 Accordingly, Stratas JA held “the rule of law concern predominates…and warrants this Court intervening to end the discord and determine the legal point once and for all.”111 To support his conclusion, he noted that Dunsmuir contemplated correctness review for questions of law which are of central importance to the legal system and outside the expertise of the decision-maker.112 In the alternative, he asserted that interpretation of the relevant Labour Code provisions required “relatively little specialized labour insight,” meaning that even if he had applied the reasonableness standard the adjudicator was entitled to “only a narrow margin of appreciation” relative to the Court’s interpretation of the Labour Code.113 But regardless of which standard applied, Stratas JA held that O’Reilly J’s interpretation of the Labour Code, which permitted “without cause” dismissals, was correct and therefore declined to intervene.
107 Wilson v Atomic Energy of Canada Ltd, 2015 FCA 17 at para 33, 249 ACWS (3d) 347.
108 Ibid at para 39.
109 Ibid at paras 43–46.
110 Ibid at para 52.
111 Ibid at para 55.
112 Ibid at para 56.
113 Ibid at para 58.
2017] RENOVATING JUDICIAL REVIEW 129
Stratas JA’s opinion weaves together two slightly different strands of argument, both of which merit further analysis. The first is that in cases where a reviewing court concludes that in cases where administrative interpretations of law conflict, it is entitled to engage in correctness review so as to settle the controversy “once and for all.” This line of reasoning would later form the basis for Côté and Brown JJ’s dissenting opinion in the Supreme Court, which I will address at greater length below.114
The second argument, which has become Stratas JA’s trademark over the past few years, is that attempts to draw a sharp distinction between different standards of review are misplaced, because judicial review is essentially a matter of degree.115 This line of reasoning emphasizes post-Dunsmuir signals from the Supreme Court that the reasonableness standard of review “takes its colour from the context.”116 But it also harkens back to Iacobucci J’s pre-Dunsmuir notion that there is a “spectrum” of review which prescribes correctness review for questions of law,117 and Sopinka J’s suggestion in CAIMAW v Paccar that the reasonableness refers to “a logical relationship between the grounds of the decision and the premises 118 thought by the court to be true.” 2017 CanLIIDocs 175
In this vein, Stratas JA asserts in both ex cathedra opinions and extra- judicial commentary that the purpose of reasonableness review is to verify whether an administrative decision falls within a range of “acceptability and defensibility” on the facts and the law119 that widens or narrows depending upon the nature of the issues involved. Thus, it is of paramount importance for reviewing courts to establish the relevant margin of appreciation at the outset before moving on to consider the procedural or substantive attributes of the decision under review. In cases where legislation “signals that the standard of review should be correctness” by providing a right of appeal,120 giving judges the power to certify a question of law for
114 See 140–142, below.
115 See e.g. First Nations Child and Family Caring Society of Canada v Canada (AG), 2013 FCA 75, 226 ACWS (3d) 813 [First Nations Child and Family Caring Society]. For academic commentary on Stratas JA’s opinion in that case, see Paul Daly, “Unreasonable Interpretations of Law” (2014) 66 SCLR (2d) 233.
116 Canada (Citizenship and Immigration) v Khosa, 2009 SCC 12 at para 59, [2009] 1 SCR 339; Catalyst Paper Corp v North Cowichan (District), 2012 SCC 2 at paras 17–18, [2012] 1 SCR 5.
117 Southam, supra note 49 at para 30, 54.
118 Paccar, supra note 47 at 1017–18: Reasonableness is not a quality that exists in isolation. When a court says that a decision under review is “reasonable” or “patently unreasonable” it is making a statement about the logical relationship between the grounds of the decision and the premises thought by the court to be true. Without the reference point of an opinion (if not a conclusion) on the merits, such a relative statement cannot be made.
119 Stratas, supra note 1 at 16. See also First Nations Child and Family Caring Society, supra note 115 at para 12.
130 UNBLJ RD UN-B [VOL/TOME 68 consideration by an appellate court,121 or setting out “recipes…that must be followed or other constraining words,”122 Stratas JA asserts that a reviewing court should allow an administrative decision-maker little or no margin of appreciation. By contrast, in cases which turn on “factual appreciation, fact-based discretions, administrative policies, or specialized experience and expertise not shared by the reviewing court on the particular point in issue,” he asserts that administrative decision-makers enjoy a relatively wide margin of appreciation.123 But the bottom line seems to be an inverse correlation between the degree to which an administrative decision involves questions of law (especially statutory provisions and common law doctrine)124 and the degree to which a court will defer to an administrative decision, because “[l]egal matters, as opposed to factual or policy matters, admit of fewer possible, acceptable outcomes.”125
In order to throw Stratas JA’s theory into stark relief, it is worth comparing the different margins of appreciation he employed in Delios v Canada (Attorney General) with the one he applied in Wilson.126 In Delios, a case concerning the interpretation of collective agreement provisions establishing an employee’s paid leave entitlement, Stratas JA held that the adjudicator was entitled to a “wide margin 2017 CanLIIDocs 175 of appreciation” because “labour adjudicators’ decisions are often protected by privative clauses” and:127
interpretations of collective agreement provisions involve elements of factual appreciation, specialization and expertise concerning collective agreements, the disputes that arise under them, the negotiations that lead up to them and, more broadly, how the management-labour dynamic swirling around them plays out in various circumstances.
120 On this point, Stratas JA approves of Slatter JA’s decision in Edmonton East, supra note 7, in which he held that a statutory right of appeal implies review on a correctness standard.
121 Stratas, supra note 1 at 4–5. See also Kanthasamy v Canada (Minister of Citizenship and Immigration), 2014 FCA 113 at paras 30–37, 372 DLR (4th) 539.
122 Stratas, supra note 1 at 15. See also Canada (Minister of Transport, Infrastructure and Communities) v Farwaha, 2014 FCA 56 at para 91, 238 ACWS (3d) 282; Canada (AG) v Boogaard, 2015 FCA 150 at para 43, 87 Admin LR (4th) 175; Canada (AG) v Almon Equipment Limited, 2010 FCA 193 at para 53, [2011] 4 FCR 203.
123 First Nations Child and Family Caring Society, supra note 115 at para 14; Forest Ethics Advocacy Association v Canada (National Energy Board), 2014 FCA 245 at para 82, 246 ACWS (3d) 191.
124 First Nations Child and Family Caring Society of Canada, supra note 115 at paras 14–15; Canada (AG) v Abraham, 2012 FCA 266 at para 45, [2013] 1 CTC 69 [Abraham]; Paradis Honey Ltd v Canada, 2015 FCA 89 at para 136, 382 DLR (4th) 720.
125 Abraham, supra note 124 at para 45; Hupacasath First Nation v Canada (AG), 2015 FCA 4 at para 66, 379 DLR (4th) 737.
126 Delios v Canada (AG), 2015 FCA 117, 100 Admin LR (5th) 301.
127 Ibid at paras 20–21.
2017] RENOVATING JUDICIAL REVIEW 131
By contrast, in Wilson Stratas JA does not mention the relevance of the privative clause and noted that “the statutory interpretation point before us involves relatively little specialized labour insight beyond the regular means the courts have at hand when interpreting a statutory provision.”128 Without further explanation, it is certainly debatable whether the margin of appreciation should vary depending on whether a labour adjudicator is interpreting collective agreement provisions regarding paid-leave entitlements as opposed to unjust dismissal provisions in the Labour Code. The unavoidable inference from cases like Delios, Wilson and others is that whenever administrative officials interpret legislation, Stratas JA’s analysis gravitates towards correctness review or, alternatively, a form of deference which is indistinguishable from it.
(b) Judicial Review Means Reasonable Pluralism…Unless There’s Only One Right Answer
When framed in this way, Stratas JA’s proposal for reform is only one small step removed from the doctrine of jurisdictional error, in which judges are entitled to 2017 CanLIIDocs 175 review questions of law on a correctness standard. At most, he is prepared to allow administrative officials a narrow margin of appreciation when interpreting legislation, but it is doubtful whether this makes any practical difference a case like Wilson because “whether we conduct reasonableness review or correctness review, the outcome of the appeal would be the same.”129 But such an approach discounts the fact that Parliament empowered adjudicators to interpret and apply the Labour Code, and the fact that the adjudicator in this case seemed to be exercising that power in a way which was rationally defensible in light of the Labour Code’s objectives, namely to resolve employment disputes in an efficient, but conciliatory, manner. Finally, it seems to conflict with the landmark ruling in cases like CUPE and Baker that judges should defer to administrative interpretations of broadly worded statutory provisions, provided they are “rationally supported by the relevant legislation.”130
In the Supreme Court, a majority led by Abella J concluded that the Federal Court of Appeal had erred in applying the correctness standard of review. On the standard of review issue, Abella J pointed out that both parties had agreed that the Adjudicator’s decision should be reviewed on a reasonableness standard. But she also relied on two recent Supreme Court decisions – Dunsmuir and Nor-Man Regional Health Authority Inc v Manitoba Association of Health Care Professionals – as authority for the proposition that the decisions of labour adjudicators and arbitrators should generally be afforded a significant degree of deference.131 Abella
128 Re Wilson and Atomic Energy, supra note 96 at para 58.
129 Ibid at para 58.
130 CUPE, supra note 42 at 237.
131 Wilson, supra note 7 at para 15. In another case from 2015, Kathasamy, supra note 7 the Supreme Court rejected Stratas JA’s suggestion that the Federal Court of Appeal should review certified questions of law on a correctness standard.
132 UNBLJ RD UN-B [VOL/TOME 68
J’s reliance on these two cases is noteworthy, because they both apply the more deferential standard of review to administrative interpretations of enabling legislation (Dunsmuir) and common law doctrine (Nor-Man), which undercuts the assertion that the degree of judicial deference narrows as the quotient of law involved in an administrative decision increases. And in another case from 2015, Kanthasamy v Canada (Citizenship and Immigration), Abella J authored another majority opinion which rejected the suggestion that the correctness standard ought to apply to cases in which a Federal Court had certified a question of law for consideration by the Federal Court of Appeal.132 Therefore, at least to this point the Supreme Court is not inclined to adopt Stratas JA’s proposal for renovating the standard of review analysis. To underline this point in Wilson, Abella J stated that an “attempt to calibrate reasonableness by applying a potentially indeterminate number of varying degrees of deference” was unhelpful, because it would further complicate an area of law was “in need of greater simplicity.”133
Regarding the Adjudicator’s decision in Wilson, Abella J concluded that it was reasonable, and therefore sent the matter back for a decision regarding the appropriate remedy. However, it remains unclear whether Abella J upheld the 2017 CanLIIDocs 175 decision simply because she agreed with the Adjudicator or whether she thought the Adjudicator’s had articulated a reasonable legal justification for his decision. When it came time to apply the reasonableness standard, Abella J stated:134
The issue here is whether the Adjudicator’s interpretation of ss 240 to 246 of the Code was reasonable. The text, the context, the statements of the Minister when the legislation was introduced, and the views of the overwhelming majority of arbitrators and labour law scholars, confirm that the entire purpose of the statutory scheme was to ensure that non- unionized federal employees would be entitled to protection from being dismissed without cause under Part III of the Code. The alternative approach of severance pay in lieu falls outside the range of “possible, acceptable outcomes which are defensible in respect of the facts and law” because it completely undermines this purpose by permitting employers, at their option, to deprive employees of the full remedial package Parliament created for them. The rights of employees should be based on what Parliament intended, not on the idiosyncratic view of the individual employer or adjudicator.
Adjudicator Schiff’s decision was, therefore, reasonable.
This passage suggests that, instead of beginning with an attempt to understand the Adjudicator’s reasons on their own terms, Abella J immediately applied traditional canons of statutory construction by engaging in a holistic analysis of the text,
132 Kanthasamy, supra note 7 at paras 42–44.
133 Wilson, supra note 7 at para 18.
134 Ibid at paras 39–40.
2017] RENOVATING JUDICIAL REVIEW 133 context, and legislative history in order to determine Parliamentary intent.135 Once she concluded that Parliament had not intended to allow employers to dismiss on a without cause basis, she considered the Adjudicator’s decision to be reasonable because he arrived at the same outcome. Thus, despite disagreeing about the appropriate standard of review, Abella J and Stratas JA agreed that the legality of the Adjudicator’s decision should be assessed according to a reviewing judge’s understanding of the Labour Code –they just parted ways over how to interpret the statute.136
The problem with this approach to reasonableness review is that it only pays lip service to the idea that administrative law has a meaningful impact on judicial reasoning about what the law requires. If one unpacks the reasonableness standard in this way, then any quibbling over the standard of review seems both vacuous and misleading, because judges can defer by first determining how the legislation should be interpreted and then verifying that the administrative decision- maker construed the legislation in substantially the same terms.
While Abella J could have rested at this point, she nevertheless offered 2017 CanLIIDocs 175 another proposal for reforming the standard of review analysis in obiter dicta.137 She began by noting that the Dunsmuir framework had fallen short of its objectives, because “where once the confusion was over the difference between patent unreasonableness and reasonableness simpliciter, we now find ourselves struggling over the different between reasonableness and correctness.”138 The consequence of this confusion was that judges and counsel engaged in protracted “terminological battles regarding the standard of review at the outset of judicial review, with relatively little regard for the merits of a particular administrative decision.139 Then, after noting several academic critiques of the Dunsmuir framework by leading administrative scholars, she raised the prospect of adopting a universal standard of reasonableness, saying:140
Nothing Dunsmuir says about the rule of law suggests that constitutional compliance dictates how many standards are required. The only requirement, in fact, is that there be judicial review in order to ensure, in particular that decision-makers do not exercise authority they do not have. I see nothing in its elaboration of rule of law principles that precludes the adoption of a single standard of review, so long as it accommodates the
135 Another prominent example of this reasoning is Canada (AG) v Mowat, 2011 SCC 53, [2011] 3 SCR 471, a case that is frequently held up as an example of disguised correctness review. See Mullan, “Unresolved Issues”, supra note 46 at 58–59; Matthew Lewans, “Deference and Reasonableness Since Dunsmuir” (2012) 38 Queen’s LJ 59 at 90–92.
136 Jones, supra note 9 at 38.
137 Wilson, supra note 7 at para 19.
138 Ibid at para 26.
139 Ibid at para 25.
140 Ibid at para 31.
134 UNBLJ RD UN-B [VOL/TOME 68
ability to continue to protect both deference and the possibility of a single answer where the rule of law demands it, as in the four categories singled out for correctness review.
So instead of Stratas JA’s proposal to draw more fine-grained distinctions regarding varying margins of appreciation, Abella J proposes to collapse different standards of review into one universal reasonableness standard which can accommodate a plurality of reasonable administrative decisions…unless a reviewing court concludes there is only one right answer.141
While moving to a universal standard of reasonableness might avoid protracted, unproductive handwringing over which standard of review to apply, defining the reasonableness in these terms will only render the application of that standard more fraught with contradiction. At best, Abella J’s proposal might avoid problems associated with applying different standards of review to multiple issues within the same administrative decision.142 But because the reasonableness standard is construed as requiring deferential assessment of administrative reasons unless the
court thinks there is only one correct outcome, there is a substantial risk that the “one 2017 CanLIIDocs 175 right answer” qualification will simply become an “analytical Trojan Horse that invites correctness review under the guise of reasonableness analysis.”143 As the history of judicial review amply demonstrates, judges are prone to overreaching their constitutional role by asserting there is only one right answer on questions of statutory interpretation, even when enabling legislation is cast in relatively general terms and assigns authority to administrative officials to resolve interpretive disputes. In those cases, where a court is convinced that an administrative decision offends the principle of legality, it should at least explain why that decision is unfair or cannot be rationally supported in light of the relevant legal principles.
Finally, it is revealing that Abella J ’s proposal would not have made any difference on the way the Wilson case unfolded. Despite applying the reasonableness standard at first instance, O’Reilly J held that the Adjudicator’s decision was unreasonable because it conflicted with what he deemed to be a proper reading of the Labour Code (i.e. that it entitled AECL to dismiss Wilson on a without cause basis). Stratas JA, for his part, thought it made no difference whether the standard of review was correctness or reasonableness, but ultimately agreed with O’Reilly J’s general interpretation of the Labour Code. While Abella J held that the Federal Court of Appeal applied the wrong standard of review, she essentially engaged in the same exercise, because she measured the legality of the Adjudicator’s decision against her own interpretation of the Labour Code. Thus, the apparent disagreement between O’Reilly J, Stratas JA, and Abella J over the appropriate standard of review masks
141 British Columbia (Securities Commission) v McLean, 2013 SCC 67 at paras 37–38, [2013] 3 SCR 895. For an excellent critique of the ‘one right answer’ gloss on reasonableness review, see Falzon, supra note 9.
142 See, e.g. Canadian Broadcasting Corp v SODRAC 2003 Inc, 2015 SCC 57, [2015] 3 SCR 615.
143 Falzon, supra note 9 at 160.
2017] RENOVATING JUDICIAL REVIEW 135 the fact that they all essentially believed there was one right answer to the question posed to the Adjudicator – they just disagreed about what that one right answer was. But by approaching judicial review in this way, all three judges effectively treat the Adjudicator’s decision as an ephemeral piece of advice instead of a piece of administrative law which merits judicial respect. This makes one wonder whether all this talk about deference is merely a rhetorical veneer designed to deflect concerns about judicial overreach.
(c) Let Sleeping Dogs Lie
Perhaps one should not put too much stock in Abella J’s proposal just yet, because the five other five concurring justices distanced themselves from her obiter comments. While McLachlin CJ (along with Karakatsanis, Wagner, and Gascon JJ, concurring) appreciated Abella J’s “efforts to stimulate a discussion on how to clarify or simplify our standard of review jurisprudence to better promote certainty and predictability,” she said she was “not prepared to endorse any particular proposal 144 to redraw our current standard of review framework at this time.” Similarly, while 2017 CanLIIDocs 175 Cromwell J agreed that “developing new and apparently unlimited numbers of gradations of reasonableness review”145 would not improve matters, he cautioned that “our standard of review jurisprudence does not need yet another overhaul.”146 Although many observers expected the Court to discuss Abella J’s proposal in Edmonton (City) v Edmonton East (Capilano) Shopping Centres Ltd, it ultimately deferred that discussion to a later date. Writing for a bare majority, Karakatsanis J noted in that case that the day for reforming the standard of review analysis “has not come, but it may be approaching,” which suggests that it may take up the cudgel again in the coming year.147
While it is difficult to speculate about where the Supreme Court might turn next, the prevailing sense of confusion regarding the standard of review is palpable. As I suggested earlier, the problems associated with the categorical approach to the standard of review in Dunsmuir continue to generate considerable litigation. Despite its attempt to signal a preference for reasonableness review in all cases where an administrative agency is interpreting its home statute, this has not deterred litigants and appellate courts from testing the limits of correctness review. In Wilson, the Federal Court of Appeal expanded correctness review to include situations where a reviewing court perceives a rift in administrative law; and in Edmonton East, the Alberta Court of Appeal expanded correctness review to cases where the enabling legislation includes a statutory right of appeal.148 Despite ample case law from the
144 Wilson, supra note 7 at para 70.
145 Ibid at para 73.
146 Ibid at para 72.
147 Edmonton East, supra note 7 at para 20.
148 Re Wilson and Atomic Energy, supra note 96 at para 39; Edmonton East, supra note 7 at para 24.
136 UNBLJ RD UN-B [VOL/TOME 68
Supreme Court that the appropriate standard of review is reasonableness in both scenarios,149 both appellate courts interpreted Dunsmuir as authorizing correctness oversight. And as long as litigants perceive they might find a receptive audience for expanding Dunsmuir’s categories of correctness review, even in cases concerning the interpretation of an administrator’s home statute, they will continue to test the Supreme Court’s resolve regarding administrative interpretations of law.
Nevertheless, the majority’s reasoning in Edmonton East provides a glimmer of hope that reengagement with the underlying principles of fairness and reasonable justification might yet provide helpful guidance about how to verify the legality of an administrative decision without embarking on a quixotic quest to determine the correct outcome to the question posed to the primary decision-maker. The case concerned a decision of the Edmonton Assessment Review Board to increase the assessed value of a shopping mall. The mall had originally been assessed as having a value of $31 million, which the owner challenged on the grounds that it exceeded the market value of the property and/or was inequitable in light of the assessed value of similar properties. However, while preparing for the hearing, the municipality discovered that the mall had been mistakenly classified as a 2017 CanLIIDocs 175 “community centre,” which meant that its rental value had been grossly undervalued. The City informed the owner of its mistake, and indicated that it would ask the Assessment Review Board to increase the assessed value of the property at the appeal hearing. During the hearing, counsel for the property owner conceded that the Board had the legal authority to increase the assessed value of the property, because the enabling legislation gave it the power to “make a change to an assessment roll or tax roll.”150 However, when the Board increased the assessed value of the property, the owner sought leave to appeal to the Court of Queen’s Bench on “a question of law or jurisdiction of sufficient importance” to warrant judicial intervention.151
At first instance, Rooke ACJ held that the appropriate standard of review was correctness, because the Board’s decision involved a jurisdictional question about whether it could increase an assessment that had been challenged by a property owner.152 But the Alberta Court of Appeal (perhaps sensing that the parameters of “jurisdictional error” had waned since Dunsmuir),153 held that the correctness standard applied because the statutory right of appeal implied that “the Legislature clearly intended that the administrative decision maker make the initial decision,
149 See, e.g. Domtar Inc v Quebec (Commission d’appel en matière de lesions professionelles), [1993] 2 SCR 756, 105 DLR (4th) 385; Pezim, supra note 48; Southam, supra note 49; Smith v Alliance Pipeline Ltd, 2011 SCC 7, [2011] 1 SCR 160; Ontario (Energy Board) v Ontario Power Generation Inc, 2015 SCC 45, [2015] 3 SCR 147; ATCO Gas and Pipelines Ltd v Alberta (Utility Commission), 2015 SCC 45, [2015] 3 SCR 219.
150 Municipal Government Act, RSA 2000, c M-26, s 467.
151 Ibid at s 470(5).
152 Edmonton East (Capilano) Shopping Centres Ltd v Edmonton (City), 2013 ABQB 526, 570 AR 208.
153 See supra note 84.
2017] RENOVATING JUDICIAL REVIEW 137 subject to review by the court.”154 In doing so, it proposed to expand Dunsmuir’s categories of correctness review to include cases where an administrative decision was subject to a statutory right of appeal,155 and cited the Supreme Court’s recent decision in Tervita Corp v Canada (Commissioner of Competition) as authority for this addendum.156
In a highly anticipated decision, the Supreme Court restored the Board’s decision by a narrow 5-4 margin. In her majority opinion, Karakatsanis J applied what she took to be the predominant approach to judicial review in the post- Dunsmuir era. After briefly reviewing the Board’s powers under the Municipal Government Act, including its statutory power to “change” an assessment, Karakatsanis J held that a presumption of deference applied so as to convey judicial respect for “the principle of legislative supremacy and the choice made to delegate decision making to a tribunal” and foster access to justice administered by a “flexible and expert tribunal.”157 Moreover, she quickly disposed of the submission that the Board’s decision involved a jurisdictional question, by saying that this category had been narrowly construed in recent years, and reiterating that the Board’s 158 interpretation of its home statute merited a deferential standard of review. In 2017 CanLIIDocs 175 response to the Court of Appeal’s suggestion that a statutory right of appeal should be construed as signaling the legislature’s preference for correctness oversight, Karakatsanis J noted that such a move conflicted with “strong jurisprudence from this Court.”159
But the most interesting aspect of Karakatsanis J’s decision concerns her method of assessing whether the Board’s decision was reasonably justified. Instead of applying ordinary canons of statutory construction to determine directly whether the legislation empowered the Board to increase the property’s assessed value, she set out to verify “the existence of justification, transparency and intelligibility within the decision-making process.”160 She began by noting that the City had apprised the owner that it was going to ask the Board to increase the assessed value, and the owner had filed lengthy submissions in response. More importantly, Karakatsanis J highlighted how the property owner’s counsel had conceded at the initial Board hearing that the enabling legislation empowered the Board to “change” to the assessment roll, which meant it had the power to increase the assessed value.
154 Edmonton East, supra note 7 at para 24.
155 Ibid.
156 Edmonton East, supra note 7 at para 25, citing Tervita Corp v Canada (Commissioner of Competition), 2015 SCC 3, [2015] 1 SCR 161.
157 Edmonton East, supra note 7 at para 22.
158 Ibid at para 22.
159 Ibid at paras 28–31.
160 Dunsmuir, supra note 5 at para 47.
138 UNBLJ RD UN-B [VOL/TOME 68
Therefore, the Board’s decision-making process had clearly been fair and transparent.161
But was the Board’s decision reasonable in light of the available evidence and relevant law? On this point, Karakatsanis J admitted that when an administrative decision-maker fails to give reasons, “it makes the task of determining the justification and intelligibility of the decision more challenging.”162 While she conceded that complete administrative silence would be a reviewable in cases where the duty of fairness required “some form of reasons,” she noted that the duty to provide reasons was not a universal requirement.163 Thus, she asserted that “when a tribunal’s failure to provide any reasons does not breach procedural fairness, the reviewing court may consider the reasons ‘which could be offered’ in support of the decision,” including (but not limited to) existing administrative law which articulates reasons which lend support for the decision under review.164 After reiterating that the Board had treated the property owner fairly by inviting submissions on whether the Board had the power to increase the initial assessment, Karakatsanis J proceeded to review the overarching objectives of the statute, specific statutory provisions, and existing administrative law regarding the purpose and function of the Assessment 2017 CanLIIDocs 175 Review Board in order to verify whether the Board’s decision was rationally defensible. She pointed out that the language of the specific provision gave the Board relatively broad powers to “change” an assessment in order to achieve the statutory objective of ensuring that properties are assessed in a fair and equitable manner.165 Furthermore, she noted a prior decision of the Alberta Municipal Government Board, which had interpreted the provision as enabling the Assessment Review Board to increase or decrease an assessment in order to arrive at a fair and equitable assessment.166 Therefore, even though she did not define the true meaning of s. 467 of the Municipal Government Act, Karakatsanis J was nevertheless able to verify that the Board’s decision was reasonable because its interpretation was “consistent with the ordinary meaning of ‘change’ and the overarching policy goal of the MGA, to ensure assessments are correct, fair and equitable.”167
What is most remarkable about Karakatsanis J’s opinion is the attempt to articulate an integrated account reasonableness review – one which draws a connection between judicial deference, fair process, and substantive review. The starting point is a reminder that the purpose of judicial review is to ensure the
161 Edmonton East, supra note 7 at paras 36–40.
162 Ibid at para 36.
163 Ibid at para 37; as L’Heureux-Dubé J noted in Baker, supra note 3 at para 43 “in certain circumstances, the duty of procedural fairness will require the provision of a written explanation for a decision.”
164 Edmonton East, supra note 7 at para 38.
165 See, eg Municipal Government Act, supra note 150 at ss 293(1), 324(1), and 467(3).
166 Edmonton East, supra note 7 at para 44.
167 Ibid at para 61.
2017] RENOVATING JUDICIAL REVIEW 139 legality of an administrative decision, rather than to prioritize judicial interpretations of law. The second point is that the legality of an administrative decision depends partially on whether a party has been given an opportunity to provide submissions regarding how the law should be applied in light of the particular facts and circumstances. Finally, the legality of an administrative decision depends on whether it can be rationally defended in light of the relevant underlying principles of the relevant legal framework for the decision. When viewed in this way, Karakatsanis J’s opinion provides a robust, but nuanced, approach to reasonableness review. As in Baker, Karakatsanis J leaves aside the question whether the decision under review concerns a question or issue which qualifies for judicial deference in the abstract, and instead asks whether the decision-maker had been attentive to the litigant’s submissions and had provided a decision which was capable of being justified in light of relevant law (including administrative law on the subject).
Nevertheless, because Karakatsanis J held that the Board was not required to provide reasons and instead opted to supplement the Board’s decision with reasons which might have been offered in support of the decision, her decision does not vindicate the values articulated in Baker – especially the value of enabling a party 2017 CanLIIDocs 175 “to see that the applicable issues have been carefully considered.”168 By affirming the proposition that judges can consider reasons “which could be offered” (but were not, in fact, offered by the decision-maker),169 Karakatsanis J’s opinion provides a perverse incentive for decision-makers to issue conclusions without a supporting legal rationale in the hopes that a reviewing court might rectify their omission. From an evaluative perspective – one that takes its cues from Baker – the duty to provide reasons is significant because it serves to verify that an administrative authority is “exercised in accordance with the boundaries imposed in the statute, the principles of the rule of law, the principles of administrative law, the fundamental values of Canadian society, and the principles of the Charter.”170 Therefore, one hopes that the Court will revisit this issue in greater depth in future cases, because it will likely yield important insights about the connection between judicial deference and the fundamental values which legitimate administrative decisions and warrant judicial respect.171
(d) The Rule of Law Requires Correctness Review
Up to this point, it appears that although judges generally acknowledge that they should defer to administrative decisions, they often renege on that commitment because they remain wedded (consciously or otherwise) to the formalist notion that
168 Baker, supra note 3 at para 39.
169 Alberta Teachers’ Association, supra note 79 at paras 51–53.
170 Baker, supra note 3 at para 56.
171 For a more thorough scholarly analysis of this connection, see Dyzenhaus & Fox-Decent, supra note 4 at 218–238.
140 UNBLJ RD UN-B [VOL/TOME 68 questions of law should generally be determined by the judiciary.172 Sometimes this tendency is couched in terms of allowing administrative officials a narrow margin of appreciation when interpreting legislation and common law principles; sometimes it confuses how judges assess the reasonableness of an administrative decision, even to the point of saying that judges should defer to administrative decisions unless they believe there is only one right answer. But the last proposal for reforming the law of judicial review seeks to clarify it by dispensing with the pretense of judicial deference to administrative decisions on matters of substance altogether.
Recall that when the Federal Court of Appeal dismissed the appeal in Wilson, Stratas JA held that the court should review the Adjudicator’s decision on a correctness standard because “the current state of adjudicators jurisprudence is one of persistent discord.”173 Therefore, despite recognizing that “Parliament has vested jurisdiction in adjudicators under the Code to decide questions of statutory interpretation” Stratas JA concluded that “the rule of law concern predominates in this case and warrants this Court intervening to end the discord and determine the legal point once and for all. We have to act as a tie-breaker.”174 To support this conclusion, he noted that the Dunsmuir framework prescribed correctness review for 2017 CanLIIDocs 175 questions of law which are of general importance to the legal system and outside the expertise of the decision-maker. In his view, conflicting administrative decisions raises a question of general importance, because the outcome of employment disputes would depend on the identity of the decision-maker;175 and since the issue concerned a matter of statutory interpretation, he thought that “the statutory interpretation point before us involves relatively little specialized labour insight beyond the regular means the courts have at hand when interpreting a statutory provision.”176
In a remarkable dissenting opinion, three members of the Supreme Court agreed that the Adjudicator’s decision should be reviewed on a correctness standard. Like Stratas JA, Côté and Brown JJ (with Moldaver J, concurring) expressed the concern that different Adjudicators had reached different conclusions about whether the Labour Code permitted federally regulated employers to dismiss without cause. For the dissenting justices, this raised a “serious concern for the rule of law,”177 because “what the law means depends on whether one’s case is decided by one decision-maker or another.”178 Thus, despite the fact that the Adjudicator was
172 Harry Arthurs, “Rethinking Administrative Law: A Slightly Dicey Business” (1979) 17 Osgoode Hall LJ 1.
173 Re Wilson and Atomic Energy, supra note 96 at para 52.
174 Ibid at para 55.
175 For a similar argument regarding judicial review of refugee claims, see Sean Rehaag, “Judicial Review of Refugee Determinations: The Luck of the Draw?” 38 Queen’s LJ 1.
176 Re Wilson and Atomic Energy, supra note 96 at para 58.
177 Wilson, supra note 7 at para 74.
2017] RENOVATING JUDICIAL REVIEW 141 interpreting his home statute, Côté and Brown JJ concluded that the Court ought to impose its own interpretation of the Code because rule of law values trumped what would otherwise be tantamount to “indiscriminate deference to the administrative state.”179
In order to appreciate the full import of the dissenting opinion, we need to revisit the statutory framework which authorized the Adjudicator’s decision. To recap: the Labour Code gives non-unionized employees a statutory right to “make a complaint in writing to an inspector if the employee has been dismissed and considers the dismissal to be unjust”;180 the filing of a complaint triggers a statutory duty for the employer “to provide a written statement giving the reasons for the dismissal”;181 if the parties cannot resolve the complaint, the inspector is required to report the matter to the Federal Minister of Labour,182 who has the discretionary power to appoint “any person that the Minister considers appropriate as an adjudicator to hear and adjudicate on the complaint”;183 an adjudicator has the power to “determine the procedure to be followed, but shall give full opportunity to the parties to the complaint to present evidence and make submissions”184 and determine 185 “whether the dismissal of the person who made the complaint was unjust”; and the 2017 CanLIIDocs 175 adjudicator’s determination is deemed to be “final and shall not be questioned or reviewed in any court.”186
Given this legal framework, it seems odd that the dissenting opinion would resort directly to correctness review without even pausing to note that Professor Schiff had been appointed by the Minister of Labour to adjudicate Wilson’s complaint, and Parliament had instructed courts to exercise restraint when reviewing the Adjudicator’s decision. To characterize reasonableness review as “indiscriminate” under these circumstances conveys remarkable disregard for the will of Parliament, the objectives of the Labour Code, and the Adjudicator’s reasons for upholding Wilson’s complaint. Furthermore, while the dissenting opinion makes a passing reference to Dunsmuir,187 its authors expended no effort to apply the standard of review analysis to the facts at hand. So even as a piece of common law reasoning, the dissenting opinion’s standard of review analysis is rather glib. Finally,
178 Ibid at para 85.
179 Ibid at para 79.
180 Canada Labour Code, supra note 89 at s 240.
181 Ibid at s 241(1).
182 Ibid at s 241(3)
183 Ibid at s 242(1).
184 Ibid at s 242(2).
185 Ibid at s 242(3)(a).
186 Ibid at s 243.
187 Dunsmuir, supra note 5 at para 80.
142 UNBLJ RD UN-B [VOL/TOME 68 the dissenting opinion adopts an extremely low threshold for triggering correctness review, by stating “[a]s long as there is one conflicting but reasonable decision, its very existence undermines the rule of law.”188
But perhaps the most significant aspect of the dissenting opinion, is that its authors have consistently sought to revive the Anisminic approach to judicial review by expanding Dunsmuir’s presumptive categories for correctness review. In Wilson, Côté and Brown JJ (with Moldaver J, concurring) wrote a dissenting opinion which construed the Adjudicator’s decision regarding Wilson’s unjust dismissal complaint as involving “a matter of general importance, defining the basis of the employment relationship for thousands of Canadians”189 without even bothering to consider important institutional reasons for judges to respect the Adjudicator’s decision. In Edmonton East, Côté and Brown JJ wrote another dissenting opinion (with McLachlin CJ and Moldaver J, concurring), which construed a statutory right of appeal as a signal from the legislature not to defer the decisions of the Edmonton Assessment Review Board’s interpretation of provisions in its home statute regarding its power to “change” the assessed value of property. And in Commission scolaire de Laval v Syndicat de l’enseignement de la region de Laval, Côté J wrote a 2017 CanLIIDocs 175 dissenting opinion (with Wagner and Brown JJ, concurring), which concluded that an arbitrator’s decision regarding the admission of evidence at a hearing raises a general questions of law which are of central importance to the administration of justice as a whole and on which an arbitrator has no expertise.
All of these cases raised issues falling squarely within the statutory mandate of various administrative officials, agencies, and institutions. And in all of these cases, the administrative decision-making processes were fair and generated decisions which were justifiable in light of legislative objectives and other sources of administrative law such as prior tribunal decisions or collective agreements. Nevertheless, each of these cases contained a dissenting opinion which attracted 2–4 justices who took the view that the correctness standard applied because the decision either involved a general question of law or the administrative decision-maker lacked relative expertise. While the cleavage on the Supreme Court on matters pertaining to the standard of review still favours a deferential standard of review when administrator’s interpret their home statute, there is enough disagreement to entice litigants to push the boundaries of Dunsmuir’s categories of correctness review in future cases.
IV. Conclusion
At this point, it seems Canadian administrative law has arrived at yet another crossroad. The question is whether the Supreme Court will stick to the path laid out in post-Dunsmuir case law by establishing a deferential baseline for reviewing
188 Wilson, supra note 7 at para 89.
189 Ibid at para 91.
2017] RENOVATING JUDICIAL REVIEW 143 administrative decisions, or chart a new course which further complicates the threshold exercise of identifying the standard of review. Judging from the cases on this year’s docket, it appears that the Court will confront this issue in the context of administrative decisions concerning human rights.190 If so, resorting to precedent will be of little use, because there is genuine confusion over whether judges should defer to the decisions of human rights tribunals when they interpret their home statute. 191 And while the Supreme Court held in Doré v Barreau du Québec192 that judges should assess the legality of administrative decisions concerning Charter values according to a reasonableness standard, its resolve on that issue seems to have weakened in Loyola High School v Quebec (Attorney General).193 Furthermore, the presumptive categories in Dunsmuir will be of little use, because such cases require a nuanced understanding of an administrator’s home statute, but also raise human rights issues which are of general importance. Similarly, if the Court resorts to Dunsmuir’s contextual standard of review analysis the contextual factors will likely pull in different directions. The bottom line is that, as long as the Dunsmuir framework remains in place, there is a strong possibility that the Court will remain split over whether judges should review administrative decisions concerning human rights on a correctness or reasonableness standard. 2017 CanLIIDocs 175
One proposal for escaping this predicament is to eschew attempts to define formal categories of administrative decisions which warrant judicial deference, and reinvigorate an evaluative understanding of administrative law in which judges relinquish their interpretive monopoly over questions of law. Such an approach would focus instead on assessing the legality of administrative decisions in light of fundamental values like procedural fairness and reasoned justification. The outline for such an approach can be found in Abella J’s proposal for a universal reasonableness standard coupled with L’Heureux-Dubé J’s understanding of judicial review as a means for ensuring the legitimacy of administrative law. Such an approach would impose important procedural and justificatory burdens on administrative officials who decide human rights issues, burdens which are analogous to the burdens imposed by courts under s. 1 of the Charter.194 But even though these justificatory burdens entail robust limits on administrative action, they do not empower judges to determine outcomes for administrative officials to implement; rather, they serve to ensure that administrative officials exercise their statutorily delegated authority in a manner which is reasonably justifiable in a free and democratic society. If one takes this proposal seriously, then the perennial
190 Stewart, supra note 7; Trinity Western University v Law Society of Upper Canada, supra note 7; Trinity Western University v Law Society of British Columbia, supra note 7.
191 See, e.g. Canada (Human Rights Commission) v Canada (AG), 2016 FCA 200, 402 DLR (4th) 160; Andrew Pinto, “Deference in Disarray: Standards of Review of Human Rights Tribunal Decisions across Canada” online:
192 Doré, supra note 80.
193 Loyola High School, supra note 80.
194 Dyzenhaus & Fox-Decent, supra note 4 at 240.
144 UNBLJ RD UN-B [VOL/TOME 68 haggling over the standard of review is rendered obsolete, and judges and lawyers can proceed to ask more important and interesting questions regarding the legality of administrative action.
2017 CanLIIDocs 175
IDENTIFYING THE REVIEW STANDARD: ADMINISTRATIVE DEFERENCE IN A NUTSHELL
The Hon. Joseph T. Robertson, Q.C.*
Overview
The volume of Supreme Court jurisprudence dealing with the review of decisions rendered by administrative decision-makers and the application of the deference doctrine is overwhelming. While Canadian Union of Public Employees, Local 963 v New Brunswick Liquor Corporation1 and Dunsmuir v New Brunswick2 are often regarded as lead decisions, there are close to 200 precedents, spanning six decades, underscoring the doctrine’s evolution. Most were rendered over the last 35 years and most proved not to be of long-term precedential significance. They simply demonstrate the proper application of the deference doctrine, as it stood at the time 2017 CanLIIDocs 175 the case was decided, while affirming the Court’s error-correcting role.3 Regrettably, the doctrine has been unable to escape criticism.
Professor Paul Daly writes of the Court’s struggle to achieve “coherence” in Canadian administrative law. Justice David Stratas, writing in his personal capacity, has concluded that: “Doctrinal incoherence and inconsistency plague the Canadian law of judicial review.”4 Prior to those observations, it was David Mullan who isolated fifteen issues that remained outstanding following the release of Dunsmuir.5 Such commentaries suggest that the prospect of accurately distilling the tenets of the
* Formerly of the Federal Court of Appeal (1992-2000) and the New Brunswick Court of Appeal (2000- 2014) and presently Jurist-in-Residence, Faculty of Law, University of New Brunswick. The subject matter of this paper was originally presented at the Mid-Winter meeting of the Canadian Bar Association, NB Branch, February 6, 2016, and has been revised in both content and style in order to reflect the jurisprudence of the Supreme Court as of January 1, 2017. My gratitude to Christopher Pelkey (Class of 2018) for his assistance and patience.
1 Canadian Union of Public Employees, Local 963 v New Brunswick Liquor Corporation, [1979] 2 SCR 227, 25 NBR (2d) 237 [New Brunswick Liquor].
2 Dunsmuir v New Brunswick, 2008 SCC 9, [2008] 1 SCR 190 [Dunsmuir].
3 For a historical account of the Supreme Court’s deference jurisprudence see: JT Robertson, “Judicial Deference to Administrative Tribunals: A Guide to 60 Years of Supreme Court Jurisprudence” in Joseph Robertson, Peter A Gall & Paul Daly, eds, Judicial Deference to Administrative Tribunals in Canada: Its History and Future (Toronto: LexisNexis, 2014) 1 [Robertson, “Judicial Deference”]. This book was also published as (2014) 66 SCLR (2d) 1.
4 David Stratas, “The Canadian Law of Judicial Review: A Plea for Doctrinal Coherence and Consistency” (20 February 2016), online:
5 See David Mullan, “Unresolved Issues On Standard of Review In Canadian Judicial Review Of Administrative Action – The Top Fifteen!” (2013) 42 Adv Q 1; Paul Daly, Administrative Law Matters (blog), online:
146 UNBLJ RD UN-B [VOL/TOME 68 deference doctrine into a nutshell format is misguided. Thankfully, the administrative lawyer, in search of go-to answers, knows of the risks inherent in any presentation that oversimplifies the law.
Neither the volume of litigation nor academic commentary surrounding administrative deference detracts from the need for an analytical framework that enables reviewing courts to address two essential questions. The first is whether the decision is owed deference on the review standard of reasonableness. Otherwise, the correctness standard applies. Second, assuming the deferential standard applies, there is an obvious need to know how reviewing courts are to assess the reasonableness of an administrative decision.
Due to time restraints, this presentation focuses on only the first question. Admittedly, the task of providing a go-to answer for the second question is riddled with difficulty from both a practical and theoretical perspective. In particular, the application of the deferential standard of review to decisions that involve the interpretation of the decision-maker’s enabling legislation has been largely ignored.
However, it was agreed that today’s presentation would focus on standard of review 2017 CanLIIDocs 175 issues.
This presentation draws a bright-line distinction drawn between the decisions of specialized tribunals (e.g., labour boards) and those made by other statutory delegates (e.g., Ministers and officers of the Crown). Admittedly, when it comes to those falling within the residual category, the analysis is as argumentative as it is descriptive. Regardless, the distinction is important if only because it draws attention to what some regard as a design flaw in the deference doctrine. This topic warrants separate consideration and is addressed in the latter portion of my presentation.
Stripped to its essentials, my underlying thesis is neither complicated nor controversial. Dunsmuir left us with a two-step framework for identifying the proper review standard. The first embraces the categorical approach. The second is labeled the contextual approach or what is often referred to as contextualism. Dunsmuir anticipated the categorical approach could prove “unfruitful” and, therefore, reviewing courts would have to move to the contextual one. However, the post- Dunsmuir jurisprudence reveals that contextualism is no more. The Supreme Court has consistently applied the categorical approach and expressly rejected the contextual one.
In short, under the categorical approach, the deferential standard of review applies unless the issue at hand falls within of the four correctness categories first identified in Dunsmuir. Moreover, as the correctness categories are narrow in scope, there is little room for the application of the non-deferential standard of review. Better still, it matters not whether the administrative decision-maker is a specialized tribunal. So too have other statutory delegates have been brought under the deference umbrella. Nor does it matter that the statutory delegate lacks relative expertise with
2017] IDENTIFYING THE REVIEW STANDARD 147 respect to the issue at hand (e.g., statutory interpretation). This is deference in a nutshell, at least when it comes to identifying the proper review standard.
The Demise of Contextualism
Recall that prior to Dunsmuir, the analytical framework for isolating the proper standard of review was labeled the “pragmatic and functional approach” as articulated in Pushpanathan v Canada (Minister of Citizenship and Immigration).6 Also, recall that the object of the exercise was to isolate the intent of the legislature or Parliament as to whether the tribunal decision would be owed deference. The framework involved consideration of four contextual factors: (1) the presence or absence of a privative clause in the tribunal’s home statute; (2) the purpose of the statute; (3) the expertise of the tribunal; and (4) the nature of the issue. Finally, recall that the law provided for two deferential standards of review as a result of the Supreme Court’s decision in Canada (Director of Investigation and Research, Competition Act) v Southam Inc7 (patent unreasonableness and reasonableness simpliciter). 2017 CanLIIDocs 175
Only the legal historian ever asks why the Court felt compelled to promote two distinct deferential standards of review. What really mattered was how the reviewing court would justify its decision to select one of the deferential standards over the other, once the correctness standard had been eliminated from the mix. And for nearly a decade, reviewing courts went about their business imagining that a valid distinction could be drawn between the two deferential standards.
Dunsmuir dispensed with the pragmatic and functional label and replaced it with another: "standard of review analysis." Substantively, however, nothing changed with respect to the essential elements of the deference doctrine, save for the all-important reduction in the number of deferential standards of review. Thankfully, Dunsmuir left us with only one: reasonableness. It also left us with a simplified analytical framework for identifying the proper review standard. And for the record, Dunsmuir did not abandon the understanding that the search for the proper review standard was a search for legislative intent.8 The abandonment occurred post- Dunsmuir.
Dunsmuir was the Court’s response to the doctrinal uncertainties that had accumulated over the years. Bastarache and LeBel JJ, writing for the majority of five, consolidated the doctrine’s tenets under one umbrella. It is a two-step framework for assessing whether a tribunal decision is owed deference:
6 Pushpanathan v Canada (Minister of Citizenship and Immigration), [1998] 1 SCR 982, 160 DLR (4th) 193 [Pushpanathan].
7 Canada (Director of Investigation and Research, Competition Act) v Southam Inc, [1997] 1 SCR 748, 144 DLR (4th) 1.
8 Dunsmuir, supra note 2 at paras 31 and 52.
148 UNBLJ RD UN-B [VOL/TOME 68
First, courts ascertain whether the jurisprudence has already determined in a satisfactory manner the degree of deference to be accorded with regard to a particular category of question. Second, where the first inquiry proves unfruitful, courts must proceed to an analysis of the factors making it possible to identify the proper standard of review.9
In short, the first step relieves reviewing courts of the obligation to conduct an exhaustive review under the second step. Fortunately, Dunsmuir provides us with a list of issues for which the proper review standard had already been identified in the earlier Supreme Court jurisprudence. In fact, there are two lists. One contains questions/issues for which the deferential standard of review applies. The other sets out those questions/issues for which correctness is automatically the proper review standard. The person in search of the proper review standard simply looks to the two lists to see where the issue at hand falls. This is labeled the categorical approach.
The Dunsmuir list for review on the standard of reasonableness embraces questions of fact, mixed law and fact, and decisions involving the application of policy or the exercise of discretion. We are also told that deference will “usually” result where a tribunal is interpreting its home statute or those statutes closely 2017 CanLIIDocs 175 connected to the tribunal’s functions and with which it will have particular familiarity. Parenthetically, the Court would subsequently replace the word “usually” with the word "presumptively.10
As to the other list, Dunsmuir established that correctness is automatically the proper review standard for the following: constitutional questions; questions of general law that are of central importance to the legal system as a whole and outside the tribunal’s field of expertise; questions regarding the jurisdictional lines between two or more competing specialized tribunals; and, finally, true questions of jurisdiction.
Note, that although Dunsmuir made no specific reference to the correctness standard applying to alleged breaches of the fairness duty (e.g., bias), the Supreme Court has yet to declare otherwise; a matter discussed below.11 Note also, that under the first step, the Supreme Court has on occasion turned to earlier case law, involving the same tribunal and home statute, in order to isolate the proper review standard with respect to a particular issue.12 The Court of Appeal of New Brunswick has done likewise. In this way, reviewing courts are relieved of the obligation to conduct an exhaustive review required under the second step of the standard of review analysis.
9 Ibid at para 61.
10 Edmonton (City) v Edmonton East (Capilano) Shopping Centres Ltd, 2016 SCC 47 at para 22, [2016] 2 SCR 293 [Edmonton East].
11 See discussion, infra note 30.
12 See Tervita Corp v Canada (Commissioner of Competition), 2015 SCC 3, [2015] 1 SCR 161 [Tervita], discussed infra 27.
2017] IDENTIFYING THE REVIEW STANDARD 149
For example, historically, the Court of Appeal has reviewed the decisions of the Appeals Tribunal of the Workplace Health, Safety and Compensation Commission, involving a question of law, on the standard of correctness.13 Invariably, the question of law involves the interpretation of the applicable legislation. The justification for not according deference to this specialized tribunal is embedded in the reality that, save perhaps for the Chair of the tribunal, members come to the Appeals Tribunal without legal training. However, there is a caveat: to the extent the relevant legislation has been amended to reflect the appointment of legally trained persons to the appeals tribunal, it may well be that the Court will be asked to reconsider whether that non-deferential standard of review applied to questions of law (e.g., the interpretation of the tribunal’s home statute) should remain in place. As well, some of the post-Dunsmuir decisions of the Supreme Court, to be discussed momentarily, suggest that the earlier New Brunswick jurisprudence should be revisited.
Should the first step of the standard of review analysis prove unproductive, Dunsmuir anticipated that reviewing courts would move to the second. The second step requires the examination of four contextual factors. As noted, this has been 2017 CanLIIDocs 175 labeled the contextual approach or simply contextualism. The stated objective is to identify the intent of Parliament or the legislature with respect to whether the tribunal decision warranted deference. However, no one factor was determinative of the standard of review issue.
In theory, and prior to November 4, 2016, a reviewing court might have found it necessary to move to Dunsmuir’s second step of contextualism in order to identify the proper standard of review. In practice, however, it should not have been necessary to do so. Why? Save for two cases under consideration at the time Dunsmuir was decided, the Supreme Court had not expressed the need or expressed an interest in moving to the second step.14 Over a period of eight years, the Supreme Court had consistently identified the proper standard of review by reference to the one applied in earlier jurisprudence or, alternatively, by deciding whether the issue at hand falls within one of the categories for which the correctness standard automatically applies. If the issue did not fall within one of the exceptional categories, the tribunal’s decision was accorded deference.
On November 4, 2016, the Supreme Court released its decision in Edmonton (City) v Edmonton East (Capilano) Shopping Centres Ltd.15 But for that decision, one could have spent considerable time trying to isolate the rationale underscoring the demise of contextualism. The case is one in which the majority affirmed the application of the categorical approach before rejecting the contextual
13 See Keddy v New Brunswick Workplace Health, Safety and Compensation Commission, 2002 NBCA 24, (2002) 247 NBR (2d) 284 (leave to appeal to SCC refused).
14 The two cases are Canada (Citizenship and Immigration) v Khosa, 2009 SCC 12, [2009] 1 SCR 339, and Nolan v Kerry (Canada) Inc, 2009 SCC 39, [2009] 2 SCR 678.
15 Edmonton East, note 10, but see the dissenting opinion which sought to retain the contextual approach.
150 UNBLJ RD UN-B [VOL/TOME 68 one. The majority’s rationale for the rejection was brutally short and clear: “The contextual approach can generate uncertainty and endless litigation concerning the standard of review.”16 While Dunsmuir had left the door to contextualism open, the post-Dunsmuir jurisprudence closed it. Admittedly, the Court was deeply divided (5/4).
The fact that Dunsmuir’s contextual approach was laid to rest was unexpected. The fact the deferential review standard of reasonableness remains the go-to standard was not. Indeed, it was Binnie J, in Dunsmuir,17 who prophetically observed that most decisions and rulings of specialized tribunals are owed deference. Of the approximately 60 deference cases decided by the Supreme Court, post- Dunsmuir, the deferential standard of review was chosen in all but six. As to the exceptional cases, none came as a surprise.
In one case, the British Columbia Administrative Tribunals Act required correctness of the decision at hand.18 In another, the Court was dealing with the decision of a human rights tribunal and a question of law tied to the concepts of state 19 neutrality, religious freedom and freedom of conscience. In another, the Supreme 2017 CanLIIDocs 175 Court had been applying a correctness standard with respect to questions of law decided by a federal tribunal since Southam was decided.20 In two other cases, the Court applied the correctness standard because the tribunal in question and the federal court possessed overlapping jurisdiction to decide matters under the same federal statute.21 Finally, the Court recently concluded that the matter of solicitor- client privilege fell within of the four categories for which correctness is automatically the proper review standard.22
In brief, contextualism is no more. Consequently, the search for the proper review standard is no longer a search for legislative intent. Indeed, not since Dunsmuir has the Supreme Court alluded to that principle. The categorical approach is the only option for identifying the proper review standard and that approach hinges on two presumptions: a presumption of deference to the tribunal's decisions and a presumption that the tribunal possesses a relative expertise regarding the issue at hand. Admittedly, ten years prior to Dunsmuir, the Supreme Court set out an analytical framework for assessing the relative expertise of a tribunal. The pertinent
16 Ibid at para 35.
17 Dunsmuir, supra note 2 at para 146.
18 McCormick v Fasken Martineau DuMoulin LLP, 2014 SCC 39, [2014] 2 SCR 108.
19 Mouvement laïque québécois v Saguenay (City), 2015 SCC 16, [2015] 2 SCR 3 [Saguenay].
20 Tervita, supra note 12.
21 Rogers Communication Inc v Society of Composers, Authors and Music Publishers of Canada, 2012 SCC 35, [2012] 2 SCR 283 and Canada Broadcasting Corp v SODRAC 2003 Inc, 2015 SCC 57, [2015] 3 SCR 615 [Canada Broadcasting].
22 Alberta (Information and Privacy Commissioner) v University of Calgary, 2016 SCC 53, 403 DLR (4th) 1 [University of Calgary].
2017] IDENTIFYING THE REVIEW STANDARD 151 decision is Pushpanathan.23 However, it is no longer necessary to ask whether the tribunal possesses a relative expertise. Today, the twin presumptions of deference and expertise stand unless the issue falls within one of Dunsmuir’s four correctness categories. This is what is meant by freestanding expertise.
Unfortunately, too many commentators fail to appreciate that the concept of freestanding expertise is a theoretical construct based on a false premise: namely, all tribunals are composed of experts who are experts on all matters. In point of fact, not all tribunals are created equal, yet our deference doctrine perpetuates the understanding that one-size-fits-all; a thesis I must leave for another day.
As the law presently stands, it makes no difference whether you are dealing with the decisions of our national regulators or underfunded provincial tribunals that depend heavily on lay members who are appointed for reasons unrelated to the work of the tribunal. The decisions of both are treated no differently when it comes to the application of the deference doctrine. Reasonableness is the go-to standard unless the issue at hand falls within of one Dunsmuir’s correctness categories. What matters in law is that a specialized tribunal has been established to deal with issues for which 2017 CanLIIDocs 175 reviewing courts are deemed to lack a relative expertise. And that is why freestanding expertise has replaced legislative intent as the cornerstone of administrative deference. All of this adds another dimension to a doctrine that, at times, represents a challenge to conventional understandings of the rule of law.24
The Demise of the Right of Appeal
Despite the demise of contextualism, some continue to ask whether the law continues to draw a substantive distinction between the privative clause and the right of appeal. Typically, one or the other will be found lodged within the tribunal’s enabling (home) statute. However, it matters not whether the legislation contains either provision or none. The deference obligation operates independently of the tribunal’s enabling statute now that contextual approach has been formally abandoned.
In brief, there was a time in Canadian law when the distinction between judicial and appellate review truly made a difference when it came to isolating the proper standard of review. A right of appeal automatically signaled that correctness was the proper standard of review with respect to questions of law and, in particular, those involving the interpretation of the tribunal’s home statute. On the other hand, a privative clause automatically signaled the need for curial deference. This was the law according to New Brunswick Liquor. Eventually, however, Canadian law would hold that neither a privative clause nor a right of appeal were determinative of the
23 Pushpanathan, supra note 6 at para 32.
24 See Peter Gall, “Problems with a Faith-Based Approach to Judicial Review” in Robertson, Gall & Daly, supra note 3 at 183.
152 UNBLJ RD UN-B [VOL/TOME 68 legislature’s intent regarding the issue of deference.25 That was the law at the time Dunsmuir was decided. Today, under the post-Dunsmuir jurisprudence, both provisions remain an irrelevancy when it comes to isolating the proper review standard. The following case law supports that understanding of the law.
Three post-Dunsmuir decisions speak to the significance of a statutory right of appeal lodged within the tribunal’s enabling legislation. All three asked whether the statutory right, together with other contextual factors such as tribunal expertise, warranted application of the correctness standard with respect to questions of law. In all three cases, the Supreme Court effectively answered “no.”
In Mouvement laïque québécois v Saguenay (City),26 the Supreme Court remained true to its earlier precedents and bluntly answered "no." In Edmonton (City); the majority of the Court was emphatic in declaring that a statutory right of appeal is not a “new” category of correctness that should be added to the list set out in Dunsmuir. The third decision is admittedly problematic: Tervita Corp v Canada (Commissioner of Competition).27 In that case, the majority held that, as the decisions of the Competition Tribunal on questions of law are appealable to the 2017 CanLIIDocs 175 Federal Court of Appeal as if they were “a judgement of the Federal Court,” the proper review standard had to be correctness.28 The dissenting opinion refused to attach any significance to the wording of the appeal clause and insisted that the deferential standard of reasonable applied. In response, the majority noted that the Federal Court of Appeal had consistently applied the correctness standard to questions of law decided by the Competition Tribunal.
Curiously, the Court in Tervita failed to acknowledge the relevance of an earlier and significant precedent involving a decision of the Competition Tribunal: Southam. Another forgotten precedent, but one that would have supported the Supreme Court’s decision to apply the correctness standard regarding the Tribunal’s rulings on questions of law. In short, the Court has consistently applied the correctness standard to questions of law decided by the Competition Tribunal. This approach is in keeping with the first-step of Dunsmuir’s two-step framework for identifying the proper review standard.
In brief, those who believe that the standard of review issue hinges on either the privative clause or the right of appeal will be disappointed. Neither is of any moment when it comes to isolating the proper review standard. The post-Dunsmuir jurisprudence confirms that understanding.
25 The leading Supreme Court decision is Dr Q v College of Physicians and Surgeons of British Columbia, 2003 SCC 19, [2003] 1 SCR 226.
26 Saguenay, supra note 19.
27 Tervita, supra note 12.
28 See Competition Tribunal Act, RSC 1985, c 19, ss 12 and 13.
2017] IDENTIFYING THE REVIEW STANDARD 153
Little Room for Correctness Review
It is not difficult to show the categorical approach leaves little room for the application of the correctness standard of review. Two reasons underscore that observation. First, Dunsmuir identified a limited number of correctness categories and second, those categories are narrow in scope. As to the first reason, I admit that the correctness categories are under-inclusive. As mentioned earlier, Dunsmuir made no mention of those cases in which one of the parties alleges a breach of the fairness duty.
One should think of the fairness duty as an organizing principle – one that embraces those applicable to matters such as bias and procedural fairness, and any other matters recognized at law as sufficient grounds for setting aside a tribunal decision but that do not go to the merits of the underlying decision. Within this context, it is difficult to find recent Supreme Court jurisprudence dealing with the standard of review issue. There is an obvious reason for the omission. The law has remained constant before and after New Brunswick Liquor. A breach of the duty of 2017 CanLIIDocs 175 procedural fairness (the old rules of natural justice) including freedom from a biased decision-maker, was always treated as a jurisdictional error (excess of jurisdiction).29 That is why it is generally safe to proceed on the understanding that the presumption of tribunal expertise dissipates when it comes to alleged breaches of the fairness duty. However, there is an exception to this exceptional category.
One should not presume that correctness will always be the proper review standard when it comes to breaches of the fairness duty. In cases where the tribunal is statutorily mandated to establish rules and policies with respect to the procedures to be followed in fulfilling its adjudicative mandate, the standard of review may revert to reasonableness. Room for the exceptional case was recognized in Baker v Canada,30 Council of Canadians with Disabilities v VIA Rail Canada Inc,31 and Mission Institution v Khela.32 Parenthetically, however, there is an ongoing debate as to whether reviewing courts should accord deference to all tribunal rulings involving procedural fairness issues.33
29 See generally Toronto Newspaper Guild, Local 87 v Globe Printing Company, [1953] 2 SCR 18, [1953] 3 DLR 561; Saltfleet (Township) Board of Health v Knapman, [1956] SCR 877, 6 DLR (2d) 81.
30 Baker v Canada, [1999] 2 SCR 817 at para 27, 174 DLR (4th) 193 [Baker].
31 Council of Canadians with Disabilities v VIA Rail Canada Inc, 2007 SCC 15 at para 231, [2007] 1 SCR 650.
32 Mission Institution v Khela, 2014 SCC 24 at paras 79 and 89, [2014] 1 SCR 502.
33 See Paul Daly, “Canada’s Bi-Polar Administrative Law: Time for Fusion” (2014) 40(1) Queen’s LJ 213; and Christopher D Bredt & Alice Melkov “Procedural Fairness in Administrative Decision-Making: A Principled Approach to the Standard of Review” (2015) 28 Can J Admin L & Prac 1; and compare with John M Evans, “Fair’s Fair: Judging Administrative Procedures” (2015) 28 Can J Admin L & Prac 111.
154 UNBLJ RD UN-B [VOL/TOME 68
At this point, it is necessary to look to the actual scope of the four correctness categories identified in Dunsmuir. The most problematic of the categories is the one reserved for questions of law that are of central importance to the legal system as a whole and outside the tribunal’s field of expertise. A careful reading of Dunsmuir reveals that the Court was referring to the application of common law and equitable principles. That understanding is compatible with the pre-Dunsmuir jurisprudence and the case law cited in that case.34 However, Dunsmuir acknowledged room for the exception - those cases where the tribunal has developed a particular expertise in the application of a particular principle.
All of that said, the post-Dunsmuir jurisprudence is incompatible with my understanding of what was decided in Dunsmuir. Cases such as Nor-Man Regional Health Authority Inc v Manitoba Association of Health Care Professionals35 are on point. That was a case in which the Court held that the equitable remedy of estoppel did not qualify as a question of central importance to the law. The Court also held that labour tribunals are no longer bound to apply the common law and equitable principles in the same manner as superior courts as they have a “different mission”.36
Presumably, the same does not hold true with respect to other specialized tribunals. 2017 CanLIIDocs 175
Several Supreme Court cases have sought to provide guidance on how to identify questions that are of central importance to the legal system. None speak to the need for deference in the context of applying common law and equitable principles. Some of the cases may be perceived as more helpful than others.
The first in time is Canada (Canadian Human Rights Commission) v Canada (Attorney General).37 Therein, the Court acknowledged that questions involving human rights concepts should continue to attract a correctness standard (e.g., “family status” and “discrimination”). However, that observation is consistent with the fact that the decisions of human rights tribunals involving such issues, and that qualify as a question of law, have always been reviewed on the correctness standard.38 At the same time, the Court held that the issue of whether the tribunal could award “costs” to the successful party under a provision that allowed for reimbursement for “expenses” did not qualify as a question of central importance to the legal system.
34 See Toronto (City) v CUPE, Local 79, 2003 SCC 63, [2003] 3 SCR 77; Toronto (City) Board of Education v OSSTF, District 15, [1997] 1 SCR 487, 144 DLR (4th) 385.
35 Nor-Man Regional Health Authority Inc v Manitoba Association of Health Care Professionals, 2011 SCC 59, [2011] 3 SCR 616 [Nor-Man Regional].
36 Ibid at para 5. Query: How does a labour arbitrator’s “mission” differ from that any of any other decision-maker who is called upon to act in accordance with the rule of law?
37 Canada (Canadian Human Rights Commission) v Canada (AG), 2011 SCC 53, [2011] 3 SCR 471 [Canadian Human Rights Commission].
38 See discussion below infra at note 76.
2017] IDENTIFYING THE REVIEW STANDARD 155
The Supreme Court’s decision in Mouvement laïque québécois v Saguenay (City)39 also involved a decision of a human rights tribunal, only this time the Court did identify a question that was of central importance to the legal system. The issue was whether a municipal council's decision to recite a Christian prayer before the commencement of each meeting constituted a breach of the state's duty of neutrality and interfered with applicant's freedom of conscience and religion under the Quebec Charter. The majority of the Court ruled that the presumption of deference had been rebutted. That issue qualified as one of central importance and understandably so, regardless of how one casts the threshold test.
A more helpful insight or test for identifying questions of central importance to the legal system is found in Canadian National Railway Co v Canada (Attorney General).40 In obiter, the Court held that, if the tribunal’s interpretation of its home statute was of precedential significance, outside the ambit of the statutory scheme under consideration, the interpretative issue could qualify as one of central importance to the law.
Another relevant decision, and one that offers a more restrictive approach, 2017 CanLIIDocs 175 is Commission scolaire de Laval v Syndicat de l’enseignment de la region de Laval41 Therein, the Court held that: “Questions of [central importance] are rare and tend to be limited to situations that are detrimental to ‘consistency in the country’s fundamental legal order of our country.”42 The majority held the arbitrator's decision, requiring three members of the employer's executive committee be examined with respect to the reasons underlying the committee’s in-camera decision to dismiss an employee, was to be assessed on the standard of reasonableness. Simply put, the majority declared that the procedural and evidentiary issues at stake (deliberative secrecy) did not qualify as questions of central importance to the law. Frankly, the chances of identifying questions that are detrimental to consistency in Canada’s legal order seem remote and, for the person who seeks go-to answers, unrealistic.
Most recently, in Alberta (Information and Privacy Commissioner) v University of Calgary,43 the obiter in Canadian National Railway Co v Canada (AG) was adopted. The majority of the Court held that the question of whether s. 56(3) of the Freedom of Information and Protection of Privacy Act (Alberta) allows for the review of documents, over which solicitor-client privilege has been claimed, qualifies as a question of central importance to the legal system. The question of what statutory language is sufficient to authorize administrative tribunals to infringe
39 Saguenay, supra note 19.
40 Canadian National Railway Co v Canada (AG), 2014 SCC 40 at para 60, [2014] 2 SCR 135 [Canadian National Railway].
41 Commission scolaire de Laval v Syndicat de l’enseignment de la région de Laval, 2016 SCC 8, [2016] 1 SCR 29.
42 Ibid at para 34.
43 University of Calgary, supra note 22.
156 UNBLJ RD UN-B [VOL/TOME 68 solicitor-client privilege was held to be one that has potentially wide implications on other statutes. moment. At the same time, one cannot help but observe that the Supreme Court could have achieved the same result had it held the common law principles governing solicitor-client privilege attract review for correctness.
Cases in which a provincial court of appeal has identified a question of central importance to the legal sAystem have not met with success in the Supreme Court. Two decisions warrant consideration. The first is Communications, Energy and Paperworkers Union of Canada, Local 30 v Irving Pulp & Paper, Ltd.44 Therein, both the majority and minority rejected the Court of Appeal’s attempt to apply the correctness standard having regard to the public importance of the case. The subtle distinction between issues of central importance to the legal order and those that are of public importance or of importance in a particular area of the law is worth briefly exploring.45
Irving Pulp & Paper was a case in which the employer had exercised its rule-making authority under the collective agreement to require random alcohol testing of employees who held safety-sensitive positions within a kraft paper mill. 2017 CanLIIDocs 175 Everyone agreed the mill qualified as a “dangerous workplace.” At the same time, the employer’s policy limited testing to 34 random samplings in a calendar year. The majority of the arbitration panel declared the employer rule to be unreasonable on the ground the employer was unable to establish a “significant problem” with respect to alcohol-related impairment performance at the plant. The failure to meet that threshold meant that the employer was unable to justify the infringement of an employee’s privacy rights.
By the time Irving Pulp & Paper was heard in the Supreme Court, intervener status had been granted to 22 interested groups from across the country. There were those representing the major transportation companies, including Canada's two national railways, as well as employee associations and representatives of the oil and mining industries throughout Canada, yet the majority of the Supreme Court was unequivocal in its summary rejection of the correctness standard applying to the issue of random alcohol testing in dangerous work environments.
The majority’s reasoning in Irving Pulp & Paper is blunt and unambiguous: “It cannot be seriously challenged, particularly since Dunsmuir v New Brunswick that the applicable standard of review for reviewing the decision of a labour arbitrator is reasonableness.”46 The minority agreed: “This dispute has little legal consequence outside the sphere of labour law and that, not its potential real-world
44 Communications, Energy and Paperworkers Union of Canada, Local 30 v Irving Pulp & Paper, Ltd, 2013 SCC 34, [2013] 2 SCR 458, (reversing 2011 NBCA 58, 375 NBR (2d) 92) [Irving Pulp & Paper].
45 Parenthetically, and for the record, the Court of Appeal (Robertson JA) held that, in the alternative, the tribunal decision also failed to meet the deferential threshold of reasonableness.
46 Dunsmuir, supra note 2 at para 7.
2017] IDENTIFYING THE REVIEW STANDARD 157 consequences, determines the applicable standard of review.”47 In short, the Supreme Court held, that while the issue of random alcohol testing may be of utmost importance (interest) in the context of labour relations, the issue was not of central importance to the legal system.
Martin v Alberta (Workers’ Compensation Board)48 is another Supreme Court decision that adopted a narrow approach to questions of central importance to the legal system. In that case, the issue centered on the interpretation of federal legislation governing federal employees who are eligible to apply for workers’ compensation benefits under the various provincial schemes. The issue was whether a federal employee had to meet the provincial eligibility requirements of the province in which the employee was injured or whether the eligibility requirements were to be fixed under the federal legislation such that the requirements were uniform throughout the country.
The Supreme Court held that the interpretation which the Alberta tribunal placed on the federal legislation was owed deference. The federal statute qualified as a “home” or “constituent” statute for which reasonableness was the presumptive 2017 CanLIIDocs 175 standard of review. More importantly, the question of law to be decided was held not to be of central importance to the legal system and squarely within the specialized functions of such tribunals. This was held to be so even though the Supreme Court’s ruling meant that it was possible for the provincial tribunals to adopt competing interpretations of the federal legislation. That possibility bears on what many believe to be a glaring deficiency in the law of deference. The Supreme Court has held that inconsistency in tribunal decisions is not an independent ground for moving to review for correctness. This point is worth exploring.
Take the case where a tribunal panel interprets a provision of its home statute in one manner, and the next panel adopts a conflicting interpretation of the same provision. Assume also that only the second panel decision is subjected to judicial review. Does the reviewing court owe deference to the second panel's interpretation or is the court free to resolve the conflict by adopting the review standard of correctness? The short answer is that the deferential review standard must be applied to the second tribunal decision. A brief explanation is warranted.
In Domtar Inc v Quebec (Commission d’appel en matière de lésions professionnelles),49 a unanimous Supreme Court, recognized that, although the requirement of consistency in the law was a valid objective, it could not be separated from the autonomy, expertise, and effectiveness of specialized tribunals. Prior to Domtar, academic commentators believed otherwise. However, the Court consciously chose a different path. As Professor Hawkins observed: “L'Heureux-
47 Ibid at para 66.
48 Martin v Alberta (Workers' Compensation Board), 2014 SCC 25, [2014] 1 SCR 546.
49 Domtar Inc v Quebec (Commission d’appel en matière de lésions professionnelles), [1993] 2 SCR 756, 105 DLR (4th) 385.
158 UNBLJ RD UN-B [VOL/TOME 68
Dubé J. was not prepared to compromise the principle of deference even given the argument that a ‘primary purpose of judicial review was to prevent arbitrariness.’”50
For those troubled by the understanding that inconsistent tribunal decision- making is not a sufficient ground for moving to the correctness standard of review, it is worth revisiting two other Supreme Court decisions: UES, Local 298 v Bibeault,51 and Ivanhoe Inc v United Food and Commercial Workers’ Local 500.52 Those decisions reveal that, on occasion, the Supreme Court has skirted the Domtar ruling.53
The next category for which correctness has been deemed the proper standard of review embraces constitutional rulings. No one quibbles with the requirement that such rulings, including those requiring the application of the Charter of Rights and Freedoms, must fall outside the deference obligation. Intuitively, the rule of law, however formulated, dictates that correctness must be the proper review standard. However, if one carefully sifts through the Supreme Court jurisprudence, it is possible to isolate the anomalous case.
2017 CanLIIDocs 175 In Doré v Barreau du Québec,54 the Court accorded deference to a disciplinary decision that impacted on a lawyer’s right to freedom of speech under the Charter. This was the case in which Maître Doré had sent an insulting letter, to the presiding judge, immediately following a court hearing. The lawyer was disciplined notwithstanding his plea that the letter fell within his right to freedom of speech. The Supreme Court held the disciplinary committee was no longer under an obligation to apply the Oakes test when dealing with s. 1 of the Charter (for which correctness would have been the proper review standard). The “new” administrative law approach requires the tribunal to balance the severity of the interference with the Charter protected right and the objectives of the tribunal’s home statute. As well, the Court held the committee’s ultimate ruling was owed deference. In so holding, the Court effectively overruled Slaight Communications Inc v Davidson,55 and Multani v Commission scolaire Marguerite-Bourgeoys.56
The next category for which correctness has been deemed the proper standard of review embraces the “who-gets-to-decide” cases. These are the ones in
50 RE Hawkins, “Whither Judicial Review?” (2009) 88 Can Bar Rev 603 at 631.
51 UES, Local 298 v Bibeault, [1988] 2 SCR 1048, 35 Admin LR 153 [Bibeault].
52 Ivanhoe Inc v United Food and Commercial Workers’ Local 500, 2001 SCC 47, [2001] 2 SCR 565.
53 With respect to tribunal inconsistency in decision making in New Brunswick, see Jones' Masonry Ltd v Labourers' International Union of North America, Local 900, 2013 NBCA 50, 364 DLR (4th) 4 (leave to appeal refused, [2013] SCCA No 356, 469 NR 396) and compare the majority and dissenting opinions.
54 Doré v Barreau du Québec, 2012 SCC 12, [2012] 1 SCR 395.
55 Slaight Communications Inc v Davidson, [1989] 1 SCR 1038, 59 DLR (4th) 416.
56 Multani v Commission scolaire Marguerite-Bourgeoys, 2006 SCC 6, [2006] 1 SCR 256.
2017] IDENTIFYING THE REVIEW STANDARD 159 which the reviewing court is asked to decide who has the jurisdiction to decide a particular issue. The jurisdiction may rest exclusively with the tribunal, a superior court or another tribunal. Correlatively, concurrency in jurisdiction may exist. For example, it has been asked whether an arbitrator possesses the jurisdiction to rule on the issue of employer discrimination or whether the jurisdiction rests with a human rights tribunal. Alternatively, the jurisdiction may be concurrent.57
There are also cases in which a tribunal has been asked to determine whether it possesses the jurisdiction to rule on the constitutional validity of a provision of its home statute. A negative response means the issue must be referred to the Court of Queen’s Bench. In fact, the Supreme Court has provided a qualified response. A tribunal may address the constitutional issue provided it has either the express or implied right to decide questions of law. However, and as explained earlier, even if the tribunal possesses the jurisdiction to decide the issue, the constitutional ruling must be reviewed on the standard of correctness.58
Finally, it must be acknowledged that, in those cases where two tribunals possess the concurrent jurisdiction to decide an issue and one of those tribunals has 2017 CanLIIDocs 175 already made a determination, the second tribunal will be forced to deal with the matter of issue estoppel. In other words, the second tribunal will be forced to defer to the ruling of the first tribunal, unless the tenets of the estoppel doctrine dictate otherwise.
This leaves us with the final category for which correctness has been deemed to be the proper review standard: the true jurisdictional question. It was 1983 when Professor MacLauchlan (now Premier of PEI) made an incisive observation about the law surrounding the concept of jurisdiction: “I currently favour likening thinking about jurisdiction to attempting to extract oneself from fly-paper; once you get started with the exercise it is virtually impossible to break free.”59 Admittedly, too much ink has been spilled since New Brunswick Liquor in trying to articulate a
57 See generally Dunsmuir, supra note 2 at para 61; Regina Police Assn Inc v Regina (City) Board of Police Commissioners, 2000 SCC 14, [2000] 1 SCR 360; Quebec (Commission des droits de la personne et des droits de la jeunesse) v Quebec (AG), 2004 SCC 39, [2004] 2 SCR 185; Tranchemontagne v Ontario (Director, Disability Support Program), 2006 SCC 14, [2006] 1 SCR 513 [Tranchemontage]; Syndicat des professeurs du Cégep de Sainte-Foy v Beaulieu (AG), 2010 SCC 29, [2010] 2 SCR 123; Alberta (Information and Privacy Commissioner) v Alberta Teachers’ Association, 2011 SCC 61, [2011] 3 SCR 654 [Alberta Teachers’ Association]; and British Columbia (Workers’ Compensation Board) v Figliola, 2011 SCC 52, [2011] 3 SCR 422.
58 See Westcoast Energy Inc v Canada (National Energy Board), [1998] 1 SCR 322, 156 DLR (4th) 456; Nova Scotia (Workers’ Compensation Board) v Martin, 2003 SCC 54, [2003] 2 SCR 504; Tranchemontagne, supra note 57, and R v Conway, 2010 SCC 22, [2010] 1 SCR 765.
59 Wade MacLauchlan, “Judicial Review of Administrative Interpretations of Law: How Much Formalism Can We Reasonably Bear?” (1986) 36 UTLJ 343 at note 3. Curiously, the Supreme Court no longer characterizes the “who gets to decide cases,” just discussed, as falling within this category. Fortunately, the oversight, if it is such, has no substantive impact on the deference doctrine. Above all, the oversight should not detract from the following analysis.
160 UNBLJ RD UN-B [VOL/TOME 68 legal framework for isolating the true jurisdictional question from the non- jurisdictional one.
At one time, administrative law embraced what was labeled the “preliminary and collateral approach” to defining jurisdictional questions. Applying that approach, it was all too easy to declare the issue at hand qualified as a true jurisdictional question and, therefore, correctness was the proper review standard. Take, for example, those cases where the tribunal had to decide whether an applicant had filed its review application within the time prescribed by the enabling statute. As neither the interpretative nor factual issue went to the merits of the underlying case, those issues were classified as preliminary questions and, therefore, correctness was the proper review standard.
Better still, take the case where a labour board must decide whether the affected person is an employee or an independent contractor under the enabling legislation. Of course, independent contractors fall outside the protections of the legislation. However, as the issue involved a preliminary determination that did not go to the underlying merits of the case, it would have been treated as a true 2017 CanLIIDocs 175 jurisdictional question under the preliminary and collateral framework.
The habit of isolating preliminary questions was also adopted regarding collateral questions. Such questions were collateral in the sense that the tribunal was being asked to rule on whether it possessed the jurisdiction to grant certain relief to the successful party or to impose certain sanctions against the unsuccessful one. For example, a tribunal may have concluded that one of the parties was in breach of its statutory obligations and, therefore, the successful party was entitled to relief, but a disagreement as to whether the tribunal possessed the jurisdiction to grant the relief contemplated. This type of disagreement would have been classified as collateral to the underlying dispute and, therefore, would have qualified as a true jurisdictional question under the “preliminary and collateral” framework.
Too many cases involving preliminary and collateral questions made their way to the Supreme Court. This is true even though the Supreme Court formally abandoned that approach, in 1988, with the release of UES, Local 298 v Bibeault.60 In that case, the Court adopted the “pragmatic and functional approach” to identifying jurisdictional questions. (Parenthetically, ten years later the Court would adopt that approach for purposes of identifying the proper standard of review regarding all issues.) Regrettably, the jurisprudence continued to produce conflicting decisions on whether the remedial powers of a tribunal fell within the true jurisdictional category, even with the application of the pragmatic and functional approach. Fortunately, the Court’s more recent jurisprudence brings certainty to the law.
60 Bibeault, supra note 51.
2017] IDENTIFYING THE REVIEW STANDARD 161
In Canada (Canadian Human Rights) v Canada (Attorney General),61 the Supreme Court held the authority of a human rights tribunal to award costs to the successful party was not a true jurisdictional question and, therefore, the tribunal’s decision was owed deference. And in Alberta (Information and Privacy Commissioner) v Alberta Teachers’ Association,62 the Court was unanimous in holding that a question of law dealing with the interpretation and application of a provision governing a limitation period did not qualify as a true jurisdictional question.
It should not be forgotten that the precedential significance of Alberta Teachers’ Association is greater than the narrow issue decided therein. A divided Supreme Court left unanswered whether there is a need to retain the category of true jurisdictional question. The majority could not identify a workable definition for the concept and, therefore, questioned the need for its retention. The minority argued for its preservation on historical grounds, and so a compromise of sorts was reached. In the future, there would be a presumption of deference to a tribunal’s interpretative rulings with respect to its home statute and those with which it has familiarity.
2017 CanLIIDocs 175 Those arguing that a tribunal’s interpretative ruling qualifies as a true jurisdictional question will now have to rebut the presumption of deference. How is that accomplished? With great difficulty unless the issue just happens to fall within one of the other correctness categories identified in Dunsmuir. While the pragmatist might allege circuitous reasoning on my part, we are left with a stark reality. There is little, if any, room left in administrative law for the concept of the true jurisdictional question. Arguably, the only true jurisdictional questions are those raising a constitutional issue or, correlatively, those requiring the reviewing court to determine which of two or more tribunals possesses the authority to decide a particular issue.63
In summary, the go-to standard of review is the deferential one. Unless the issue falls within one of the four (five) correctness categories, the presumptive standard of reasonableness applies. Pragmatically speaking, there is little hope of anyone identifying a true jurisdictional question so long as the deferential standard of review presumptively applies to the tribunal’s interpretations of its enabling statute and the only practical way of doing that is to how that the issue falls within one of the correctness categories identified in Dunsmuir. However, the administrative lawyer and reviewing court will encounter difficulty in identifying questions that are of central importance to the legal system and in assessing whether the issue falls outside the tribunal’s relative expertise. The tests the Supreme Court has articulated are not easy to apply and unlikely to promote consistency in the law.
61 Canadian Human Rights Commission, supra note 37.
62 Alberta Teachers’ Association, supra note 57.
63 This is dealt with in Robertson, supra note 3 at 38.
162 UNBLJ RD UN-B [VOL/TOME 68
The Segmentation Issue
To this point, the presentation might leave the impression that a tribunal’s decision is subject to only one standard of review. But this is not so. For example, the tribunal may make several factual rulings, together with a ruling on a procedural matter, followed by an interpretation of one the provisions of the tribunal’s enabling (home) statute. Collectively, those rulings support the tribunal’s ultimate disposition. What matters is that “segmentation” or “disaggregation” of a tribunal decision may result in the reasonableness standard applying to some issues while the correctness standard applies to others.
Parenthetically, segmentation of a tribunal decision has its antagonists. In the Supreme Court, Abella J has consistently maintained that a tribunal decision should be subjected to only one review standard. While the weight of Supreme Court jurisprudence does not embrace that view, it would be remiss not to acknowledge the existence of two conflicting decisions, released within a day of one another, dealing with the segmentation issue: Council of Canadians with Disabilities v VIA Rail 64 65 Canada Inc, and Lévis (City) v Fraternité des policiers de Lévis Inc. The first 2017 CanLIIDocs 175 decision rejected outright the notion of segmentation. The second embraced it wholeheartedly. However, the Supreme Court jurisprudence predating both decisions is consistent with the understanding that segmentation is permissible.66
In any event, there are two more recent decisions of the Supreme Court that expressly reject Abella J’s insistence that segmentation of a tribunal decision be proscribed: Mouvement Laïque Québécois v Saguenay (City),67 and Canada Broadcasting Corp v SODRAC 2003 Inc.68 In the latter case, Abella J, in dissent, conveniently and succinctly expressed her reasons underscoring her opposition to segmentation in the following manner: “Breaking down a decision into each of its component parts also increases the risk that a reviewing court will find an error to justify interfering in the tribunal’s decision, and may well be seen as a thinly veiled attempt to allow reviewing courts wider discretion to intervene in administrative decisions.”69
64 Council of Canadians with Disabilities v VIA Rail Canada Inc, 2007 SCC 15, [2007] 1 SCR 650.
65 Lévis (City) v Fraternitié des policiers de Lévis Inc, 2007 SCC 14, [2007] 1 SCR 591.
66 As to the earlier jurisprudence, see Canada (Deputy Minister of National Revenue) v Mattel Canada Inc, 2001 SCC 36 at para 39, [2001] 2 SCR 100; Law Society of New Brunswick v Ryan, 2003 SCC 20 at para 41, [2003] 1 SCR 247. It also bears noting that Abella J continues to insist that segmentation of tribunal decisions should not be permitted. In Irving Pulp & Paper, supra note 44 at para 54, she emphasized that a tribunal's decision should be approached as “an organic whole” and “without a line-by- line treasure hunt for error.”
67 Saguenay, supra note 19.
68 Canada Broadcasting, supra note 21.
69 Ibid at para 191. A response to Abella J’s objections to segmentation can be found in Robertson, supra note 3.
2017] IDENTIFYING THE REVIEW STANDARD 163
The “Palpable & Overriding Error” Threshold
Putting aside the issue of segmentation, there remains another pressing question. Accepting that questions of fact and questions of mixed law and fact are owed deference on the review standard of “reasonableness”, one cannot help but ask whether that deferential standard is equivalent to the “palpable and overriding error” standard being applied in the context of civil appeals under the principles articulated in Housen v Nikolaisen,70 and HL v Canada (Attorney General).71 The short answer is “no”.
In Agraira v Canada (Public Safety and Emergency Preparedness),72 the Supreme Court cautioned that the appellate standards of “correctness” and “palpable and overriding error” and the administrative law standards of “correctness” and “reasonableness” are not interchangeable. More recently Mouvement Laïque Québécois v Saguenay (City),73 the Court held that the palpable and overriding test had no application in the context of administrative appeals. That being so, it should follow that the threshold test for setting aside a finding of fact, or a finding of mixed law and fact, is different in the context of civil appeals (palpable and overriding 2017 CanLIIDocs 175 error) than it is in the context of judicial review by way of appeal (reasonableness). However, one is at a loss to explain how in practical terms the two deferential standards differ.
Other Statutory Delegates
This presentation does not account for those cases in which the tribunal is chaired by a sitting judge.74 Nor does it take into account the unique treatment which the Supreme Court has accorded to the decisions of human rights tribunals. Historically, the Court has consistently applied the correctness standard to tribunal rulings that qualify as questions of law and, in particular, those involving, for example, the definition of discrimination in one of its various manifestations.75 This
70 Housen v Nikolaisen, 2002 SCC 33, [2002] 2 SCR 235.
71 HL v Canada (AG), 2005 SCC 25, [2005] 1 SCR 401.
72 Agraira v Canada (Public Safety and Emergency Preparedness) 2013 SCC 36 at para 45 [Agraira].
73 Saguenay, supra note 19.
74 See generally Robertson, supra note 3 at 90.
75 As to the pre-Saguenay jurisprudence, see Canadian Human Rights Commission, supra note 37 and the earlier jurisprudence beginning with Ross v New Brunswick School District No 15, [1996] 1 SCR 825, 171 NBR (2d) 321, which deals with a question of fact, and cases such as Gould v Yukon Order of Pioneers, [1996] 1 SCR 5571, 18 BCLR (3d) 1, which consolidates the earlier precedents including University of British Columbia v Berg, [1993] 2 SCR 353, 79 BCLR (2d) 273; Dickason v University of Alberta, [1992] 2 SCR 1103, 95 DLR (4th) 439; Zurich Insurance Co v Ontario (Human Rights Commission), [1992] 2 SCR 321, 9 OR (3d) 224; and Canada (AG) v Mossop, [1993] 1 SCR 554, 100 DLR (4th) 658.
164 UNBLJ RD UN-B [VOL/TOME 68 understanding, however, is difficult to reconcile with the Court’s recent decision in Mouvement laïque québécois v Saguenay (City).76
As noted earlier, in Saguenay (City), the Court was dealing with a decision of the Quebec’s Human Rights Tribunal involving the state’s duty of “religious neutrality” that flows from freedom of conscience and religion. Therein, the Court accepted that the proper standard of review regarding that issue was correctness,77 but it did so on the basis that the issue was of central importance to the legal system. No reference was made to the substantial body of earlier Supreme Court jurisprudence holding that correctness is the proper review standard for questions of law that are decided by a human rights tribunal. Despite that omission it seems as though the Court is moving to a deference doctrine in which human rights tribunals are to be placed on the same dais as other specialized tribunals.78
To this point, the presentation has focused on the application of the deference doctrine with respect to the adjudicative decisions of specialized tribunals. This leaves for consideration the doctrine’s application in the context of administrative decisions made by other statutory delegates. The list includes 2017 CanLIIDocs 175 delegates such as government Ministers, civil servants that act as sub-delegates, and those who hold office under a statutory regime (e.g., Registrar of Land Titles).
This part of my presentation is important to those practicing administrative law in New Brunswick. This is because there is recent Supreme Court jurisprudence that effectively overrules three earlier precedents of the New Brunswick Court of Appeal. O’Dell v New Brunswick (Minister of the Environment and Local Government);79 Greenisle Environmental Inc. v New Brunswick (Minister of the Environment and Local Government);80 Carter Brothers Ltd v New Brunswick (Registrar of Motor Vehicles).81 Parenthetically, in Hovey v Registrar General of Land Titles,82 Walsh J noted the potential conflict between the decisions of the Court of Appeal and the three Supreme
76 Saguenay, supra note 19.
77 Ibid at para 49.
78 See Canadian Human Rights Commission, supra note 37, and Saskatchewan (Human Rights Commission) v Whatcott, 2013 SCC 11, [2013] 1 SCR 467.
79 O’Dell v New Brunswick (Minister of the Environment and Local Government), 2005 NBCA 58, 286 NBR (2d) 115.
80 Greenisle Environmental Inc v New Brunswick (Minister of the Environment and Local Government), 2007 NBCA 9, 311 NBR (2d) 161.
81 Carter Brothers Ltd v New Brunswick (Registrar of Motor Vehicles), 2011 NBCA 81, 377 NBR (2d) 29. In the same vein, see Takeda Canada Inc v Canada (Minister of Health), 2013 FCA 13, [2014] 3 FCR 70 (Stratas JA in dissent); and Prescient Foundation v Canada (Minister of National Revenue), 2013 FCA 120 at para 13, 358 DLR (4th) 541.
82 Hovey v Registrar General of Land Titles, 2014 NBQB 118, 420 NBR (2d) 201.
2017] IDENTIFYING THE REVIEW STANDARD 165
Court cases to be discussed. However, on the facts, he was not required to address the issue.
Between 2005 and 2011, the New Brunswick Court of Appeal had consistently held that a statutory delegate, such as a Minister of the Crown, is owed no deference when it comes to the interpretation of his or her home statute. Correlatively, the Court was not prepared to grant deference to the interpretative decisions rendered by government officers such as the Registrar of Motor Vehicles and the Registrar of Land Titles. Otherwise, the Court would be granting deference to government lawyers who provide legal advice to statutory delegates under the guise of “institutional expertise.” However, if the decision under review involved, for example, the exercise of Ministerial discretion, deference is required in accordance with the Supreme Court’s decision in Baker v Canada.83 Leaving aside that type of case, there are four recent Supreme Court cases which support the contention that the New Brunswick jurisprudence has been overtaken. The Supreme Court precedents were released between 2013 and 2016.
84 In Agraira v Canada (Public Safety and Emergency Preparedness), the 2017 CanLIIDocs 175 Federal Court of Appeal had reviewed the federal Minister’s interpretation of the term “national interest” on the standard of correctness, while according deference to the Minister’s application of that interpretation to the facts of the case (a question of mixed law and fact). On further appeal, the Supreme Court applied the first step of the Dunsmuir framework to hold that the Minister’s interpretation was to be assessed on the deferential standard of reasonableness. Why? The Court answered: […] because such a decision involves the interpretation of the term “national interest” in s. 34(2), it may be said that it involves a decision maker “interpreting its own statute or statutes closely connected to its function, with which it will have particular familiarity.”85
In Canadian National Railway Co. v Canada (Attorney General),86 one of the issues was whether s. 40 of the Canadian Transportation Act vested the Governor in Council with authority to vary or rescind a decision of the Canadian Transportation Agency on a point of law. The Governor in Council (federal cabinet) had so concluded. In upholding that interpretative decision, the Supreme Court held the standard of review analysis set out in Dunsmuir applies to the interpretative decisions of Cabinet. The Court’s reasoning is oracular in nature: “Dunsmuir is not limited to judicial review of tribunal decisions”.87 The decision of the Federal Court of Appeal in Public Mobile Inc v Canada (Attorney General),88 was the only
83 Baker, supra note 30.
84 Agraira, supra note 72.
85 Quoting Dunsmuir, supra note 2 at para 54.
86 Canadian National Railway, supra note 40.
87 Ibid at para 53.
166 UNBLJ RD UN-B [VOL/TOME 68 decision cited in support of that proposition.89 The Supreme Court went on to hold that, as the interpretative issue did not fall within one of the categories for which correctness is automatically the proper review standard, the Council’s interpretative decision had to be assessed on the standard of reasonableness.
The Supreme Court’s decision in Wilson v British Columbia (Superintendent of Motor Vehicles)90 is consistent with the Court’s two earlier rulings. In that case, the Court upheld the interpretation which the Superintendent of Motor Vehicles had placed on a provision of the Motor Vehicle Act. Curiously, the deferential standard of review was chosen by analogy to the standard of review applicable to specialized tribunals when interpreting the provisions of their enabling legislation.
Collectively, the rulings in Agraira, Canadian National Railway and Wilson recognize that the deference doctrine has been extended to embrace all administrative decision-makers and not just those falling within the category of specialized tribunals. The difficulty I have is that the extension is based on a false premise: namely, all statutory delegates possess a relative expertise when it comes to 2017 CanLIIDocs 175 interpreting statutes. Indeed, no one has been as critical of this aspect of the deference doctrine than Professor Daly. His influential administrative law blog offers an incisive analysis with respect to a Federal Court of Appeal decision in which the review standard of reasonableness was applied to the decision made by a frontline immigration officer with no legal training: MPSEP v Tran.91
The Federal Court decision involved the interpretation of the Immigration and Refugee Protection Act, a pure question of law, by those with no apparent legal training.92 While Professor Daly’s comment merits reading in its entirety, one need only reproduce his closing observations to reinforce the understanding that the Supreme Court precedents mentioned earlier are, to say the least, troubling decisions:
But if Tran is right, then deference is due to decision-makers who have no legal expertise, who do not address relevant arguments expressly in their reasons, and who may reasonably come to diametrically opposed conclusions as to similarly situated individuals. And the courts cannot intervene to resolve the issues authoritatively even though there is a strong
88 Public Mobile Inc v Canada (AG), 2011 FCA 194, [2011] 3 FCR 344.
89 Parenthetically, and with respect, I do not read that decision as establishing that the interpretative decisions of the Governor in Council are owed deference.
90 Wilson v British Columbia (Superintendent of Motor Vehicles), 2015 SCC 47, [2015] 3 SCR 300.
91 Canada (Public Safety and Emergency Preparedness) v Tran, 2015 FCA 237, 392 DLR (4th) 351.
92 Professor Daly’s case comment is instructive: Paul Daly, “A Snapshot of What’s Wrong with Canadian Administrative Law: MPSEP v Tran 2015 FCA 237” (13 November 2015), Administrative Law Matters (blog), online:
2017] IDENTIFYING THE REVIEW STANDARD 167
indication that parliament intended for them to do so. Somewhere along the line, something has gone rather badly wrong.93
However, shorty after the release of the Federal Court’s decision in Tran, the Supreme Court released yet another decision that affirms the understanding that the interpretative decisions of all statutory delegates are owed deference. This is true even though the delegate may have no expertise in legal matters. The decision is Kanthasamy v Canada (Citizenship and Immigration).94
In Kanthasamy, the majority adopted the deferential review standard of reasonableness with respect to an issue of law involving s. 25(1) of the Immigration and Refugee Protection Act as interpreted by the Minister and a front-line immigration officer. This is so despite the fact that an earlier Supreme Court decision had adopted the standard of correctness in similar circumstances: Pushpanathan v Canada (Minister of Citizenship and Immigration).95 Regrettably, that decision did not make its way into the majority opinion, but there was something else that did not make its way into the reasons for judgment. Professor Daly notes that, during oral
argument before the Supreme Court, counsel emphasized the lack of legal expertise 2017 CanLIIDocs 175 of front-line immigration officers, yet there is no mention in the Court’s reasons of that argument. Professor Daly laments: “Another week, another underwhelming standard-of-review from the Supreme Court of Canada…” The social media views of others are even less kind!96
I end this presentation with three unvarnished questions: (1) what policy reasons justify deference to a civil servant’s interpretation of enabling legislation in circumstances where the delegate obviously lacks legal training; (2) why should legal counsel within the office of the Attorney General or other departmental lawyers be entitled to raise the plea of institutional expertise; (3) whatever happened to the understanding that citizens are entitled to independent and impartial decision- makers?
Of course the administrative lawyer, in search of go-to answers, knows there is no need to search for nutshell answers to my questions. They are irrelevant now that contextualism has been abandoned. The same person also knows that the New Brunswick jurisprudence, identified above, is no longer binding.
93 Ibid.
94 Kanthasamy v Canada (Citizenship and Immigration), 2015 SCC 61, [2015] 3 SCR 909.
95 Pushpanathan, supra note 6.
96 See Paul Daly, “Can This Be Correct? Kanthasamy v. Canada (Citizenship and Immigration) 2015 SCC 61” (11 December 2015), Administrative Law Matters (blog), online:
168 UNBLJ RD UN-B [VOL/TOME 68
Conclusion
Accepting that New Brunswick Liquor represents the genesis of Canada’s modern deference doctrine, it has taken our Supreme Court nearly forty years to craft a doctrine that others say is infused with incoherence and inconsistency. Fortunately, for the administrative lawyer and reviewing court in search of go-to answers, the evolution of the doctrine is not disappointing when it comes to identifying the proper review standard. It resonates with simplicity. In a nutshell: the decisions of statutory delegates are owed deference unless the issue falls within one of the correctness categories identified in Dunsmuir or unless the issue involves a breach of the fairness duty.
Even inconsistency in tribunal decision-making is not a valid reason for moving to review for correctness. Better still, the deference obligation no longer hinges on what the legislature may have intended and, hence, the presence of a privative clause or a right of appeal within the enabling or home legislation is no longer of any moment. The same holds true regarding whether the administrative decision-maker possesses a relative expertise with respect to the issue at hand. 2017 CanLIIDocs 175 Ironically, persons sitting as lay experts on a specialized tribunal are entitled to deference when it comes to assessing their interpretative decisions. Those realties flow naturally from the death of contextualism.
The administrative lawyer in search of go-to answers knows that there is little chance of identifying a true jurisdictional question now that deference is required of interpretative decisions. And those who go searching for questions of central importance to the legal system will struggle with the Supreme Court’s threshold tests. Perhaps time is better spent on the question not addressed during this presentation: What renders a tribunal decision unreasonable? Till the next time.
BUNGLED POLICE EMERGENCY CALLS AND THE PROBLEMS WITH UNIQUE DUTIES OF CARE
Bruce Feldthusen*
1. Introduction
In Michael v Constable of South Wales the Supreme Court of the United Kingdom (‘UKSC’) upheld the striking out of a negligence action brought by the estate of a murdered victim of domestic violence.1 Ms. Michael’s ex-partner, Williams, discovered her in bed with another man. Williams hit her, left to take the other man into town, and told her he would return to kill her. Michael made an emergency telephone call to the police. Her call was misrouted to the neighbouring county and answered by a police operator, Ms. Mason. Michael described the attack and told Mason that Williams was going to kill her. Mason told Michael she would notify the 2017 CanLIIDocs 175 police force in Michael’s area. She logged the call as “Grade 1,” which meant a response within 5 minutes was required. However, when Mason contacted Gould, the police operator in Michael’s area, she neglected to mention that Michael was in fear for her life. Gould therefore logged the call as “Grade 2,” which meant a response within an hour. Michael called a second time about 15 minutes after her first call. There were screams on the line and then the call ended. The event was then upgraded to “Grade 1.” The police arrived at Michael’s home 22 minutes after her first call and discovered that Williams had brutally stabbed Michael to death. Had the police not bungled her first call, it seems likely that the claimants could have established that the police would have been able to save Michael’s life. Williams pleaded guilty to murder and was sentenced to life imprisonment. Mason and Gould faced disciplinary action. The Independent Police Complaints Commission issued a report strongly criticizing Mason for breaching internal policy by failing to obtain critical information from Michael. The police force in Michael’s area was criticized for failing to respond immediately upon receiving the report from Mason, given that so much critical information was missing.2 Nevertheless, in a 5-2 decision the UKSC dismissed an action in negligence, brought on behalf of Michael’s parents and children, seeking damages against the Chief Constables of both counties.
The majority in Michael displayed little interest in Ms. Michael’s experience. Instead, it focussed on affirming a fundamental principle of UK law: that
* Professor, University of Ottawa Faculty of Law.
1 Michael v The Chief Constable of South Wales Police, [2015] UKSC 2, [2015] AC 1732 [Michael].
2 United Kingdom, Independent Police Complaints Commission, Independent Investigation into Police Contact with Joanna Michael prior to her death, online:
170 UNBLJ RD UN-B [VOL/TOME 68 a common law duty of care in negligence can never be founded on a statutory duty or power alone. Michael holds that the courts should not recognize unique public duties of care. A unique public duty is one that is imposed on governments or other public defendants where no such duty would be imposed on a private party in the same or an analogous situation.3 Instead, the liability of public defendants must be based on the application of ordinary private law principles. McBride calls this the Diceyan principle.4 Public actors should be “under the same (emphasis added) law that applies to private citizens,” a principle Dicey called the “idea of equality.”5
Like the Michael decision, this article does not deal primarily with the social problem of domestic violence. It does shed some light on the question of police responsibilities to potential victims of crime who reach them on emergency hotlines. However, the primary focus is on unique duties of care, with Michael serving as a provocative background against which to evaluate the arguments. I believe that Ms. Michael’s family was entitled to a remedy in tort. I do not believe it would be necessary in Canada to create a unique public duty to provide one.
6 Public defendants owe the same duties of care as do private citizens. In 2017 CanLIIDocs 175 Section 2, I will review the basic law of negligence pertaining to the failure of one private party to confer a benefit on another.7 The general rule is that one private party
3 Hill v Hamilton-Wentworth Regional Police Services Board, 2007 SCC 41, [2007] 3 SCR 129 [Hill] is an example of a duty of care that may only be owed by the police, but is nevertheless analogous to private party duties. The court recognized a duty of care owed by an investigating police officer to a suspect in a criminal investigation. The relationship between the parties is analogous to other special relationships of control and vulnerability between private parties where exceptional affirmative duties of care have been recognized. See 175–177, below.
4 Nicholas McBride, “Michael v Chief Constable of South Wales Police [2015] UKSC 2” (2015) University of Cambridge Legal Paper Research Series, Paper No. 21/2015 at 5, online:
5 Peter Hogg, Patrick Monahan & Wade Wright, Liability of The Crown, 4th ed (Toronto: Carswell, 2011) at 218–19 citing Albert Venn Dicey, The Law of the Constitution, 10th ed (London: McMillan, 1959) at 193.
6 However, the Canadian public defendant may enjoy an immunity from liability if the allegation of negligence concerns core government policy. See Just v British Columbia, [1989] 2 SCR 1228, 64 DLR (4th) 689 [Just] and R v Imperial Tobacco Canada, 2011 SCC 42, [2011] 3 SCR 45 [Imperial Tobacco]. Immunity is different than an objection to a unique public duty. See 172, below. It can be argued that immunizing public defendants from ordinary negligence liability is as objectionable as subjecting them to unique duties. See Bruce Feldthusen, “Public Authority Immunity from Negligence Liability: Uncertain, Unnecessary, and Unjustified” (2014) 92 Can Bar Rev 211.
7 The defendant does not perform an act that causes harm. Rather the defendant fails to prevent harm or fails to provide other benefits. This is described variously; e.g. as potential liability for nonfeasance, for omissions, or for a failure to take positive action. See Childs v Desormeaux, 2006 SCC 18 at paras 31–32, [2006] 1 SCR 643 [Childs]. Examples include duties to rescue, duties to warn, duties to protect, and duties to control. See also 195, below.
2017] BUNGLED POLICE EMERGENCY CALLS 171 does not owe an affirmative duty to confer a benefit upon another.8 There are numerous exceptions to this “no duty” rule. Canadian courts probably recognize a broader range of exceptions, and apply them less strictly than do courts in the UK.
Section 3 suggests that Michael would have been decided differently in Canada,9 and possibly should have been decided differently in the UK, based on basic private party negligence law. The Michael claim ought to have been allowed to proceed to trial. There is a sound case that the claimants could have established that the police assumed responsibility to Ms. Michael. If necessary, they might also have been able to establish that she relied on the police to her detriment. This is an important conclusion because it demonstrates that basic negligence law is not as impotent in the face of domestic abuse as Michael suggests it is in the UK. It also shows that basic negligence law can take into account unique aspects of government conduct without creating unique public duties of care.10
Sections 4 and 5 consider the alternative argument: assuming that the facts will not support a duty in private party negligence law, when, if ever, ought the law 11 to recognize unique public duties. Section 4 considers arguments in favour of 2017 CanLIIDocs 175 unique duties that are grounded in what McBride calls the “policy approach” derived from the Anns case.12 Characteristic of this approach is a presumption that government owes a duty to provide benefits to its citizens at a standard of reasonable care unless there are good reasons to deny or limit the duty.13 McBride notes that the
8 The classic authority is Osterlind v Hill, 263 Mass 73 (1928), 160 NE 301. The defendant rented a canoe to an intoxicated customer. He then ignored the customer’s cries for help when the canoe tipped. The court held that he did not owe a duty to rescue the plaintiff. This case would probably be decided differently today in Canada, based on a special relationship exception. See Childs, ibid at paras 38–40.
9 A Canadian case somewhat similar to Michael is Mooney v British Columbia (AG), 2001 BCSC 419, [2001] BCLWD 913, aff’d 2004 BCCA 402, 31 BCLR (4th) 61, leave to appeal to SCC refused, Mooney v Canada (AG) (3 March 2005), No 30546 [Mooney]. Mooney formally reported to the police credible threats of violence by her ex-partner. The police did nothing. Forty-seven days later he broke into her house, killing her friend and injuring her daughter. At trial a duty of care was recognized on the part of police to protect Mooney. On appeal the case was dismissed on the issue of causation. The question of unique duties of care was not discussed explicitly. See Margaret I Hall, “Duty, Causation, and Third-Party Perpetrators: The Bonnie Mooney Case” (2005) 50 McGill LJ 597; Elizabeth Sheehy, Defending Battered Women on Trial: Lessons from the Transcripts (Vancouver: UBC press, 2013) at ch. 2; Elizabeth Sheehy, “Causation, Common Sense and the Common Law: Replacing Unexamined Assumptions with What We Know About Male Violence Against Women or From Jane Doe to Bonnie Mooney” (2005) 17 CJWL 97; Erika Chamberlain, “Tort Claims for Failure to Protect: Reasons for (Cautious) Optimism since Mooney ” (2012) 75 Sask L Rev 245 [Chamberlain, “Optimism”]; and Julia Tolmie, “Police Negligence in Domestic Violence Cases and the Canadian Case of Mooney: What Should Have Happened, and Could It Happen in New Zealand?” 2006 NZLR 243.
10 See 178–181 and 196–200 below.
11 See text accompanying note 3.
12 McBride, “Michael Comment”, supra note 4 at 6 discussing Anns v Merton London Borough Council, [1978] AC 728, [1977] UKHL 4 [Anns].
13 See 189–195, below, where this “Good Public Samaritan” approach is criticized.
172 UNBLJ RD UN-B [VOL/TOME 68 policy approach often has resulted in the same outcome – no unique public duty – as the Diceyan approach.14 The sympathetic facts in Michael are useful to illustrate the issues. Section 5 considers the possibility of unique public duties of care in narrowly defined specific circumstances.
Although Anns has been overruled in the UK,15 Canada continues to follow robustly the Anns policy approach to duty of care.16 Not surprisingly, therefore, the Supreme Court of Canada (‘the Supreme Court’) has recognized at least 5 unique public duties of care.17 The Supreme Court also purports to follow the same rule as the UKSC, that a common law duty of care cannot be imposed on public authorities based on the words of the enabling statutes alone.18 Yet it is difficult to explain the recognized unique public duties otherwise. Rarely has the court acknowledged that is creating a unique public duty of care and never has it discussed explicitly and fully whether it is appropriate to do so, as did the court in Michael. There may exist a principled justification for imposing unique public duties, but it has never been put forth as such. Instead, I will suggest that the unique duties that Canada does recognize have emerged on an ad hoc basis, in the process damaging the critical structure of common law adjudication. 2017 CanLIIDocs 175
Issues surrounding unique public duties are sometimes confused with issues surrounding government immunity for high level policy decisions.19 This is probably because both are concerned with respecting the separation of powers between the legislative bodies and the courts. However, there is a fundamental difference. A case for immunity only arises when the public defendant would otherwise be liable for breaching a recognized duty of care.20 Immunity is a concept employed to reduce government responsibility for otherwise negligent conduct below the level of
14 McBride, “Michael Comment”, supra note 4 at 6.
15 The two-step duty framework was rejected in Caparo Industries plc v Dickman, [1990] 2 AC 605, [1990] 1 All ER 568. Liability for defective structures was rejected in Murphy v Brentwood District Council, [1991] 1 AC 398, [1991] UKHL 2.
16 This is true especially after the decision in Cooper v Hobart, 2001 SCC 79, [2001] 3 SCR 537 [Cooper].
17 The following unique duties have been identified by Bruce Feldthusen, “Unique Public Duties of Care: Judicial Activism in the Supreme Court of Canada” (2016) 53 Alberta L Rev 955: Schacht v O’Rourke, [1976] 1 SCR 53, 55 DLR (3d) 96 [Schacht]; Kamloops (City) v Nielsen, [1984] 2 SCR 2, 10 DLR (4th) 641 [Kamloops]; Just, supra note 6; Odhavji Estate v Woodhouse, 2003 SCC 69, [2003] 3 SCR 263 (the negligence holding against the Chief of Police) [Odhavji]; and Fullowka v Pinkerton’s of Canada Ltd, 2010 SCC 5, [2010] 1 SCR 132 [Fullowka].
18 R v Saskatchewan Wheat Pool, [1983] 1 SCR 205, 143 DLR (3d) 9 [Saskatchewan Wheat Pool]; Odhavji Estate, ibid. Admittedly, these decisions are difficult to reconcile with others. See e.g. Cooper, supra note 16 at para 43.
19 See also McBride, “Professional Negligence”, supra note 4 at 17 and 20.
20 The decision in Imperial Tobacco, supra note 6 provides an excellent example. The defendant Canada was held to owe a recognized prima facie duty regarding misrepresentations that caused detrimental reliance loss at Step 1 of the Anns framework. However, at Step 2 Canada was granted immunity for its alleged breach of duty because the court held that the representation was made as an exercise of high level policy.
2017] BUNGLED POLICE EMERGENCY CALLS 173 responsibility owed to others by private citizens. In contrast, unique public duties arise by definition only when the conduct at issue is not governed by ordinary private party negligence law. Whereas a claim of immunity seeks special exculpatory treatment, a unique public duty is an additional duty owed only by the public defendant. The objection to a unique public duty is that it violates Dicey’s equality principle.21 There is no other duty from which immunity could be sought.22
Although “duty of care” is a classic common law negligence question, unique public duties of care raise important questions about the separation of powers in constitutional law. When courts create unique public duties of care, I will argue that they appropriate unilaterally powers that previously and properly belonged to the legislative branch.23 I will argue that the law of negligence ought not to recognize unique public duties of care unless a principled justification that does not prove to be over-broad can be identified. No doubt there are compelling counter-arguments.24 Canadian law would benefit if these came forth explicitly.
2. The Duty to Confer Benefits in Private Party Negligence Law 2017 CanLIIDocs 175
To evaluate the case for unique public duties of care it is necessary to identify the principles that govern duties to provide benefits between private parties.25 The general rule in negligence is that one private party does not owe an affirmative duty to confer benefits upon another. I will refer to this as the “no duty” rule. There are
21 Logically, the objection to unique public duties should be extended to policy immunity. They both entail treating governments differently from private parties.
22 Prior to Imperial Tobacco, supra note 6, Just was the leading authority on policy immunity. Just created a unique public duty of care. Cory J was possibly careless in using immunity language to refer to the process of creating a unique duty. I think not. Several passages in Just suggest that Cory J did not support unique public duties and did not realize that he was creating one. See Just, supra note 6 at 1239 and 1244. The finding of proximity in Just is out of line with basic private party negligence law and with proximity decisions in most other public authority cases. See Taylor v Canada (AG), 2012 ONCA 479 at para 80, 111 OR (3d) 161. The erroneous assumption that he was dealing with a pre-existing common law duty principle is what best explains Cory J’s discussion about immunity.
23 This is a different and narrower argument than a rights-based argument that would preclude any and all judicial policy making. See 192, below.
24 I would find it more difficult to adhere to this position if I were not confident that basic Canadian negligence law would support a duty of care on the part of the police in a case such as Michael.
25 I rely on the excellent article by Peter Benson to explain the significance of this distinction in private law. At the risk of over-simplification, he says “. . . misfeasance restricts the fundamental imperative in private law to a prohibition against conduct, whether act or omission, that injures or interferes with a definite but limited kind of protected interest; namely, another’s ownership right” (person or property). Nonfeasance does not interfere with the plaintiff’s right to exclude others from her personal or proprietary interests; it fails to benefit her. See Peter Benson, “Misfeasance as an Organizing Normative Idea in Private Law” (2010) 60 UTLJ 731 at 733 and generally at 731–737. See also Childs, supra note 7 at paras 31–32; and Donal Nolan, “The Liability of Public Authorities for Failing to Confer Benefits (2011) 127 LQR 260.
174 UNBLJ RD UN-B [VOL/TOME 68 numerous exceptions. Some are unclear or contentious. Many overlap. Authors and courts classify the exceptions differently.
A. Defendant by his Fault Creates a Situation of Peril
When a defendant by his fault creates a situation of peril, the defendant comes under a duty to protect the person so-imperilled.26 Strictly speaking, this is not an exception to the “no duty” rule. The “no duty” rule does not apply to misfeasance that causes physical harm. However, the creation of the new peril frequently occurs in the course of providing a benefit to another. In Hampshire, for example, a fire department was held liable based on its decision during its intervention to turn off the sprinkler system. Turning off the sprinklers made the fire damage more extensive than it would have been had they done nothing. The department was held liable for the additional damage. 27 In Zelenko v Gimbel Brothers the defendant removed the ill plaintiff to a place where no one else could help him. The fresh harm was the defendant’s denying the plaintiff other aid.28 There is no reason to distinguish making someone worse off by denying other aid from any other manner of inflicting 2017 CanLIIDocs 175 harm.
B. Duty to Warn
There is no general duty to warn another about dangers of which one is aware. 29 However, product manufacturers and distributors owe a duty to warn of inherently dangerous products or dangerous uses of safe products. The duty arises when the defendant acquires actual knowledge of the danger, including those it discovers after sale.30 Significantly, it arises even when the defect was not caused by any fault on the part of the defendant. This exception only applies to a defendant who has created the peril. It is also relevant that the product manufacturer exception is limited to
26 Videan v British Transport Commission, [1963] 2 QB 650 at 699, [1963] 2 All ER 860, quoted with approval in Horsley v McLaren, [1972] SCR 441 at 444, 22 DLR (3d) 545 [Horsley]. Dorset Yacht Co Ltd v Home Office, [1970] AC 1004, [1972] All ER 294 [Dorset Yacht] discussed below, may also be explained this way. Horsley also establishes that if the defendant’s breach of the original duty foreseeably induces a new rescue attempt, a further duty may be owed to protect the new rescuer. The defendant’s breach of the original duty constitutes a fresh peril to the foreseeable rescuer.
27 Capital & Counties plc v Hampshire County Council, [1997] QB 1004, [1997] 2 All ER 865 (CA), discussed in Michael, supra note 1 at para 71.
28 Zelenko v Gimbel Brothers, 287 NYS 134 at 135, aff’d 287 NYS 136 (1936).
29 See Margaret I Hall, “Duty to Protect, Duty to Control and the Duty to Warn” (2003) 82 Can Bar Rev 645 at 673–79.
30 Rivtow Marine v Washington Iron Works, [1974] SCR 1189, 40 DLR (3d) 530; Lambert v Lastoplex Chemicals, [1972] SCR 569, 25 DLR (3d) 121; Hollis v Dow Corning, [1995] 4 SCR 634, 266 DLR (4th) 257.
2017] BUNGLED POLICE EMERGENCY CALLS 175 commercial defendants who create the risk.31 It is doubtful, but possible, that a duty to warn might be extended to non-commercial defendants such as motorists who create a peril by being involved in an automobile accident without any fault on their part.32
C. Special Relationships of Control:33
There exists a number of status-based “special relationships” where the more powerful party owes affirmative duties to the more vulnerable party. The term “special relationship” is not a technical term so much as a convenient label for an open-ended list of such relationships. Some are formal, ongoing status relationships like “parent-child” or “doctor-patient.” Others, like “commercial alcohol provider- customer,” are situational.34 The underlying principle is that a defendant in a position of control over a vulnerable plaintiff owes certain affirmative obligations to the plaintiff. These are true exceptions to the “no duty” rule.
Vulnerability alone cannot justify an exception to the “no duty” rule. If it 2017 CanLIIDocs 175 did there would nothing left of the rule.35 Having control simply makes an intervention easier or more likely to be effective. This is irrelevant to the “no duty” rule. Control and vulnerability must work together. Perhaps the answer lies in the plaintiff’s exclusive right to control over her own body or property. When a defendant assumes or obtains, and retains, some of what was originally the plaintiff’s exclusive right of control, the defendant is no longer a mere bystander. The relationship has become “special” because of the transfer of control. At that point, the defendant has been entrusted with some of the core rights of the plaintiff. It has also been argued that a government police force effectively monopolizes and controls a citizen’s right to protect herself from crime and that this entails special affirmative obligations.36
31 See Childs, supra note 7 at para 35.
32 See Ziemer v Wheeler, 2014 BCSC 2049, [2015] BCWLD 232. See also Oke v Weide Transport (1963), 41 DLR (2d) 53, 43 WWR 2 (Man CA) per Freedman JA dissenting.
33 See Hall, supra note 29; Nicholas McBride & Roderick Bagshaw, Tort Law, 5th ed (Harlow England: Pearson Education, 2015) at 245–253.
34 See Hall, supra note 29 at 653.
35 See Allan Beever, “The Basis of the Hedley Byrne Action” in Kitt Barker, Ross Grantham & Warren Swain, The Law of Misstatements (Oxford and Portland Oregon: Hart Publishing, 2015) 83 at 97. Beever suggests that had the passenger who fell overboard in Horsley swum nearby to another boat the vulnerability and dependency would be the same as it was when he stayed near his own boat. But those on the other boat would not owe any special duty to rescue him. See contra Andrew Robertson & Julia Wang, “The Assumption of Responsibility” in Barker, Grantham & Swain 49 at 70 who say dependency is the key.
36 This is an important idea offered in support of a unique public duty of care by Stelios Tofaris & Sandy Steel, Police Liability in Negligence for Failure to Prevent Crime: Time to Rethink (University of
176 UNBLJ RD UN-B [VOL/TOME 68
There are numerous examples of recognized special relationships of control: “employer-employee”37; “pleasure boat captain – passenger”38; “occupier of land- entrants”39; and “landlord-tenant.”40 Some of these special relationships involve public defendants including “police officer-suspect”41; “police officer-informant”42; “police force-potential victims of crime”43 and “warden-prisoner.”44
Doctors and other health care providers owe a unique obligation to perform the professional service with the reasonable care expected of similarly situated professionals.45 A doctor who voluntarily intervenes to treat an accident victim must exercise reasonable care to improve the patient’s condition, not merely to avoid making it worse.46 There are several possible explanations for this rule, none universally accepted as dominant. It is certainly relevant that medical intervention puts the patient’s health, possibly life, at risk.47 This special obligation may also be explained as a requirement derived from the professional status. This may be part of our understanding of what it means to render “professional” services. Even by contract, medical doctors are not permitted to provide bargain-basement, lower skill professional services. They must meet the standard of the profession. Or, this 2017 CanLIIDocs 175 affirmative obligation may derive from one of two possible meanings of “assumption of responsibility” in basic private party negligence. The first refers to a duty not to induce detrimental reliance loss. However, the second, an “equivalent to contract” approach to assumption of responsibility, requires the defendant to provide a positive
Cambridge Legal Studies Research Paper Series, Paper No 39/2014) at 5 [Tofaris & Steel, “Police Liability”], online:
37 Hunt v Sutton Group Incentive Reality, 60 OR (3d) 665, 215 DLR (4th) 193; Jordan House v Menow, [1974] SCR 239, 38 DLR (3d) 105 per Laskin J [Jordan House].
38 Horsley, supra note 26.
39 Depue v Flateau, 111 NW 1 (Minn SC 1907) discussed in Beever, supra note 35 at 96.
40 Q v Minto Management (1985), 49 OR (2d) 531, 15 DLR 4th 581 (SC).
41 Hill, supra note 3.
42 Robertson & Wang, supra note 35 at 78–9.
43 Jane Doe v Toronto (Metropolitan) Commissioners of Police, 39 OR (3d) 487, 72 DLR (4th) 580 (Ct J (Gen Div)).
44 Dorset Yacht, supra note 26.
45 The same may be true of other professionals such as lawyers who voluntarily provide professional services. Michael, supra note 1 at para 178 quoting Lanphier v Phipos (1838), 8 C & P 475, 479; Robert Stevens, Torts and Rights (Oxford: Oxford University Press, 2007) at 12.
46 Nolan, supra note 25 at 282.
47See Criminal Code of Canada, RSC 1985, c C-46 at s 217.
2017] BUNGLED POLICE EMERGENCY CALLS 177 benefit if that was the responsibility assumed. These two possibilities are discussed below.48
There are also numerous examples of “special relationships of control” that entail the dual duties to control the vulnerable party and to protect third parties from being injured by the vulnerable party.49 Liability to the third party will depend on the nature of both relationships, defendant-perpetrator and defendant-third party plaintiff.50 Doctors may be required to protect others from their patients.51 Parents are required to protect their child and also required to control the child so that the child does not harm others. Commercial alcohol providers owe a duty to protect their patrons,52 and also a duty to control their patrons so they do not injure others.53 Jailers owe a duty to protect their prisoners, and also a duty to control them so they do not escape and injure others.54 These duties to the third party are not pure exceptions to the no duty rule. They are all but specific examples of the broader principle discussed above that a defendant who creates a situation of peril by his fault owes a duty to protect persons so imperilled.
2017 CanLIIDocs 175 D. Assumption of Responsibility
Recall, the general rule is that one party is under no duty to confer a benefit on another. In Michael, the majority considered two recognized exceptions to the “no duty” rule, eventually holding that neither applied to the facts. The first was when one party owed a duty to control another, and thereby came under a duty to protect a third party.55 The second was when the defendant assumed responsibility to benefit the plaintiff under the Hedley Byrne principle.56
48 See 178, below.
49 See generally Hall, supra note 29. This is one of two exceptions to the “no duty” rule identified in Michael, supra note 1 at para 99.
50 Hall, ibid at 646; Michael, supra note 1 at para 99.
51 See Wenden v Trikha, 1 Alta LR (3d) 283, 124 AR 1 aff’d 135 AR 382, 14 CCLT (2d) 225 (CA), leave to appeal to the SCC refused 149 AR 160, 17 CCLT (2d) 285. See also Douglas Smith, “Wenden v Trikha and Third Party Liability of Doctors and Hospitals: What's Been Happening to Tarasoff” (1995-96), 4 Health L Rev 12.
52 Jordan House, supra note 37 at 248.
53 Stewart v Pettie, [1995] 1 SCR 131 at 143, 25 Alta LR (3d) 297.
54 Dorest Yacht, supra note 26 was so explained in Michael, supra note 1 at paras 58, 89, 142.
55 Michael, supra note 1 at para 135.
56 Ibid at para 136.
178 UNBLJ RD UN-B [VOL/TOME 68
The court in Michael stated the Hedley Byrne principle in two significantly different ways. First, commenting on what (little) Lord Goff had said about it in Spring, the court said:57
The underlying principle rested on an assumption of responsibility by the defendant towards the plaintiff, coupled with reliance by the plaintiff on the exercise by the defendant of due skill and care. The principle that a duty of care could arise in that way was not limited to a case concerned with the giving of information and advice (Hedley Byrne) but could include the performance of other services.
Under this view, the defendant assumes responsibility to exercise due skill and care, the familiar standard of care in negligence law, but not unknown in contract.58 What precisely “assumes responsibility” means is contentious in UK law.59 Michael seems to adopt the view that responsibility is actively assumed by the defendant rather than imposed by law.60 Detrimental reliance is presumably required.61 This explanation of Hedley Byrne is a variant, albeit a significant variation,62 of the “create the peril” situation.63 This is also a principle that antedates Hedley Byrne, applies to acts as well as to statements, and applies to physical harm as well as to economic loss.64 2017 CanLIIDocs 175
Later, Lord Toulson explained the Hedley Byrne principle differently:65
The principle established by Hedley Byrne is that a careless misrepresentation may give rise to a relationship akin to contract under which there is a positive duty to act. Lord Devlin spoke of "an assumption of responsibility in circumstances in which, but for the absence of consideration, there would be a contract" and he said that "wherever there is a relationship equivalent to contract, there is a duty of care" (pp 529- 530).
57 Ibid at para 67.
58 See e.g. Esso Petroleum v Mardon, [1976] QB 801, [1976] 2 All ER 5 (CA) per L Denning.
59 See generally Beever, supra note 35; Robertson & Wang, supra note 35; and Christian Witting, “What Are We Doing Here? The Relationship Between Negligence in General and Misstatements in English Law” in Barker, Grantham, & Swain, supra note 35.
60 Michael, supra note 1 at para 67
61 Ibid at para 138.
62 The plaintiff injures herself by relying on the defendant. There must be an adequate explanation of why she should be able to hold the defendant responsible for this. See Stephen R Perry, “Protected Interests and Undertakings in the Law of Negligence” (1992) 42 UTLJ 247 at 285.
63 See Nolan, supra note 25 at 278.
64 Mercer v SE&C Ry, [1922] 2 KB 549.
65 Michael, supra note 1 at para 135.
2017] BUNGLED POLICE EMERGENCY CALLS 179
The “equivalent to contract” approach has several possible implications. Presumably it would require privity, thereby limiting the ambit of responsibility. The basis of liability would not be the failure to confer the benefit, but the breach of the undertaking.66 The defendant could assume responsibility to provide the plaintiff with a positive benefit, not merely for taking due care to prevent a detrimental reliance loss.67 The court would be looking for an intention to be bound to a promise confer a benefit, not merely an intention that the plaintiff rely. The issue would go beyond whether responsibility was assumed to what responsibility was assumed.
The equivalent to contract approach enjoys powerful academic support.68 The cases said to support it are not definitive.69 Later, Lord Toulson seems to reject the broader approach without discussing it, and to require detrimental reliance. Despite the absence of clear authority, the case for a duty to provide positive benefits independent of detrimental reliance is a compelling one that is likely to be pursued in future. The court in Michael missed an excellent opportunity to consider it.
3. Michael falls within the Assumption of Responsibility Exception to the “No 2017 CanLIIDocs 175 Duty” Rule
Although there was great vulnerability to the police on the part of Ms. Michael, the police did not exercise any control over her such as would have entailed a duty to protect her. They did not create the original danger that Williams posed to Michael. Nor were they in a traditional special relationship of control with the murderer Williams based on custody. Later, I will suggest that an expanded duty to control might apply had the police been aware of the risk to Michael, but she had been unaware of it. In the actual Michael situation, the most promising avenue for establishing a duty of care under basic negligence law was to establish an assumption of responsibility on the part of one or both of the police emergency operators. The claimants must have been devastated to discover that the majority only found it necessary to devote a single paragraph to this crucial line of argument, and to dismiss it summarily. The key paragraph reads as follows:70
Mr Bowen submitted that what was said by the Gwent call handler who received Ms. Michael's 999 call was arguably sufficient to give rise to an assumption of responsibility on the Hedley Byrne principle as amplified in
66 See Allan Beever, Rediscovering the Law of Negligence (Oxford: Hart Publishing, 2007) at 222.
67 See Beever, supra note 35 at 98–99.
68 See ibid at 83, 104–105; Stevens, supra note 45 at 14; Nolan, supra note 25 at 282–83.
69 These are primarily the health care professional cases some of which may be explained otherwise. See 176–177, above. But see especially Barrett v Ministry of Defence, [1995] 3 All ER 87, [1995] 1 WLR 1217. See also McBride & Bagshaw, supra note 33 at 230.
70 Michael, supra note 1 at para 138.
180 UNBLJ RD UN-B [VOL/TOME 68
Spring v Guardian Assurance Plc. I agree with the Court of Appeal that the argument is not tenable. The only assurance which the call handler gave to Ms. Michael was that she would pass on the call to the South Wales Police. She gave no promise how quickly they would respond. She told Ms. Michael that they would want to call her back and asked her to keep her phone free, but this did not amount to advising or instructing her to remain in her house, as was suggested. Ms. Michael's call was made on her mobile phone. Nor did the call handler's inquiry whether Ms. Michael could lock the house amount to advising or instructing her to remain there. The case is very different from Kent v Griffiths where the call handler gave misleading assurances that an ambulance would be arriving shortly.
This is virtually all the court had to say about the application of private party negligence law to what actually happened to Ms. Michael. The court’s lengthy and detailed discussion of the (arguably) grander principles of unique public duties, and the rights versus policy debate, overwhelmed discussion of the outrageous circumstances of the murder.71 But there is nothing inconsistent with supporting the Diceyan approach or the rights-based approach on the one hand, and giving proper
consideration to the question of what responsibilities in ordinary private party 2017 CanLIIDocs 175 negligence, if any, a police emergency operator assumes to a citizen who calls in distress on the other. The conclusion may well be “none.” However, this paragraph offers scant and superficial justification for that conclusion.
The Supreme Court of Canada no longer follows Hedley Byrne in negligent misrepresentation cases. Today, a plaintiff need only establish that the defendant ought to have reasonably foreseen that the plaintiff would rely on the information or advice to their detriment, and that reliance in the particular case was reasonable. The Supreme Court restricted the scope of the duty by adopting a transaction-specific “end and aim” test to control indeterminate liability.72 Possibly, the reliance approach in Hercules has replaced the assumption of responsibility approach across the board.73 Either way, the claimants in Michael have a strong case. It is obvious
71 Lord Toulson’s reasons constitute a dream come true for supporters of the Diceyan approach, and supporters of the rights-based approach to negligence law. One hundred and thirty-nine paragraphs of obiter dicta would have been less compelling. Even McBride’s usually sharp style of criticism appears somewhat muted by his undisguised joy at the triumph of the Diceyan and rights-based approaches. In “Michael Comment”, supra note 4 at 9–10 he says with uncharacteristic understatement: But the UKSC was taking a bit of a chance by making these findings without the benefit of a full hearing. Perhaps the need to lay down a strong line in Michael on the future of liabilities of public bodies in omissions cases justified doing this, but I do feel some unease at this aspect of the decision, particularly as there was evidence that Joanna’s neighbours could hear what was going on, and were concerned enough that they called the police themselves (though their calls were mis-routed to the Gwent police as well).
72 Hercules Managements v Ernst & Young, [1997] 2 SCR 165 at 200, 146 DLR (4th) 577. This is developed by Bruce Feldthusen, “Hedley Byrne: Misused, then Exiled by the Supreme Court of Canada” in Barker, Grantham & Swain, supra note 35, ch 11.
73 That said, there is ample support for the assumption of responsibility principle that has never been overruled or even criticized. See e.g. Welbridge Holdings Ltd v Greater Winnipeg, [1971] SCR 957, 22 DLR (3d) 470; J Nunes Diamonds Ltd v Dominion Electric Protection Co, [1972] SCR 769, 26 DLR (3d)
2017] BUNGLED POLICE EMERGENCY CALLS 181 that someone in the position of Ms. Mason, the first call operator, knows that emergency callers are relying on her. The claim is transaction-specific. The very nature of the service is to intend, induce and invite specific reliance. Citizens call the emergency line expecting a proper response to the emergency. Relying on the police to provide one is entirely reasonable.
The claim in Michael was struck out on a different basis. The majority held that Ms. Mason made no relevant statements or promises whatsoever. The majority conceded that if Mason told Michael that help would arrive shortly, there might have been a duty, and liability if reliance on that statement had caused additional harm.74 True, Mason only gave assurance that she would pass on the call to the proper police department. But it is doubtful the claim would have been dismissed had Mason simply called Gould, told him there was a call for him and passed on Michael’s number. Surely Ms. Michael was entitled to believe at a minimum that Mason undertook to convey the full and relevant details of her call according to established police procedure. This Mason failed to do.
When Mason asked Michael whether she was able to lock the house, it 2017 CanLIIDocs 175 would have been reasonable for Michael, beaten, terrified, alone, and as vulnerable to Mason as she can possibly be, to assume that she was being advised to stay in the house and lock up. The problem is a line of UK authority that requires in police cases that only an unambiguous explicit promise can constitute an assumption of responsibility.75 Lord Kerr effectively criticized this rule in dissent.76 Employing it in Michael is the ultimate irony. The entire majority judgment is devoted to championing the Diceyan approach. Yet, when it came to the actual claim the majority retained a rule that protects the police from the ordinary rules of negligence with a unique and strictly limited public duty.
Many, if not most, of the significant facts referred to immediately above are uniquely associated with the police response to an emergency call. However, the duty that I argue should result is not a unique public duty. The principle that liability will be imposed based on an assumption of responsibility is a principle of basic negligence law. The very nature of the public emergency service is to intend, induce and invite specific reliance, the touchstones of an assumption of responsibility. In this way, the common law may take into account factual matrixes that are unique to
699 per Spence J (dissenting), citing John Fleming, The Law of Torts, 4th ed (Sydney: Law Book Co, 1971) at 564; Hodgins v Nepean (Township) Hydro-Electric Commission, [1976] 2 SCR 501, 60 DLR (3d) 1; Carman Construction Ltd v Canadian Pacific Railway Co, [1982] 1 SCR 958, 136 DLR (3d) 193; Central Trust Co v Rafuse, [1986] 2 SCR 147, 31 DLR (4th) 481.
74 Michael, supra note 1 at para 138; this was the case in Kent v Griffiths (2000), [2001] QB 36, [2000] 2 All ER 474 (CA).
75 See Michael, supra note 1 at para 164. This rule is consistent with the “contract without consideration” approach to Hedley Byrne discussed immediately below.
76 See supra note 1, especially at paras 165–68.
182 UNBLJ RD UN-B [VOL/TOME 68 government activity and address the understandable view that there are situations where we might expect more from the government than from a private citizen. This is how the argument above is constructed.77 The important thing is that the principle of law is the same. In contrast, to impose liability on the police in the absence of an assumption of responsibility would require a unique principle of public liability.
The next question is whether the claimants are required to prove that the failure to communicate the emergency properly, or the advice to stay in the house, or any other assumptions of responsibility that might have been revealed at trial, put Ms. Michael in a worse position than she would have been in had she not called the police. There is strong academic support for a “contract without consideration” approach to assumption of responsibility. Under that approach, it is the breach of an undertaking, not the infliction of detrimental reliance loss, which constitutes the actionable wrong with assumptions of responsibility.78 A party may be held liable for failing to deliver a benefit based on an undertaking to do exactly that. Arguably, the undertaking to forward the full and relevant details of Michael’s original call to the proper police station was an undertaking to provide a benefit. The police promised to help her and their failure to do so allowed Williams to murder her. This is an 2017 CanLIIDocs 175 argument that ought to have been considered by the court, and the claimants ought to have been allowed to develop an evidentiary foundation for it at trial.
Recall also that a doctor’s duty to use reasonable care to improve the patient’s situation may be based on the risk to health and life with medical interventions.79 If so, an analogy might be drawn to the case of emergency responders who begin a professional interaction with a person who is in the midst of a life-threatening emergency. This would be consistent with the third requirement for an exception to the “no duty” rule favoured by Tofaris and Steel, “A's status creates an obligation to protect B from that danger.”80
Finally, assuming the court had considered and rejected these two arguments and insisted that detrimental reliance loss was an essential element of the claim, it is arguable that there was such detrimental reliance. There were concerned neighbours in the near vicinity to whom Michael might have turned had she not reasonably understood that she had been advised to stay in the house.81 The claimants should have been allowed to develop this argument at trial.
77 The same is true of the decision in Hill, supra note 3 at para 27.
78 Detrimental reliance probably must be foreseeable for the duty to arise, but that is different than requiring actual detrimental reliance as a condition of duty. See Nolan, supra note 25 at 285–86.
79 Supra note 45.
80 See Tofaris & Steel, “Police Liability”, supra note 36 at 5; and Tofaris & Steel, “Omissions”, supra note 36 at 128. See also Michael, supra note 1 at para 197 per Lady Hale, and at paras 178–181 per Lord Kerr dissenting.
81 McBride, “Michael Comment”, supra note 4 at 9–10.
2017] BUNGLED POLICE EMERGENCY CALLS 183
The reasons for judgment in Michael constitute an impressive review of the authorities, an overwhelming endorsement of the Diceyan approach, and a rejection of the policy approach to establishing new duties of care. They also constitute an inexplicably dismissive application of the law to the facts. What is needed to resolve this claim is evidence produced by the parties, tested under oath and woven into submissions by trained advocates. The claim ought to have been allowed to proceed to trial. It is unlikely that a Canadian court would have stopped this claim on a preliminary motion,82 and surprising that the UKSC did so. There is good reason to be uneasy about the result.
4. The Difficulties with Unique Public Duties
A. Introduction
I assume that a Canadian court would have held for the claimants in Michael under ordinary negligence law. If not, the question would arise whether the court should 2017 CanLIIDocs 175 instead impose a unique public duty. This question may seem odd given how little attention the courts have paid to unique public duties in recent times. Subsection B below begins by pointing out that although unique public duties do exist in Canadian negligence law today, at one time Canadian courts refused to recognize them. Most legislative bodies and the courts defined the distinction between the political and the judicial realm according to the Diceyan principle that governments should be governed by the same rules of negligence law as private parties.
Subsection C considers the general objections to unique duties of care. It suggests that the distribution of government largess should be, subject to citizens’ rights-based claims, a purely political function. It takes no position on the broader question that arises in rights-based negligence discourse about whether courts should seek to effect distributive justice between citizens themselves. It identifies the “Good Public Samaritan” principle as the apparent justification for the unique public duties currently recognized in Canada. This principle states that once a public defendant decides to confer a benefit, it then comes under a duty of care to render the benefit with reasonable care. It appears that this principle, suspect in its own right, is applied on an ad hoc basis. Were it applied to all “like cases,” as it should be with a common law commitment to stare decisis, the shift in power from the legislatures to the courts, effected by the unconstrained courts themselves, would constitute a dramatic change to our constitutional democracy.
82 See Mooney, supra note 9. See also Sunny Dhillon, “Woman sues Surrey RCMP officer for failure to act on reported assault,” The Globe & Mail, (14 March 2016), online:
184 UNBLJ RD UN-B [VOL/TOME 68
Subsection D explores whether much more limited principles might be identified to allow for unique public duties in exceptional cases. I remain open to such a possibility, but conclude that attempts to justify principled exceptions that are not over-broad remain a work in progress.
B. Unique Public Duties in Canada: Then and Now
Historically, the rule in Canada was the same as the majority decision in Michael – the courts should not recognize unique public duties of care. Today there are at least five Supreme Court of Canada precedents for unique public duties,83 including two imposed on the police.84 This transition to unique public duties has never been openly justified as such by the Supreme Court, let alone considered in depth as it was in Michael. Canadian negligence law and Canadian constitutional law would benefit from fresh consideration of unique public duties of care.
The point is debatable, but I would suggest that most Canadians do not 2017 CanLIIDocs 175 conceive of their constitution as consisting of one set of rules for private citizens and another set for public officials. True, it is often appealing to prefer that a loss be shifted to a deeper pocket, especially when an avoidable loss is suffered by a vulnerable plaintiff. Michael is a sympathetic case. But the temptation to create unique public duties must be measured against the fact that a society that normalizes unique obligations will also normalize unique public immunities and privileges. A special UK rule limiting police liability was invoked to defeat Ms. Michael’s claim.85 I am even more confident that Canadians do not generally approve of unique excuses for government negligence.86 The question must be considered as a broader one than that of unique liabilities. We are really talking about conceptualizing government and private parties as separate and distinct in private law. This has not been our tradition. Our tradition has been that public actors should be “under the same
83 Supra note 17.
84 See Schacht and Odhavji, supra note 17.
85 Michael, supra note 1 at para 164, see also text accompanying note 75.
86 One example is core government policy immunity, supra note 6. Many public authorities are shielded by legislation requiring proof of gross negligence or bad faith. See e.g. Vancouver (City) v Ward, 2010 SCC 27, [2010] 2 SCR 28. See also Jacobs v Ottawa (Police Service), 2016 ONCA 345 at para 12, 400 DLR (4th) 148 where the court affirmed that a police disciplinary offence had to be proven on a special and higher standard of proof than the established civil standard of a balance of probabilities. Dissappointed plaintiff’s lawyer Lawrence Greenspon was quoted as saying “we’ve got one law for the police, and another law for everyone else” in “Supreme Court dismisses Ottawa police appeal to lower standard of proof for officers,” Ottawa Citizen (13 January 2017) online:
2017] BUNGLED POLICE EMERGENCY CALLS 185
(emphasis added) law that applies to private citizens.”87 This is a fundamental political principle.88
Crown Liability legislation by which the Crown surrendered its historical immunity from liability in tort evidences Canada’s historical commitment to the Diceyan principle. The federal Crown, and the Crown in 8 of the 9 common law provinces, consented only to being held vicariously liable for torts committed by their servants or agents.89 They did not accept to be held liable for “peculiarly governmental activity” where there exists “no clear private analogue.”90 British Columbia also consented to “direct” liability, which might be interpreted to include unique public duties.91 In Swinamer, the Supreme Court collapsed the distinction between direct and vicarious liability without considering explicitly a challenge to unique public duties.92 The dissenting judgment in Schacht v O’Rourke was the closest the Supreme Court has ever come to considering whether Crown Liability legislation precludes unique duties of care.93 The majority did not address the issue. Arguments based on the Crown liability statutes are seldom raised and seldom successful today.94 The legislatures have apparently acquiesced.95 It is highly 2017 CanLIIDocs 175
87 Hogg, Monahan and Wade, supra note 5 at 218–19.
88 This is a principle of formal equality. As such it is vulnerable to the criticism that employing it to defeat a claim such as Michael allows form to triumph over substance. In part, the answer is that a unique public duty is not necessary to allow Michael claimants to succeed. In part, the answer is that formal divisions of constitutional power are core principles of how we are governed. They are not “mere” formalities and toying with them is very likely to produce substantial social change. As such, the question of whether the courts ought to create unique public duties is deserving of transparent debate, something that has not happened in the Canadian courts.
89 Crown Liability and Proceedings Act, SC 1990, c 8, s 20; amended SC 2001, c 4, s 3(b)(1); Proceedings Against the Crown Act, RSA 2000, c P-25, s 5(1)(a); Proceedings Against the Crown Act, CCSM, c P140, s 4(1)(a) (excluding liability for economic loss); Proceedings Against the Crown Act, RSNL 1990, c P-26; Proceedings Against the Crown Act, RSNS 1989, c 360, s 5(1)(a); Crown Proceedings Act, RSPEI 1988, c C-32, s 4(1)(a); Proceedings Against the Crown Act, RSS 1978, c P-27, s 5(1)(a); and Proceedings Against the Crown Act, RSO 1990, c P27, s 5(1)(a). Canada and all the common law provinces except BC also have a provision substantially identical to s 10 of the federal Act which reads as follows: 10. No proceedings lie against the Crown . . . in respect of any act or omission of a servant of the Crown unless the act or omission would, apart from the provisions of this Act, have given rise to a cause of action for liability against that servant or the servant’s personal representative or succession. This historical immunity was constitutional, based on separation of powers, not on tort doctrine.
90 Hogg, Monohan & Wade, supra note 5 at 261 use these terms and support this proposition. See also Norman Siebrasse, “Liability of Public Authorities and Duties of Affirmative Action” (2007) 57 UNBLJ 84; and Lewis Klar, “Tort Liability of the Crown: Back to Canada: Saskatchewan Wheat Pool” (2006- 2007) 32 Advocates Q 293 at 294.
91 Crown Proceeding Act, RSBC 1996, c 89, s 2(c).
92 Swinamer v Nova Scotia (AG), [1994] 1 SCR 445 at 462–3, 129 NSR (2d) 321.
93 Schacht, supra note 17.
186 UNBLJ RD UN-B [VOL/TOME 68 doubtful an argument that unique public duties are precluded by the Crown Liability legislation would succeed today. My point is simply that the idea of limiting government liability to the same rules that govern private parties is not foreign to Canadian constitutional law.
Moreover, until at least 1989, the Supreme Court also acknowledged a common law rule precluding recognition of unique public duties in municipal government cases where the Crown Liability statutes do not apply. The famous decision Welbridge Holdings v Winnipeg applied the rule to deny liability.96 Kamloops v Nielsen recognized the same rule but did not apply it.97 Justice Cory also appeared to favour the Diceyan approach in Just.98 This common law prohibition against unique public duties has never been explicitly overruled, or even criticized. It simply disappeared. It too can no longer be considered a viable legal argument. The most one can say is that concerns about unique public duties are neither novel in Canadian law, nor radical. At one time this was how courts themselves defined the division of power between courts and legislative bodies.
Today there are numerous Supreme Court precedents for the recognition of 2017 CanLIIDocs 175 unique public duties. This in itself legitimizes an attempt to establish a unique police duty to govern the Michael situation. However, none of these precedents are directly on point. A quick summary below of the unique police duty cases suggests that any unique public duty created to govern the Michael situation would be a “novel” duty requiring a full Anns/Cooper analysis.99
The first unique public duty of care in Canada was established in a police case, Schacht v O’Rourke.100 A police officer attended at a traffic accident and then
94 No one pays any attention to the distinction today. See Hogg, Monahan & Wade supra note 5 at 182–3; Williams v Canada (AG), 76 OR (3d) 763, 257 DLR (4th) 704, varied on other grounds; and Davidson v Canada, 2015 ONSC 8008, 262 ACWS (3d) 648.
95 It is difficult to feel any sympathy for governments who could have, and presumably still could, put an end to unique public duties by legislation. This article is not premised on the need to protect the government. It is premised on the need to distinguish properly governance from common law adjudication in the interests of all citizens. A proper distinction benefits both the legislatures and the courts.
96 Welbridge Holdings v Winnipeg, supra note 73. Welbridge has never been overruled and may be undergoing a modern revival. See, e.g. 118143 Ontario Inc v Mississauga (City), 2016 ONCA 620, 405 DLR (4th) 338.
97 Justice Wilson expressed concern about creating unique public duties pertaining to pure economic loss in Kamloops, supra note 17 at 27. She did not raise that concern on the basic by-law enforcement issue. One possible explanation may be that liability in Kamloops was premised on improper government conduct, not simple negligence. See infra note 131.
98 Just, supra note 6.
99 Anns, supra note 12. In fact, the SCC regards every case involving a public defendant as a “novel” case. See Jost Blom, “Do We Really Need the Anns Test for Duty of Care in Negligence?” (2016) 53 Alta L R 895 at 905.
100 Schacht, supra note 17.
2017] BUNGLED POLICE EMERGENCY CALLS 187 left the scene without notifying the proper authorities about the remaining dangers at the scene. The court recognized a new common law duty based on a statutory duty to maintain a “traffic patrol.” The majority held that this included a duty to notify possible road users of any foreseeable dangers arising from the original accident. The decision in Schacht was clearly statute-specific and thus not a direct precedent for the Michael scenario.101 Moreover, Schacht might not be decided the same way today. The Supreme Court has subsequently held that it is impermissible to create a new common law duty of care based on the words of the statute alone, as it did in Schacht.102
In the most recent unique police duty case, Ohavji, a deceased criminal suspect’s family was allowed to proceed with an action for psychiatric damage against the Chief of Police. The action was based on the Chief’s failure to compel his officers to assist with an SIU investigation into the suspect’s death at the hands of the police.103 The basis of decision on the negligence point was unclear, highly unusual, and of no direct relevance to a case like Michael.
The recognition of a novel duty of care owed by investigating police 2017 CanLIIDocs 175 officers to criminal suspects recognized in Hill v Hamilton is not a unique public duty. That duty is entirely consistent with the principle in private party negligence law that imposes affirmative obligations in special relationships of control.104
Jane Doe v Toronto105 is a lower court decision. It is otherwise the most useful precedent for a unique police duty in a Michael situation, and as I will argue later, more broadly.106 The action was brought by a woman who had been raped by an intruder whom the police suspected to be a serial rapist with an established modis operandi based on entry via lower floor climbable apartment balconies within a small geographical area. The court held that the police owed a unique public duty to a limited class of potential rape victims to investigate crime with due care, and specifically owed a duty to warn the plaintiffs so they could protect themselves. There was an additional finding of liability based on a breach of the plaintiff’s Charter right to gender equality resulting in an identical award of damages. Adherence to rape myths as well as sexist stereotypical reasoning about
101 The majority suggested that there was common law authority for a unique public duty in such a case. Ibid at 86. The cases cited show only that a police officer has special public duties, but are not authorities for unique public tort duties.
102 See, e.g. Saskatchewan Wheat Pool, supra note 18.
103 Odhavji, supra note 17 at para 4. This decision is better known for its elaboration on the intentional tort of misfeasance in public office. No criticism is intended to the holding on that point, which is not relevant to the unique negligence duty.
104 See 175–177, above.
105 Jane Doe, supra note 43.
106 See infra note 167.
188 UNBLJ RD UN-B [VOL/TOME 68 rape, about women, and about women who are raped motivated and informed the failure to warn. Women were treated differently because some members of the force adhered to sexist notions that, if warned, women would panic and scare off the attacker. The negligence analysis was also infused with concern about the discriminatory treatment. It is unclear whether the negligence claim would have been decided the same way absent the gender discrimination.107
Police failures to respond effectively to domestic violence complaints have often been studied through the lens of systemic gender discrimination.108 If supporting evidence of unlawful discrimination is available, it would certainly help advance the effort to establish a unique police duty in a case like Michael. Moreover, the Charter breach aspect would remove any constitutional objection to the recognition of the unique duty. Courts have every right to address Charter breaches. The only remaining question would be whether a negligence duty is necessary, or whether the Charter remedy is sufficient.
C. The General Challenges with Unique Duties of Care 2017 CanLIIDocs 175
The basic question with unique public duties concerns the relationship between the government and its citizens. Relational matters are properly dealt with as questions of proximity. The distinction between inflicting harm and failing to prevent it is a fundamental distinction in our law.109 The proximity test is different for nonfeasance than for misfeasance. Proximity is a challenging concept to define succinctly. McLachlin CJC put it this way in Childs v Desormeaux:110
The law of negligence not only considers the plaintiff’s loss, but explains why it is just and fair to impose the cost of that loss on the particular defendant before the court. The proximity requirement captures this two- sided face of negligence.
As explained in the earlier review of private party negligence law, the general rule is that one party does not owe a duty to convey a benefit to another. This is so even when the parties are in as close a relationship to one another as they were in Michael. The plaintiff must establish a special relationship of control or an assumption of responsibility to succeed. The failure to have done so is what would drive the plaintiff to seek to establish a unique public duty. The argument for a unique public
107 But see Mooney, supra note 9. At trial a police officer was held to owe a duty of care to investigate and prevent crime, but the claim failed on causation on appeal with little discussion of the duty itself.
108 Mooney, supra note 9.
109 Supra note 25 and Childs, supra note 7.
110 Childs, supra note 7 at para 25. Childs involved a claim against a social host brought by a victim of an auto accident caused by an impaired guest. The claim was dismissed on proximity grounds, not step two Anns grounds. The proximity grounds were rooted in the distinction between causing harm and failing to provide a benefit. See also Cooper, supra note 16 at paras 34–35.
2017] BUNGLED POLICE EMERGENCY CALLS 189 duty must depend on arguments that distinguish public from private defendants. The plaintiff may not rely solely on statutory public duties to draw that distinction.111
The generally accepted rationales for the “no duty” rule are respect for individual autonomy,112 concerns about potentially indeterminate liability, and the difficulty of singling out one defendant among many who might have conferred the benefit.113 None of these rationales apply to public defendants who fail to protect citizens from physical harm. This clears the deck for an argument in favour of unique duties relating to public benefits. It does not, however, make that case.
A number of arguments for unique public duties of general application have been advanced.114 In my opinion, none are compelling in substance. Even if they were, imagine the outcome if these general principles were applied across the board to all public benefits, as they ought to be according to the principle of stare decisis. In such a scenario, it is difficult to identify what would remain of the distinction between a judicial function and a legislative one.
One argument in favour of unique public duties generally is that public 2017 CanLIIDocs 175 benefits are not purely gratuitous. Citizens pay tax to support public benefits. However, taxes are not paid in exchange for a specific public benefit, or for a particular quality of a specific public benefit, as would be the case, for example, in a contract. Paying tax does not convey a right to receive a particular benefit. While governments may not deserve the praise that might be accorded to a Good Private Samaritan, from the recipients’ point of view a government benefit is still a gratuitous benefit.
A similar argument based on “general reliance” is sometimes raised in support of unique public duties. General reliance usually means only that citizens expect public authorities to conduct their operations without negligence.115 Assuming
111 See e.g. Saskatchewan Wheat Pool, supra note 18.
112 Of the three, autonomy is paramount. In Michael, supra note 1 at para 177, Lord Kerr dissenting said “whereas it is arguable that a private individual's freedom has an intrinsic value in its contribution to an autonomous life, the value of the state's freedom is instrumental and lies in the contribution that it makes to the fulfilment of its proper functions.”
113 McBride, “Michael Comment”, supra note 4 at 9, summarizing Tofaris & Steel, “Police Liability”, supra note 36. See also Tofaris & Steel, “Omissions”, supra note 36 at 129–133; McBride & Bagshaw, supra note 33 at 216–219; and Siebrasse, supra note 90 at 87.
114 By general arguments I mean those that apply across the board to any situation to which the relevant principle applies. For example, the duty to avoid causing foreseeable physical damage to a foreseeable plaintiff applies generally to almost every potential defendant and plaintiff. It is not limited to product manufactures and consumers, the situation from which it was derived. Duties based on assumptions of responsibly apply generally to almost every potential defendant and plaintiff, not only to bankers or professional persons. See, e.g. Mutual Life and Citizen’s Assurance Co v Evatt, [1971] AC 793 (PC). The principle takes into account the factual differences between professional advisors and others.
190 UNBLJ RD UN-B [VOL/TOME 68 that were true, as the Chief Justice reminded us, it remains to justify imposing the cost of their disappointments on the defendant.116 General reliance is not situation- specific reliance between parties in a closely proximate relationship based on an assumption of responsibility that produces detrimental reliance.
Most of the recognized unique public duties in Canada are based on the general principle that once a public defendant begins to exercise a discretionary power, it then comes under a duty to exercise the power with reasonable care. I call this the Good Public Samaritan liability principle.117 A clear example is the recognized unique public duty imposed on municipalities that decide to exercise their discretionary statutory power to operate a residential home inspection program. Once the municipality implements the program, the Canadian courts require that the inspections be performed with reasonable care. This is a unique public duty. The idea that a Good Private Samaritan should incur legal responsibilities simply by beginning to offer a benefit, while one who does nothing would not, has never been established in private party negligence law,118 except in the case of professionals like doctors and lawyers.119
2017 CanLIIDocs 175 There are problems with this Good Public Samaritan liability principle. It became prominent after a simple, unsupported conclusion drawn by Lord Wilberforce in Anns.120 Neither he, nor the many Canadian judges who have
115 Invercargill City Council v Hamilton, [1994] 3 NZLR 513 (CA); aff’d [1996] 2 WLR 367 (PC). Although general reliance typically arises in public authority negligence actions, it is discussed in both the private and public sectors in Childs, supra note 7 at para 40.
116 This is similar to the argument that while it might be reasonable to rely on financial reports prepared by a national accounting firm as having been prepared with due care, that does not justify liability unless the firm did something to intend or induce the reliance. See Perry, supra note 62 at 285.
117 So far, the Supreme Court has only applied this principle where there is a safety rationale. See 194– 196, below.
118 See especially Horsely v MacLaren, [1970] 2 OR 487, 11 DLR (3d) 277 (CA) per Jessup J. See also HR Moch v Rensselaer Water, 159 NE 896 (1928) (NYCA). Whenever the Good Samaritan liability principle it is mentioned in private party negligence law it usually turns out that the decision was based on a sounder principle. See e.g. Zelenko v Gimbel Brothers, supra note 28. If it were established in the UK, presumably the unique public duty discussion in Michael would have been unnecessary. If the Good Samaritan liability rule were adopted generally as opposed to exceptionally in private law, I believe it would then be necessary to argue that the rule not apply to public defendants because it would be over- broad in application to governments who are constantly delivering public benefits. It is not necessary to deal with this complication here.
119 Later I will suggest that a good case could be made for extending this line of authority to the police. See 198, below.
120 “Passing then to the duty as regards inspection, if made. On principle there must surely be a duty to exercise reasonable care.” Anns, supra note 12 at 755. This conclusion would also seem to violate the rule now followed in England and Canada that a common law duty cannot be based on a statutory public duty alone. See supra note 18. The Good Public Samaritan rule has since been adopted regularly by the Supreme Court of Canada with no further justification or explanation. See e.g. Kamloops, supra note at 17; Just, supra note 6 at 1242–3; Rothfield v Manolakos, [1989] 2 SCR 1259 at 1266, 41 BCLR (2d) 374; Ingles v Tutkaluk Construction, 2000 SCC 12 at para 17, [2001] 1 SCR 298. There are many pre-Michael
2017] BUNGLED POLICE EMERGENCY CALLS 191 followed him, have ever offered any justification, doctrinal or otherwise, for such a rule. Anns itself has since been overruled in the UK.121 There is no apparent reason why the defendant’s voluntary provision of a gratuitous benefit, standing alone, should confer a right on an unharmed plaintiff to receive the benefit. Nor is it sound policy to discourage Good Samaritans.122 It is better to allow the public to obtain some public benefits, albeit imperfect, than none.123 That said, it is not my purpose to suggest that the housing inspection or similar highway maintenance lines of authority should be overruled. I do not foresee this happening.124 Rather, I want to use the Good Public Samaritan liability principle to illustrate a more fundamental constitutional problem with unique public duties of care.
Governments are in the business of providing public benefits: health, safety, education, transportation, housing, culture, recreation, and so on. Governments have limited budgets. Allocating a limited budget among competing claims for public benefits is a political task. Judicial supervision exists quite properly to ensure that the government respects the recognized legal rights of its citizens. These include the right not to be discriminated against on prohibited grounds, the right to be governed honestly and in good faith, the right to receive mandatory entitlements, and the right 2017 CanLIIDocs 175 to enjoy the same protections from government as from private citizens in basic tort law. Imposing a standard of “reasonable care” on the provision of discretionary benefits absent an interference with such rights is different.
What does it mean to provide a reasonable public benefit? This does not pose much difficulty in a case like Michael. Recognized public service providers like police or firefighters develop professional standards for basic services. What it means to perform a discretionary municipal house inspection, or highway inspection, with reasonable care is a different question. There is no professional standard, contract or representation to set the standard. Later, I will suggest that professional status may help found a principled distinction within the ordinary common law between public rescue providers and others.125 For now, I will concentrate on the inspection-type cases. applications of the Good Public Samaritan rule in the UK. See generally McBride, “Michael Comment”, supra note 4.
121 Supra note 15.
122 Rights theorists abhor such instrumentalist arguments. See e.g. Donal Nolan, “Revisiting the Liability of Public Authorities for Omissions” (2014) 130 LQR 21 at 24. However, it is pointless to ignore economic reality. A public authority who failed to include potential liability in predicting the cost of a discretionary program or action would be negligent.
123 See McBride, “Michael Comment”, supra note 4 at 10.
124 It is more realistic to suggest that the Good Public Samaritan liability principle will not readily be further extended. See e.g. Vlanich v Typhair, 2016 ONCA 517, 131 OR (3d) 353 [Vlanich].
125 See 199, below. As to the importance of there being a recognized standard of behaviour to justify judicial intervention, see David Cohen, “The Public and Private Law Dimensions of the UFFI Problem” (1984) 8 Can Bus LJ 410 at 421.
192 UNBLJ RD UN-B [VOL/TOME 68
For the sake of argument assume accepted standards exist within the for- profit house inspection industry and that the courts could adopt those as a definition of “reasonable.” But why should the government be required to meet a private market standard? Why, as an exercise in governance, should the government not be permitted to adopt instead a lesser standard of inspection, call it a “modest” standard.126 The government could combine this with modest benefits to tennis players who use public courts and modest benefits to library users, and so on. Or the combination could include modest benefits to home buyers and generous benefits to tennis players, and no benefits to library users. A great many different combinations are possible given the number of benefits being distributed.
When a court decides that new home owners are entitled to “industry- standard” or otherwise defined “reasonable” inspections, not the modest programs preferred by the legislature, additional costs will be imposed on the municipalities. These costs are not related to the infringements of recognized rights. These costs might encourage the defendant to stop providing the discretionary benefit altogether. They might encourage “better” inspection practices and better outcomes for some fortunate home buyers. If so, some other claimants, library users perhaps, will 2017 CanLIIDocs 175 receive fewer benefits than the municipality would have given them otherwise. Or the municipality might raise taxes. Are these outcomes more “reasonable” than the benefit combination decided upon by the municipality? What is a reasonable benefit allocation? Surely these are political questions, not justiciable questions.
A more deferential approach might be for the courts to take the “modest” benefit program as a given, but require the government to deliver that program with reasonable care. This is how some would define actionable “operational” negligence.127 The policy-operational continuum was introduced to Canadian law via Anns as a vehicle with which to determine whether the government ought to be immune from tort law in particular cases. The continuum approach has since been abandoned in the UK,128 and the quest to identify operational negligence has been abandoned in Canada.129 Liability for unreasonable program implementation is no different than liability for unreasonable program definition.130 They both amount to
126 There are different ways in which an inspection program might fall below industry standard. For example, there could be fewer inspections, less extensive inspections, or a less adequately supervised programs with a higher error rate than in the private sector.
127 This was introduced by McLachlin J as she then was at trial in Just v BC, [1985] 5 WWR 570 at 576, 64 BCLR 34 (SC).
128 As pointed out in R v Imperial Tobacco Canada Ltd, supra note 6 at para 78, the House of Lords declared the “policy/operational distinction unworkable in difficult cases, a point said to be evidenced by the Canadian jurisprudence: Stovin v Wise, [1996] AC 923 (HL), per Lord Hoffmann.”
129 See e.g. Imperial Tobacco, supra note 6 at para 78..
130 Anns itself was equivocal about whether policy implementation should be assessed on a standard of reasonable care (hence based on a unique public duty) or a standard of good faith (an entirely appropriate condition precedent for the legitimate exercise of government power). See Anns, supra note 12 at 755. Kamloops carried this equivocation into Canadian law. See Kamloops, supra note 17 at 24–25. The actual decision in Kamloops was based on improper conduct by the municipality, not negligence. See Kamloops,
2017] BUNGLED POLICE EMERGENCY CALLS 193 the court requiring the government to redefine its public benefit package, abandoning some programs, allocating more money to some and less to others. They both require the public defendant to deliver a certain standard of gratuitous service to an unharmed citizen. 131
Courts lack the institutional competence to allocate public benefits amongst competing claims.132 Two-party litigation is an inadequate vantage point from which to take all relevant considerations into account.133 But the more fundamental reason why courts should decline to become involved in allocative policy that does not infringe the rights of citizens is constitutional. Allocative policy with respect to public benefits, good or bad, is the essence of governance. Governments should be entitled to select whom to benefit and how. That is politics. Courts should make sure governments respect the public law and private law rights of citizens when governments define and deliver the benefits. That is the judicial function.
Although the argument against unique public duties derives support from rights-based torts scholars, it is a different argument that can stand by itself. Rights- based scholars object to the courts attempting to achieve distributive justice by 2017 CanLIIDocs 175 pursuing policy goals. They believe that the function of negligence law is corrective justice. Citizens enjoy a primary right to security of the person and property. A negligent defendant who interferes with these rights should be required to restore the status quo ante by paying restitutionary damages.134 The case against unique public duties, in contrast, must take as given all established private party tort duties including those based on distributive policy goals to which the rights-based scholarships would object. For example, if the common law were to recognize generally a legal duty based on a moral obligation to rescue another from injury or death, under the Diceyan approach that duty should also apply to public
supra note 17 at 24. The dissent in Kamloops did not believe the evidence supported a finding of impropriety and therefore declined to recognize a unique public duty. Since then, Canadian courts have varied in the degree to which bad faith is stressed over negligence. On the bad faith side see e.g. Froese v Hik, 78 BCLR (2d) 389, [1993] BCWLD 1405 per Huddart J as she then was, approved in Foley v Shamess, 2008 ONCA 588, 297 DLR (4th) 287. See also City of Toronto v Polai, [1970] 1 OR 483, 8 DLR (3d) 689, upheld in [1973] SCR 38, 28 DLR (3d) 638. On the simple negligence side see e.g. Oosthoek v Corporation of the City of Thunder Bay, 30 OR (3d) 323, 139 DLR (4th) 611(CA), leave to appeal to the Supreme Court of Canada dismissed September 26, 1997; and Rausch v Pickering (City), 2013 ONCA 740, 369 DLR (4th) 691.
131 In contrast, with an assumption of responsibility approach based on induced actual reliance the defendant may undertake to perform its services according to its accepted practices.
132 David Cohen & JC Smith, “Entitlement and the Body Politic: Rethinking Negligence in Public Law” (1986) 64 Can Bar Rev 1 at 10–11, and sources cited therein at n 73.
133 Ibid at 8.
134 See e.g. Beever, supra note 66 at 211; Benson, supra note 25; Ernest J Weinrib, The Idea of Private Law (Cambridge: Harvard University Press, 1995).
194 UNBLJ RD UN-B [VOL/TOME 68 defendants.135 The case against judicial policy-making in private party negligence is a different case that must be made separately. It need not be made at all for the purposes of this article.
We can look at this another way. When a court imposes negligence liability on policy grounds it will affect the distribution of wealth. It is understandable that some people believe that courts and legislatures should each have a role to play in effecting distributive justice generally. Unique public duties, however, affect the distribution of wealth in a particular way, by affecting the distribution of government benefits. Government benefits, by definition, are created and delivered by the government. Citizens enjoy substantial rights against their governments. The courts’ job is to enforce those rights, not to supervise discretionary benefits beyond this.
The principled objection to unique public duties is grounded in Dicey’s equality principle. There are also policy concerns. Courts do continue to employ the Good Public Samarian principle in new situations when it suits them.136 There might come a tipping point at some stage. However, I do not base my concerns about unique public duties on any expectation that they will cause an imminent 2017 CanLIIDocs 175 constitutional or financial crisis. The courts are usually very cautious about extending the Good Public Samaritan liability principle as far as formal logic would otherwise take it. This necessary judicial deference leads to the definition and distribution of public benefits on an apparently ad hoc basis. In so doing the courts undermine their own credibility.
The doctrine of stare decisis requires courts to treat like cases alike.137 There are literally thousands of public benefits regularly conveyed to citizens by their governments. Many are distinguishable from the inspection situation. Many are not. Strictly speaking, the courts ought to extend the Good Public Samaritan liability principle across the board to any situation in which the government operates a public benefit program that is in principle similar to the inspection benefit. Thanks to the good sense of most judges, this has not happened. If it did, many, perhaps most government benefits would have to achieve a court-defined standard of reasonable care. Public administration would become overwhelmingly judicialized. The existing division of powers between the courts and the legislatures would not merely shift, it would be entirely redefined.
Voluntary judicial restraint in employing the Good Public Samaritan principle across the board is necessary and commendable, but it comes with a cost. It
135 Tofaris and Steel make this argument to support a unique duty of care. See “Omissions”, supra note 36 at 142–145. Later I will suggest the same outcome can be accomplished via basic common law negligence.
136 See e.g. Fullowka, supra note 17; Rausch, supra note 130.
137 True, this is an uncertain and fungible exercise. Some would argue stare decisis is more an exercise in justification than in logic. Regardless, there are limits to the fungibility. For example, I suggest these limits were exceeded in Vlanich, supra note 124.
2017] BUNGLED POLICE EMERGENCY CALLS 195 is difficult, perhaps impossible, to predict or justify when the Good Public Samaritan principle will be invoked and when it will not.138 For example, the Supreme Court has recognized unique public duties in the areas of highway safety,139 workplace safety,140 and building construction.141 There seems to be a “safety” theme. However, Health Canada does not owe a duty to consumers when it approves for sale defective medical devices such as jaw or breast implants.142 Why is there proximity between the police or highway maintenance crews and ordinary road users, even in the absence of specific reliance, but not between consumers of medical devices who did specifically rely on the defendants’ product approval posted on their website? Why are highway accidents, where insurance cover is prevalent, or structural building defects that are easily discovered by consumers hiring a private inspector, more worthy of exceptional intervention than other cases like the medical device cases? No principled answers to these questions appear in the reasons for judgment. The Canadian exceptions appear to be purely ad hoc.143 This is damaging to the structure of the common law.144
5. Difficulties with Unique Public Duties in Exceptional Circumstances 2017 CanLIIDocs 175
Suppose that one accepts the aversion to unique rules of general application for public defendants. Should there be exceptions in exceptional cases? The substantive case against unique public duties is based on a particular view about the distinction between the legislative function and the judicial function. I cannot identify an exceptional case that would not violate this distinction. Therefore, I would prefer a
138 Unfairness among potential beneficiaries is a problem. See McBride and Bagshaw, supra note 33 at 218.
139 Schacht, supra note 17; Just, supra note 6.
140 Fullowka, supra note 17
141 Kamloops, supra note 17.
142 Drady v Canada, 2008 ONCA 659 at para 52, 300 DLR (4th) 443, leave to appeal refused [2008] SCCA No 492; and Attis v Canada, 2008 ONCA 660 at para 77, 93 OR (3d) 35, leave to appeal refused [2008] SCCA No 491.
143 See Erika Chamberlain “To Serve and Protect Whom? Proximity in Cases of Police Failure to Protect” (2016) 53 Alta L Rev 977 [Chamberlain, “Serve and Protect”] where the author exposes several poorly reasoned lower court decisions about police duties to protect victims of crime.
144 Consider the dilemma the court faced in Vlanich, supra note 124. The plaintiff relied on the unique inspection cases and the underlying Good Public Samaritan liability principle to argue that the defendant owed it a duty to enforce its bylaws with reasonable care. The Court of Appeal felt compelled to “distinguish” the inspection cases by stating that the inspection authorities had “invited the injured party to rely on an inspection, and it has assumed responsibility for avoiding the risk.” See para 32. Of course, the inspection duties were not based on either induced transaction-specific reliance or an assumption of responsibility. Suggesting otherwise was probably the only way in which the Court of Appeal could reconcile the lack of proximity with the doctrine of stare decisis.
196 UNBLJ RD UN-B [VOL/TOME 68 clean and clear across-the-board prohibition against unique public duties.145 However, others will prefer to draw the line differently in exceptional cases based on the quest for “good” judicial interventions into what otherwise would be the legislative sphere. I suggest that acceptable exceptions must 1) distinguish meaningfully those cases where an exception is justified; and 2) keep the exception sufficiently narrow that the existing division of powers between the legislative and judicial branch is not radically altered.146 It may be possible to meet these conditions, but I have only seen one, a proposal by Tofaris and Steel, that has done so.147 It justifies the exception based on important differences between the status of the police and that of ordinary citizens. Ironically, their proposal is unnecessary. I conclude this section by suggesting that any plaintiff who can satisfy the Tofaris and Steel conditions for a unique public duty would succeed, probably more easily, under basic Canadian private party negligence law.
The Canadian unique public duty cases tend to be safety-based, intended to reduce personal injury and death. The Canadian Charter of Rights and Freedoms provides a strong precedent for prioritizing personal security over property rights.
General negligence law recognizes fundamental distinctions between physical 2017 CanLIIDocs 175 damage and pure economic loss based on both principle and policy.148 There seems to be little judicial appetite for expanding recovery for pure economic loss beyond its relatively limited scope today.149 All unique public duties effectively create a limited taxpayer-funded insurance scheme.150 The arguments for having to insure someone in Ms. Michael’s position are compelling. The arguments for insuring someone’s interest in receiving a gratuitous financial benefit are less so, especially if they are effectively able to obtain such protection themselves.151 Refusing to recognize novel unique public duties concerning public benefits relating to property or purely economic interests would be a step in the right direction.152
Recall that governments will routinely be held liable for misfeasance that interferes with the right to personal security. Unique public duties would be directed
145 I would also prefer to employ this approach instead of the immunity approach.
146 These are limits I would propose for unilateral action by the courts. The legislature is free, subject to claims of right, to select which constituencies it wishes to benefit and which not.
147 Tofaris & Steel, “Police Liability”, supra note 36; and Tofaris & Steel, “Omissions”, supra note 36.
148 See e.g. Benson, supra note 25 at 865; Beever, supra note 66 at 214; Canadian National Railway Co v Norsk Pacific Steamship Co, [1992] 1 SCR 1021, 91 DLR (4th) 289 per La Forest J dissenting, eventually adopted by the court in Bow Valley Husky (Bermuda) Ltd v Saint John Shipbuilding Ltd, [1997] 3 SCR 1210, 153 DLR (4th) 385.
149 Recovery for one type of pure economic loss, dangerously defective housing construction, is well- entrenched in the inspection cases. This would have to be carved off somehow.
150 See e.g. Cooper, supra note 16 at para 55.
151 There is a significant difference in being able to purchase a standard home inspection and being able to replace police emergency services on the private market. 152 See Vlanich, supra note 124 at para 31.
2017] BUNGLED POLICE EMERGENCY CALLS 197 to the failure to take reasonable steps to protect citizens from such interferences in advance. The problem is that governments are deeply involved in benefit programs with a safety rationale. In order to avoid judicializing an enormous range of government power, we cannot accept a general health and safety justification, or even a safety justification alone, for unique duties of care. We have to further narrow the justification. It is, however, difficult in a common law system to distinguish in principle an imminent risk from a long term risk, or a safety rationale from a health rationale. The home inspection duty allows recovery of pure economic loss on a safety rationale. Based on stare decisis this could be extended. It remains to be seen whether some safety-based benefits can be distinguished in principle from others.
One possibility is to attempt to draw a distinction between “targeted” public duties and duties that exist for the benefit of the general public. This distinction was adopted by the Supreme Court of Canada in Cooper v Hobart153 and is frequently invoked by the courts. It may be a principled distinction, but it is totally conclusory. Every duty derived from a statute has a general public aspect, and, like in Cooper, a class of persons who are more directly affected than members of the general public.
Is the purpose of the emergency service in Michael targeted towards potential 2017 CanLIIDocs 175 victims of crime, or towards crime prevention generally?154 Does that question really help to distinguish Cooper from Michael? If the legislature intends to confer private benefits on targeted individuals, it should create statutory torts to that effect. The search for targeted duties adds nothing to the discussion.155
A proposal by Tofaris and Steel to adopt a particular duty of care owed by the police to protect citizens from the criminal acts of others has received favorable attention in the UK.156 In 2014 the authors stated their proposition as follows:157
. . . a finding of proximity should arise where the following factors are satisfied: (i) The claimant is at a special risk of personal harm, i.e., a greater risk than the general public. The circumstances in which the risk will be special must be left to the courts to develop on a case-by-case basis. Guidance on this can be found in the New Zealand case Couch v Attorney-General, where the majority held that “the necessary risk must be… special in the sense that the plaintiff’s individual circumstances, or her membership of the necessary class rendered her particularly vulnerable to suffering harm
153 Cooper, supra note 16 at para 38.
154 Tofaris and Steel claim it is the former. See “Omissions”, supra note 36 at 154–155. I suggest there is no possible way to answer the question definitively.
155 See also Nolan, supra note 122.
156 The recommendations made in Tofaris & Steel, “Police Liability”, supra note 36 at 5 were quoted in dissent by Lady Hale in Michael, supra note 1 at para 189. See also at para 197 per Lady Hale, and at paras 178–181 per Lord Kerr dissenting.
157 Tofaris & Steel, “Police Liability”, supra note 36 at 5.
198 UNBLJ RD UN-B [VOL/TOME 68
of the relevant kind” from the third party. In any case, there is no doubt that a person facing a specific threat to her physical safety from a specific individual is at a special risk. (ii) The police are aware or should have reasonably been aware that the claimant is at a special risk of personal harm. (iii) The police are given special powers by law to protect the class of persons to which the claimant belongs, i.e., members of the public at a special risk of physical harm. (iv) The claimant is dependent upon the police as regards protection against the risk on the basis of the legal and civic duties imposed on her to inform the police about the incident and to refrain from taking measures beyond reasonable self-protection and/or her vulnerability in the sense that she cannot be reasonably expected to protect herself adequately against that risk.
Inexplicably, the 2016 version of their proposal omits the following requirement: “and/or her vulnerability in the sense that she cannot be reasonably expected to protect herself adequately against that risk.”158
2017 CanLIIDocs 175 This is a “good” proposal, excellent according to my untrained eye. It deals only with special risks of personal harm, a type of risk with more moral and legal significance than property damage or economic loss. Restricting the duty to police, and to a limited class in a limited set of circumstances, certainly minimizes the scope of the judicial intrusion into the legislative realm. But my objection to unique public duties is not based on whether or not they are “good” duties. Some are, some are not, and usually it is difficult to tell.
One may quibble with the proposal. The more specific we are in defining the exception, the more likely that we will exclude other situations that are not distinguishable in principle. Lords Toulson’s and Kerr’s sparring over several options in Michael illustrates this. Why are dangers posed by a third party different from other risks to life and limb? Why police, but not other emergency responders like firefighters? Does an imminent safety risk warrant different treatment from an imminent health risk, for example, or even a long term risk of industrial disease? One may still ask why, unless the police were in some way specifically responsible for the risk of criminal attacks by others, ought there to be a unique public duty? How confident are we that the proposed rules will actually increase the degree of protection the police currently offer to potential victims of criminal attacks?159
Despite these quibbles, I find it difficult to argue strenuously against liability under the Tofaris and Steel conditions. The more interesting question is whether this specific and narrowly crafted unique public duty is necessary, at least in Canada. I suggest not. Basic common law negligence should be able to accomplish the same goals.
158 Tofaris & Steel, “Omissions”, supra note 36 at 151.
159 For a fascinating take on this problem see Margaret Hall, “Theorizing the Institutional Tortfeasor” (2016) 53 Alta L Rev 995.
2017] BUNGLED POLICE EMERGENCY CALLS 199
Tofaris and Steel propose their novel police duty against the background of botched emergency responses such as those in Michael. As previously discussed, that particular situation can be effectively resolved by the basic rules of Canadian negligence law governing an assumption of responsibility. Canadian courts already require a special relationship of proximity between the citizen and the public defendant that is closer than the relationship between the defendant and the general public or even from other members of the class most directly affected by the public benefit at issue.160 Near privity is required in a negligent misrepresentation action. The most the plaintiff should have to establish is that the defendant intended, induced or invited her to rely on the defendant; that she reasonably and foreseeably did so; and possibly that she suffered foreseeable detrimental reliance loss as a consequence. The Tofaris and Steel conditions would be more demanding of claimants.
It is difficult to imagine that a party who reaches a police emergency line and sets out a request for police protection from imminent physical harm or death could fail to meet the first two Tofaris and Steel conditions. The authors believe their proposal goes further than the common law because it would apply in the case where 2017 CanLIIDocs 175 the emergency operator was listening to music and simply did not answer the call. I would argue that situation could also culminate in a duty of care under basic Canadian negligence law. By providing the service the force is assuming responsibility by intending and inducing a limited and vulnerable class of persons whose personal safety is at risk to rely on the service being provided with reasonable care. Listening to music instead of answering emergency calls breaches that duty. Refusing to answer emergency calls without relevant justification should also be actionable as “bad faith.”161
It is a further question whether the plaintiff’s recovery should be limited to detrimental reliance loss, or extended to the full loss of the benefit. By analogy to the exceptional professional duties of affirmative action one could argue it is a standing promise to provide to provide the benefit, open to being accepted to create a “contract without consideration” unless and until it is withdrawn. As noted in Michael:
Every person who enters into a learned profession undertakes to bring to the exercise of it a reasonable degree of care and skill. He does not undertake, if he is an attorney, that at all events you shall gain your case, nor does a surgeon undertake that he will perform a cure, nor does he undertake to use the highest possible degree of care and skill.162
160 Cooper, supra note 16 is the leading example. There is an expansive summary of the law on point in Taylor v Canada, supra note 22 at paras 75–91. The exceptions are the unique public duty cases identified supra note 17.
161 See Anns, supra note 12 at 102.
162 Michael, supra note 1 at para 178 quoting Tindal CJ in Lanphier v Phipos (1838) 8 C & P 475, 479.
200 UNBLJ RD UN-B [VOL/TOME 68
This principle has been extended to lawyers, doctors, other health care providers163 and ambulance service providers.164 It has not been extended to police or firefighters, but it could be. As noted above, with recognized rescue professionals such as the police there should be no difficulty in determining what a reasonable police officer ought to have done.165
The Tofaris and Steel proposal covers more ground than assumption of responsibility where the plaintiff is unaware of the special danger, but the police department is. The alleged negligence would be the failure to warn the plaintiff, or to control the potential assailant. General Canadian common law recognizes numerous exceptions to the “no duty” rule based on special relationships of control.166 I suspect a good personal injury lawyer, especially one working in a “no unique public duty” jurisdiction, would be happy to bring a case against the police seeking to expand this line of authority beyond a duty to control a person already in custody.167
Tofaris and Steel’s proposal is specific to the police, and they emphasize that the “special status” of the police is what justifies their proposed unique police 168 duty. My preferred approach is to work with basic common law negligence and to 2017 CanLIIDocs 175 allow the flexibility of the common law to take into account special facts relevant to the status of the public defendant when applying the general principles. I believe that what drives Tofaris and Steel to the preference for a unique public duty is the need to circumvent a narrow definition of assumption of responsibility specific to the police in the UK.169 There are no police-specific definitions of assumption of responsibility in Canada. This makes our ordinary common law more adaptable to the special arguments in favour of police liability in the types of circumstances envisaged by Tofaras and Steel. Neither Tofaris and Steel’s proposal nor basic private party
163 Michael, supra note 1 at para 112.
164 Ibid at para 81.
165 See 190–191, above.
166 See 175, above. A case like Jane Doe, supra note 43, absent the Charter element, might also be explained along these lines.
167 This is discussed in McBride & Bagshaw, supra note 33 at 245 where the authors also refer to John Goldberg & Benjamin Zipursky “Intervening Wrongdoing in Tort: The Restatement (Third)’s Unfortunate Embrace of Negligent Enabling” (2009) 44 Wake Forest L Rev 1211 at 1240, n 121 in support of an expanded notion of control beyond custody. Chamberlain, “Serve and Protect”, supra note 143 cites several lower court decisions that take a very expansive view along these lines. We should also keep in mind that there has long existed in common law support for a “Bad Samaritan” liability rule that would impose liability for the failure to perform an easy rescue from a situation threatening life or serious bodily harm to another. Such provisions do exist in many civil law jurisdictions. See, e.g. Allen Linden, “Toward Tort Liability for Bad Samaritans” (2016) 53 Alta L Rev 837.
168 Tofaris & Steel, “Omissions”, supra note 36 at 129.
169 Ibid at 150. Tofaris and Steel also discuss their preference for unique duty over assumption of responsibility at Tofaris & Steel, “Police Liability”, supra note 36 at 23–4 based on the debate about whether a responsibility is truly assumed or imposed by law in the UK.
2017] BUNGLED POLICE EMERGENCY CALLS 201
Canadian negligence law explain or justify the decisions in the classic unique public duty cases decided by the Supreme Court of Canada.170
6. Conclusions
Unique public duties pose more problems than they solve. They allow the courts on their own initiative to shift classic governance functions to the courts. They damage the structure of the common law. Most importantly, where they appear to be most needed, in Michael for example, they are unnecessary. Private party negligence can address the issues, and do so in a way that takes into account the unique aspects of the government’s role in perpetrating the alleged wrong. There are also other options. Perhaps a better answer lies in public law?171 If unique responses to police failure to address domestic violence are necessary, the legislature should provide them. There is no constitutional objection to the legislature creating a statutory cause of action to deal with a duty to prevent crime generally, or a duty to prevent domestic violence particularly. The complaint under the Human Rights Act, 1998 for breach of the defendants' duties as public authorities to protect Ms. Michael's right to life under 2017 CanLIIDocs 175 article 2 of the European Convention on Human Rights was allowed to proceed.172 A Canadian plaintiff might succeed in an action for damages under the Charter.173 The police misconduct in Michael was the subject of a damning public inquiry, discipline and in one case dismissal.174 We should resist the temptation to resolve Ms. Michael’s case with a unique public duty. Hard cases make bad law.
I do not purport to have offered the definitive word on unique duties of care. Nor do I believe that Canadian judges have yet developed principled guidelines for creating unique duties of care. A full and open discussion would be welcome.
170 Supra note 17.
171 Nolan, supra note 25 at 291 quoting Beever, “Rediscovering”, supra note 66 at 340. See also Paradis Honey v Canada, 2015 FCA 89 at paras 119–146, 382 DLR (4th) 720.
172 The implications of this are considered in McBride, “Michael Comment”, supra note 4.
173 See e.g. Jane Doe, supra note 43 and Dudley v British Columbia, 2013 BCSC 1005, 49 BCLR (5th) 382 [Dudley]. This may be a better way to attack systemic wrongdoing.
174 This was not the case in Dudley, ibid, nor in Mooney, supra note 9. See Chamberlain, “Optimism”, supra note 9 at para 33. Police accountability is one of the reasons Tofaris & Steel, “Omissions”, supra note 36 offer in support of a unique police duty of care. It is possible to do this otherwise.
THE COMPLICATED INTERSECTION OF POLITICS, ADMINISTRATIVE AND CONSTITUTIONAL LAW IN NUNAVUT’S ENVIRONMENTAL IMPACTS ASSESSMENT REGIME
Daniel W. Dylan1
INTRODUCTION ...... 202 I. BRIEF HISTORY OF NUNAVUT AND THE NLCA ...... 204 II. SUMMARY OF IMPACTS ASSESSMENT REGIME ...... 206 III. LAND USE PLAN CONFORMITY ...... 212 A. MINOR VARIANCES ...... 213 B. AMENDMENTS ...... 213
C. LAND USE PLAN EXEMPTIONS ...... 214 2017 CanLIIDocs 175 IV. IDENTIFICATION OF THE PROBLEM(S) BY EXAMPLE ...... 215 A. BAFFINLAND IRON MINES CORP AND THE MARY RIVER IRON ORE PROJECT ...... 215 B. AREVA AND THE KIGGAVIK PROJECT...... 225 C. CLYDE RIVER AND SEISMIC TESTING ...... 227 V. WHY THESE ISSUES ARE IMPORTANT ...... 229 CONCLUSION...... 230
INTRODUCTION
In Canada, it is somewhat unique that environmental impacts assessment is carried out by both the federal and provincial governments, under separate but complimentary assessment regimes. Even more unique is the environmental impacts assessment regime which exists in the territory of Nunavut, Canada’s youngest jurisdiction. What makes environmental impacts assessment most unique in Nunavut is the complicated intersection of land use planning, administrative rules and procedures derived from a constitutionally-protected modern day treaty—the Nunavut Land Claims Agreement (NLCA or Agreement),2 and the infusion of fundamental, rudimentary or bare politics. Whereas the majority of jurisdictions in Canada base their environmental impacts assessment regimes on a commonly
1 LLM, JD (United States), LLB (Canada), BA (Hons); Assistant Professor of Law, Bora Laskin Faculty of Law, Lakehead University, Thunder Bay, Ontario, Canada.
2 Agreement Between the Inuit of the Nunavut Settlement Area and Her Majesty The Queen in Right of Canada, 25 May 1993, online:
2017] NUNAVUT’S ENVIRONMENTAL IMPACTS ASSESSMENT REGIME 203 developed and accepted framework that divides assessment based on jurisdiction and project type—and leave land use planning matters to provincial municipalities—the regime in Nunavut is based almost entirely on a process laid out in the NLCA and has as its aims the preservation of the Aboriginal rights and interests of Inuit beneficiaries protected by and under the Agreement. Upon closer scrutiny, however, it is apparent that the environmental impacts assessment and land use planning regimes prescribed in the NLCA, and supplemented by the federal Nunavut Project Planning and Assessment Act (NUPPAA),3 are not only significantly different from other regimes in Canada but, also, that these very regimes which were designed or intended to promote and protect those rights and interests which Inuit obtained through negotiation and ratification of the NLCA may in fact undermine and fragment them. That is to say, the regimes may fragment those participatory rights as well as neighbouring or undergirding ones such as wildlife, hunting and fishing rights.
The fragmentation of these rights and interests is aptly illustrated by the NLCA process in which a Minister may grant a project proponent an exemption from land use plan conformity, land use conformity being the crucial first obstacle that 2017 CanLIIDocs 175 nearly every resource development project in Nunavut must overcome.4 By looking at three real-world examples: Baffinland Iron Mines Corp.’s Mary River Iron Ore Project, Areva Resources Canada Inc.’s Kiggavik Project, and the Clyde River case heard by the Supreme Court of Canada in November 2016,5 this paper will canvass the Ministerial land use exemption process, illustrate the intersection of politics, administrative and constitutional law in Nunavut’s impacts assessment regime, and explicate the manner in which Inuit rights and interests are being resultantly fragmented.
It is my thesis that while the NLCA provides to Inuit in Nunavut participation in decision-making processes respecting projects and their resulting environmental and socio-economic impacts, it does not provide final decision- making power itself, and that the absence of such decision-making power undermines—even dilutes—the very rights, interests, and culture which this participation and the NLCA was designed to promote and protect. Stated another way, the Inuit of Nunavut, by virtue of the NLCA, or lack thereof, seem to enjoy no greater right to accommodation under the duty to consult jurisprudence than do other Aboriginal groups in Canada. This circumstance is only further exacerbated where politics collides with administrative and constitutional law in Nunavut’s environmental impacts assessment regime.
3 Nunavut Planning and Project Assessment Act, SC 2013, c 14, s 2 [NUPPAA].
4 NLCA, supra note 2 at arts 11.5.11 – 11.5.13.
5 Hamlet of Clyde River v TGS-NOPEC Geophysical Company ASA (TGS), SCC Judgement Reserved, 2015 FCA 179, 474 NR 96 [Clyde River].
204 UNBLJ RD UN-B [VOL/TOME 68
The legal issues explored here are not, however, the product of incomplete legal research, and are canvassed because of the practical effects they currently have in the territory and the possibility that they may produce serious, complex, and unnecessary litigation in the future.6
Equally relevant to the issues explored in this article is the absence of any statutory or policy guidance on the exemption process (and the other jurisdictional questions presented here) to help resolve the small but crucial ambiguities in the statutory text which are, in way or another, likely to have significant and long-term ramifications for the Inuit of Nunavut and the environment in Nunavut. As development opportunities and projects continue to emerge in Nunavut, the statutory interpretation problems have the potential to become acuter and to further manifest in different, non-mutually exclusive ways.7
This paper will proceed in five parts. I begin by providing a brief history of Nunavut and the NLCA in order to contextualize my analysis. Part II follow this with a brief summary of the environmental impacts assessment regime in Nunavut in order to properly situate the examination of land use plan exemptions. In Part III, I 2017 CanLIIDocs 175 examine the land use conformity regime and then in Part IV I illustrate the problems with the examples mentioned earlier. In the final part, I offer some insight into why these emerging issues are important. Altogether, while the intersection of politics and administrative and constitutional law in an environmental impacts assessment regime is not unique to Nunavut, I aim to show that the Inuit rights and interests which the Nunavut regime was designed to protect are ultimately being fragmented in the regime’s current incarnation.
I. BRIEF HISTORY OF NUNAVUT AND THE NLCA
In Inuktitut, the language of the Inuit in the eastern Arctic,8 Nunavut means “our land.” Nunavut, Canada’s newest jurisdiction and territory, was created by several constitutional and statutory instruments. The Nunavut Land Claims Agreement, signed with apparent finality in 1993, settled the land claim of the Inuit of Nunavut9
6 See e.g. Nunavut Tunngavik Inc v Canada (AG), 2014 NUCA 02, 580 AR 75 [Nunavut Tunngavik].
7 In this paper I argue that the issues which I discuss have the potential to produce serious, complex and unnecessary litigation in the future, but the possibilities that the ambiguities can be exploited by project proponents to maximize their chances of receiving an exemption, or that the ambiguities allow government officials to champion development to the detriment of the environment, or that the ambiguities allow the federal government to consistently override Nunavut or Inuit interests all have the potential to manifest as problems.
8 Official Languages Act, RSNWT 1988, c O–1 (section 4 recognizes Cree, Chipewyan, Dogrib, Gwich’in, Inuktitut, North Slavery and South Slavery, French and English as the Official Languages of Nunavut), see also Inuit Language Protection Act, S Nu 2008, c17.
9 The Inuit of Nunavut are to be distinguished from the Inuit of Nunavik, for example, and from other Inuit Peoples in Canada.
2017] NUNAVUT’S ENVIRONMENTAL IMPACTS ASSESSMENT REGIME 205 against the Government of Canada.10 As such, it is considered a modern-day treaty and is constitutionally protected by section 35 of the Constitution Act, 1982.11 Generally speaking, the Agreement sets out the various terms and conditions in which the Inuit of Nunavut agreed to surrender any future claim to Aboriginal title in Nunavut in exchange for title to defined parcels of land, rights respecting wildlife and natural resource management, among others.12 The Agreement consists of 42 Articles, including ones devoted to, for example, wildlife, parks, conservation areas, land use, development, water rights, impact benefit agreements, among others, and is approximately 300 pages in length. The Nunavut Land Claims Agreement Act gives legal force to the NLCA,13 and the Nunavut Act, legally creates Nunavut and gives it legal existence in Canada’s constitutional order.14 As a result, Nunavut legally joined the Canadian federal-provincial-territorial family in 1999.
Geographically, Nunavut consists of approximately 20% of Canada’s land mass and demographically is inhabited by approximately 37,146 people comprising approximately 0.10% of the Canadian population as a whole.15 Of this population, according to the Government of Nunavut Bureau of Statistics, approximately 81% 16 are Inuit. The majority of the land that comprises Nunavut is Crown land and is 2017 CanLIIDocs 175 therefore under the dominion and jurisdiction of the federal government. Some lands have been ceded by the Crown to the territorial government, the Government of Nunavut, and are known as “Commissioner’s Lands,” and are essentially governed by the territorial Commissioner’s Lands Act.17 The Inuit of Nunavut own title to 19% of the land in Nunavut, including mineral rights to 2% of Nunavut.18 This means that Inuit own title to, in similar fashion to a fee simple, surface rights to nearly twenty percent of the territory and only surface and subsurface rights in two percent. The lands which are owned by the Inuit of Nunavut are held in trust by Nunavut
10 NLCA, supra note 2 (the Preamble to the Agreement provides: “…the Parties agree on the desirability of negotiating a land claims agreement through which Inuit shall receive defined rights and benefits in exchange for surrender of any claims, rights, title and interests based on their assertion of an aboriginal title…”; see also Nunavut Tunngavik, supra note 6.
11 Constitution Act, 1982, s 35, being Schedule B to the Canada Act 1982 (UK), 1982, c 11.
12 NLCA, supra note 2 at art 2.7.1.
13 NLCA Act, supra note 2.
14 Nunavut Act, SC 1993, c 28.
15 Nunavut, Nunavut Bureau of Statistics, Stats Update: “Total Population by Inuit and Non-Inuit for Nunavut, Region and Community, 2001 to 2016, as of July 1”, (Iqaluit: Bureau of Statistics, 2016) online:
16 Ibid.
17 Commissioner's Land Act, RSNWT 1988, c C–11.
18 Nunavut Tunngavik Incorporated, “Explore the Potential of Inuit Owned Lands” (March 2011) Nunavut Tunngavik Incorporated (blog), online:
206 UNBLJ RD UN-B [VOL/TOME 68
Tunngavik Inc. (NTI) and the Regional Inuit Associations (RIA) on behalf of and for the benefit of all Inuit in Nunavut,19 and are generally known and referred to as “Inuit Owned Lands” or “IOL”.
The Nunavut Act creates for Nunavut a Legislative Assembly similar in principle to that of a provincial legislature. The Legislative Assembly of Nunavut is populated with members elected by constituents in Nunavut’s 22 ridings or constituencies. Although not entirely unique in Canada but nevertheless rare, members of the Legislative Assembly are not elected as representatives of any one or other Canadian political party (although they may be members of one) but essentially as independent members. The independent members of the Legislative Assembly select among themselves who should be designated a member of the Executive Council (the equivalent of the Governor in Council or Cabinet) and who, ultimately, will be the Premier of Nunavut. Once selected, the Premier of Nunavut assigns the various government portfolios to the various members who have been selected to serve on the Executive Council. Overall, the intent of the governance structure is to provide the framework for a consensus government.20
2017 CanLIIDocs 175 The Government of Nunavut is not, however, composed as an ethnic government of the Inuit of Nunavut, though Article 23 of the NLCA is devoted to ensuring employment for Inuit in Government of Nunavut positions.21 Rather, the Government of Nunavut is a public government and is constitutionally obligated to act like any other public government in Canada. Nevertheless, it can be said, to some degree at least, that the Inuit of Nunavut enjoy some sort of self-determination via the territorial government.22
II. SUMMARY OF IMPACTS ASSESSMENT REGIME
The preamble to the Agreement provides, in part, that:
Inuit shall receive defined rights and benefits in exchange for surrender of any claims, rights, title and interests based on their assertion of an aboriginal title…and...have negotiated [the] Agreement based on and reflecting…certainty and clarity of rights to ownership and use of lands and resources, and of rights for Inuit to participate in decision-making
19 Ibid (“The Regional Inuit Associations (RIAs) – Kitikmeot, Kivalliq and Qikitani – were designated as the Inuit Organizations in which surface title to Inuit Owned Lands in each representative region would vest. The RIAs administer access through the issuance of Land Use Licences and Surface Leases as well as other forms of authorization. Where subsurface title to Inuit-owned lands is held by Inuit, it is vested in NTI. Inuit Owned Lands are held in trust by NTI and the RIAs on behalf and for the benefit of all Inuit.”).
20 See David M Brock, “Power in Consensus Government”, Policy Magazine (July/August 2014), online:
21 NLCA, supra note 2 at art 23.
22 See Jack Hicks & Graham White, “Nunavut: Inuit self-determination through a land claim and public government?” in Jens Dahl, Jack Hicks & Peter Jull, eds, Nunavut: Inuit Regain Control of Their Lands and Their Lives (Copenhagen: International Work Group for Indigenous Affairs, 2000) 30 at 117.
2017] NUNAVUT’S ENVIRONMENTAL IMPACTS ASSESSMENT REGIME 207
concerning the use, management and conservation of land, water and resources, including the offshore…23
Of particular interest and relevance here, in the present context, is the phrase “...of rights for Inuit to participate in decision-making concerning the use, management and conservation of land, water and resources...”
The rights for Inuit to participate in decision-making concerning the use, management, and conservation of land, water and resources, including the offshore manifests most vividly in Article 11 of the NLCA, devoted to land use planning, and Article 12 of the NLCA, devoted to development impact. Both Articles make up the most significant parts of the environmental impacts assessment regime in Nunavut. I canvass both by explicating how a project in Nunavut comes to fruition, or to the contrary, does not come to fruition, once having undergone the environmental impacts assessment process outlined in the NLCA and, following that, to a limited degree, comparatively in NUPPAA.24 More specifically, I aim to illustrate how the rights of Inuit to “participate” in decision-making concerning the use, management
and conservation of land, water and resources are, in current operation effectively 2017 CanLIIDocs 175 circumscribed.25 Stated another way, while Inuit have the right to participate in decision-making, the actual power to decide which projects in Nunavut come to fruition and which do not, at law, rests with Ottawa, not with the Inuit of Nunavut. That being the case, the neighbouring rights which the regime is meant to protect, such as those pertaining to wildlife, for example, in turn, are also—perhaps inevitably—fragmented and diluted.
The environmental impacts assessment process in Nunavut is trigged when a project proponent (proponent) develops a project proposal (proposal). The proposal must adhere to a specific set of criteria set out by the Nunavut Impact Review Board (NIRB or Board), under the NLCA, and in a similar fashion, if applicable, to certain criteria specified under NUPPAA.26 The first step of the process requires the proponent to submit the proposal to the Nunavut Planning Commission (NPC or Commission), the administrative body responsible for land use planning in Nunavut. In contrast, the NIRB is the administrative body responsible for screening and reviewing project proposals for eco-systemic and socio-economic impacts in Nunavut.27 Both are known in Nunavut as Institutes of Public Government (IPG), as
23 NLCA, supra note 2 at Preamble (this part of the preamble states similarly for wildlife harvesting and management: “to provide Inuit with wildlife harvesting rights and rights to participate in decision- making concerning wildlife harvesting…”).
24 NUPPAA, supra note 3.
25 NLCA, supra note 2 at art 1 defines resources as “…for the purpose of Articles 25 to 27, coal, petroleum, precious and base metals and other naturally occurring substances that can be mined, but does not include specified substances.” Notably it does not include wildlife or wildlife habitat.
26 NUPPAA, supra note 3, was only declared in force on July 9, 2015 and thus is not applicable to very many current projects in Nunavut.
208 UNBLJ RD UN-B [VOL/TOME 68 are the Nunavut Water Board (NWB), the Nunavut Wildlife Management Board (NWMB), and the Nunavut Surface Rights Tribunal (NSRT).
Upon receipt of the proposal by the NPC from the proponent, the Commission must determine whether the project proposed in the proposal conforms to an applicable land use plan. While the NPC has, for many years now, been developing a Nunavut Draft Land Use Plan (essentially a Nunavut-wide land use plan), it has not yet completed it.28 The only land use plans currently operational in Nunavut are the North Baffin Region Land Use Plan (NBRLUP) and the Keewatin Regional Land Use Plan (KRLUP).29 If the project proposed by a proponent in the proposal is not geographically located in an area which is covered by either of these two existing land use plans, the NPC will confirm that no land use plan exists that is applicable to the project and simply forward the proposal to the NIRB for an impacts assessment screening/review. If, however, a land use plan is applicable to the project proposed in the proposal, then the NPC will make a determination as to whether or not the proposed project conforms to that land use plan. Should the NPC determine that the proposal does not conform to the applicable land use plan the NPC will not 30 forward the proposal to the NIRB. Should the NPC determine that the project is not 2017 CanLIIDocs 175 in conformity with the applicable land use plan, a number of options become available to the proponent to overcome the non-conformity or negative conformity determination. I will discuss these options in greater detail later, and one of them is, in fact, the precise focus of this article: ministerial exemptions.31 Should, on the other hand, the NPC determine that the project proposed in the proposal conforms to the applicable land use plan it will forward the proposal to the NIRB.32
Upon receipt of the proposal by the NIRB from the NPC, the NIRB will undertake an initial assessment of the proposal and determine an appropriate method of screening/review for the proposal.33 First, it may (for a variety of reasons but
27 See NLCA, supra note 2 at art 12.2.3. The extent to which NIRB can impose socio-economic mitigation requirements is very limited. 28 NLCA, supra note 2 at art 11.5.1 (“A Nunavut land use plan shall be formulated by the NPC in accordance with Section 11.5.4 to guide and direct short term and long term development in the Nunavut Settlement Area. Regional or sub-regional components of the land use plan shall be implemented where approved pursuant to Section 11.5.9.”).
29 See Nunavut Planning Commission, “Approved Plans,” (March 2017), Nunavut Planning Commission (blog), NPC online:
30 NLCA, supra note 2 at art 12.3.4 (“NIRB shall not screen project proposals that are not in conformity with land use plans, unless an exemption has been received under 11.5.11 or a variance has been approved under Section 11.5.10.”).
31 The other options open to the proponent are to seek a minor variance or to seek a Land Use Plan Amendment.
32 NLCA, supra note 2 at art 12.3.1 (“Where the NPC determines, pursuant to Section 11.5.10, that a project proposal is in conformity with the land use plans, or a variance has been approved, the NPC shall, subject to Sections 12.3.2, 12.3.3 and 12.4.3, forward the project proposal with its determination and recommendations to NIRB for screening.”).
2017] NUNAVUT’S ENVIRONMENTAL IMPACTS ASSESSMENT REGIME 209 commonly because a proposal only has minor or minimal impacts) determine that the proposal does not require review/screening and simply forward its recommendation to the Minister for approval.34 Because the vast majority of land in Nunavut is Crown land, the majority of projects take place on Crown land and therefore require a federal land use permit, the “Minister” tends to be the federal Minister of Indigenous and Northern Affairs Canada (INAC). But not all projects do, in fact, take place on Crown land—some take place on IOL, and in those cases, the proponent is required to obtain a land use permit from the RIA. Yet, NIRB still would send its recommendations to the INAC Minister. This raises a related but subsidiary question to the one primarily addressed here; that subsidiary question being, namely how does the INAC Minister attract or gain jurisdiction to receive the recommendation and on what jurisdictional basis decide whether a project should proceed or not.35 As will become clear in this article, much hinges upon the ambiguity of the meaning of “the Minister” and the phrase “the jurisdictional responsibility for authorizing a project to proceed.” These ambiguities, though subtle, portend serious consequences if left unaddressed and it is the purpose of this paper to explain them.
Returning to the initial assessment conducted by the NIRB upon receipt of a 2017 CanLIIDocs 175 proposal from the NPC, the NIRB may, as a second option, send the proposal to a federal environmental assessment panel for a Part 6 review.36 Part 6 reviews are conducted differently than Part 5 reviews and for the sake of efficiency, are not discussed here at this juncture.
Third, the NIRB may determine that a Part 5 review is required, which consists of a formal review by the NIRB itself in which the proponent explains the project in greater detail and various governmental, non-governmental and community groups, including the Inuit of Nunavut, represented either by NTI or an RIA (or both), intervene in the review process and are given the opportunity to make submissions on matters that are of concern to them.37 Part 5 review is the most common type of project review in Nunavut. Upon completion of the review, the NIRB makes a recommendation to “the Minister having the jurisdictional responsibility for authorizing a project to proceed”38 as to whether the project should
33 NLCA, supra note 2 at art 12.4.4. See also Article 12.3.2 excludes certain projects from screening altogether, but if the NPC has concerns about cumulative impacts, it may refer the proposal to NIRB despite 12.3.2.
34 Ibid at art 12.4.4.
35 Any of the ambiguities that may exist with respect to the interpretation of Article 12 provisions, could be easily resolved if both the territorial and federal governments, per Article 12.1., appointed and/or enumerated specific Ministers “to perform all functions assigned to ‘the Minister.’” A similarly easy solution is not, however, available under Article 11.
36 NLCA, supra note 2 at art 12.4.1. See also NUPPAA, supra note 3, s 2, s 161
37 NLCA, supra note 2 at art 12.4.2
38 Ibid at art 12.1.1
210 UNBLJ RD UN-B [VOL/TOME 68 or should not proceed. If the NIRB’s recommendation is that the project should proceed, the NIRB typically includes terms and conditions which are to be added to the project certificate which enables the project to move forward.
Upon receipt of the NIRB’s recommendation following the impacts review assessment conducted by the NIRB in a Part 5 review, there are two options open to “the Minister.” The Minister may accept the NIRB’s recommendation that the project should or should not proceed, whatever the case may be, and such acceptance by the Minister ends the initial environmental impacts review assessment process.39 If the project is recommended to proceed, the proponent is legally obligated to comply with the terms and conditions contained in the project certificate which may and often does include ongoing environmental and wildlife monitoring and impacts mitigation requirements, among other requirements. In contrast, the NLCA prescribes specific reasons under which the Minister may reject the recommendation that the project should proceed or should not proceed.
For example, if the NIRB determines that the project should proceed, the
Minister may reject that determination on the basis that the proposal is not in the 2017 CanLIIDocs 175 national or regional interest.40 It is important to note that the Minister is obligated to consider the project as it relates not only to the regional interest, but also to the national interest. This is interesting for the reason that the NLCA is a treaty which settled the land claims of the Inuit, one of Canada’s Aboriginal Peoples. One must wonder why the Minister would be obligated to consider the national interest, when it may be different from or inconsistent with the regional interest, especially given that the Agreement is meant to settle the claims of and grant rights, specifically, to the Inuit. Moreover, one would also have to wonder what the standard by which to determine what is in the national interest is and how such a standard is in fact determined. Finally, one would also wonder in a territory as vast and as sparsely populated as is Nunavut, what exactly the regional interest is or might be, and whether the regional interest ought to be read as synonymous with Inuit intra- territorial and/or territorial interests.41
The Minister may also reject the NIRB’s recommendation that the project should proceed on the basis that the terms and conditions contained in the recommendation are more onerous than necessary or insufficiently mitigate the ecosystemic and socioeconomic impacts, or, on the basis that the terms and conditions are so onerous that they would undermine the viability of a project that is, again, in the national or regional interest.42 In such situations, NIRB is expected to reconsider its recommended terms and conditions based on the Minister’s reasons for rejecting the recommendation.43
39 Ibid at art 12.5.7(a).
40 Ibid at art 12.5.7(b).
41 These are questions which I undertake elsewhere in another paper.
42 NLCA, supra note 2 at art 12.5.7(c).
43 Ibid at art 12.5.7(c).
2017] NUNAVUT’S ENVIRONMENTAL IMPACTS ASSESSMENT REGIME 211
Similarly, where NIRB recommends that the project should not proceed, the Minister may reject that recommendation on the grounds that the project should have been approved because of its importance to, again, the national or regional interest.44 In such a situation, the Minister will refer the report contained in the recommendation back to NIRB to consider terms and conditions which should be attached to the project’s certificate.45
Finally, where the Minister determines that NIRB’s report is deficient with respect to ecosystemic and socioeconomic issues, the Minister may refer the report back to NIRB for further review or public hearings.46 In such a situation, upon such further review or hearings, NIRB is expected to submit a revised recommendation to the Minister, and the Minister must again accept or reject that recommendation in accordance with the provisions above.47
What is particularly noteworthy about all of these provisions is that the ultimate decision as to whether a project proposed in Nunavut will proceed lies with a federal Minister in Ottawa, and not with those Inuit of Nunavut who are beneficiaries of the Agreement, or with the Government of Nunavut. In other words, 2017 CanLIIDocs 175 by design, the NLCA leaves the final decision about whether a project will proceed or not to Ottawa. In the whole regime, this is not the only power which lies with Ottawa. But we are left wondering, in the absence of a specific Minister appointed to do so, which Minister in Ottawa properly exercises the powers assigned to the “the Minister” in Article 12.48
NUPPAA, or the Nunavut Planning and Project Assessment Act, is, as its preamble states, an act “respecting land use planning and the assessment of ecosystemic and socioeconomic impacts of projects in the Nunavut Settlement Area.”49 NUPPAA purports to fill certain gaps contained in the NLCA in respect of land use planning and development, but in the event of an inconsistency between the NLCA and NUPPAA, the NLCA prevails.50 The NUPPAA came into force in July 2015. Sections 104 to 108 of NUPPAA provide very similarly as do the NLCA
44 Ibid at art 12.5.7(d).
45 Ibid.
46 Ibid at art 12.5.7(e).
47 Ibid.
48 Again, as I have noted elsewhere in this paper, any interpretation problems with respect to “the Minister,” could be resolved by either or both levels of government appointing/enumerating a specific Minister to exercise those powers. The problem of determining jurisdiction would be diminished, but not entirely extinguished.
49 NUPPAA, supra note 3, preamble.
50 NLCA, supra note 2 at art 12.2 (“Where there is any inconsistency or conflict between any federal, territorial and local government laws, and the Agreement, the Agreement shall prevail to the extent of the inconsistency or conflict. Cite to inconsistency provision in NUPPAA.”).
212 UNBLJ RD UN-B [VOL/TOME 68 provisions discussed above, but they are not exactly the same. It is not my intention to comprehensively focus on the similarities or dissimilarities between these two instruments at this instance, and will return to the NIRB process again later. Instead, in the next section I focus on land use planning, land use planning conformity determinations and land use planning exemptions in the NLCA, and note similarities to NUPPAA where relevant to the issues under examination here.
III. LAND USE PLAN CONFORMITY
The NLCA provides that:
…the primary purpose of land use planning in the Nunavut Settlement Area shall be to protect and promote the existing and future well being of those persons ordinarily resident and communities of the Nunavut Settlement Area taking into account the interests of all Canadians; special attention shall be devoted to protecting and promoting the existing and future well-being of Inuit and Inuit Owned Lands…51 2017 CanLIIDocs 175 We should note that this Article does not explicitly refer to land, land use, environmental concerns, wildlife, resource extraction, sustainable development, or anything related to land use planning or development impact; however, it does provide that “special attention shall be devoted to protecting and promoting the existing and future well-being of Inuit and Inuit Owned Lands.” Can the promotion and protection of those areas of concern be inferred from that part of the provision which states its purpose is “…to protect and promote the existing and future well being of those persons ordinarily resident and communities of the Nunavut Settlement Area…or…the future well-being of Inuit and Inuit Owned Lands”? For the time being, we must infer this to be the case based on the way the NLCA is organized. In essence, we must infer that NPC’s goals in making land use planning conformity determinations are, if nothing else, animated by the desire to protect land, wildlife, and the interests of Inuit; however, the existing and future well being of those persons ordinarily resident in Nunavut are also goals of land use planning in Nunavut.52 Where the NPC has determined that a proposal is not in conformity with an applicable land use plan, three options in response to that determination become available to a proponent. The purpose of this part is to explicate these options and bring the interpretive problems into focus.
51 NLCA, supra note 2 at art 11.2.1(b) [emphasis added] (it should be noted that the Nunavut Settlement Area is slightly geographically different than Nunavut, but such a discussion is not needed here; also not just Inuit, but persons, meaning non-beneficiaries as well).
52 The NLCA does not define “person,” nor does it distinguish between persons and aboriginal persons, or Inuit. That being the case, the “existing and future well being of those persons ordinarily resident in Nunavut” should be read, it seems, to include Inuit existing and future well being.
2017] NUNAVUT’S ENVIRONMENTAL IMPACTS ASSESSMENT REGIME 213
A. MINOR VARIANCES
The first of these options is a minor variance. As discussed earlier, upon receipt and review of a proposal, the NPC is required to determine whether a proposal is in conformity with an applicable and existing land use plan and must forward the proposal with its determination and any recommendations to the “appropriate federal and territorial agencies”—namely the NIRB.53 If the NPC determines that a proposal is not in conformity, then provided the applicable existing land use plan authorizes it, the NPC may approve a minor variance to the applicable existing land use plan thereby allowing the proposal to be forwarded to the NIRB.54 Typically, the proponent applies for the minor variance—that is, again, if the land use plan authorizes the NPC to issue minor variances.55
B. AMENDMENTS
Another option available to a proponent and, interestingly, to others such as the
Government, a Designated Inuit Organization (DIO), or any person affected by a 2017 CanLIIDocs 175 plan, is to propose to NPC an amendment to the applicable existing land use plan.56 It is remarkable that the standing requirement to request an amendment is minimal: “any person affected by a plan” may propose such an amendment. Naturally, the question will arise as to who is “any person affected by a plan?” and whether the answer to that question would include a resident of say, Victoria, BC, or Halifax, NS, for example, and whether that answer also includes corporations as persons. It does not seem unreasonable to conclude so if “the interests of all Canadians” are to be “taken into account” in land use planning conformity determinations. While the use of the word “person” in these provisions presents numerous statutory interpretation questions such as these, more detailed answers to those questions and rationales for them must be left for another day.
Nevertheless, NPC must consider the proposed amendment and, if it deems a review appropriate, review the proposal publicly.57 Such a public review gives numerous stakeholders the opportunity to comment upon the amendment application. Upon completion of the proposed amendment review process, the NPC is, according to Article 11.6.3, required to make a recommendation to the Minister of Indigenous Affairs and Northern Development and the Territorial Government Minister responsible for Environment whether (a) the proposed amendment be rejected in
53 Ibid at art 11.5.10.
54 Ibid.
55 It is not clear if the two land use plans currently in effect in Nunavut do provide as such, and this question is the subject of discussion elsewhere in another paper.
56 NLCA, supra note 2 at art 11.6.1.
57 Ibid at art 11.6.2.
214 UNBLJ RD UN-B [VOL/TOME 68 whole or in part; or (b) the proposed amendment be accepted, in whole or in part.58 Article 11.6.3 specifically names these Ministers with respect to land use plan amendments, but can be said to anomalously do so when compared to the provision respecting land use plan exemptions found in Article 11.5.11. These options in 11.6.3, however, present the first illustration of the intersection between politics and law in Nunavut’s impacts assessment regime, for two reasons.
First, land use plans are approved by these Ministers in the first place.59 Whether the will exists to amend a land use plan which a Minister had already earlier approved in order to bring a given proponent’s proposal into land use conformity is distinctly a political, not a legal question—the NLCA evidently permits it.
Second, this political question is further complicated where the Minister of one government possesses such will, and the Minister of the other government does not. Stated more simply, the NLCA provides little guidance on what happens if one Minister wishes to reject the recommendation and one wishes to accept it. The answer to this question is not clear, there is no statutory or policy guidance available, and we must, therefore, it seems, conclude that the signatories of the NLCA 2017 CanLIIDocs 175 intentionally left resolutions of such situations to be sorted out politically.
C. LAND USE PLAN EXEMPTIONS
With the third option, which is the lynchpin of this paper, the problematic intersection between politics and law in Nunavut’s impacts assessment regime is plainly displayed. NLCA Article 11.5.11 provides that where “the NPC has determined that a project proposal is not in conformity with [a land use] plan, the proponent may apply to the appropriate Minister for exemption.”60 Furthermore, upon receipt of such an exemption application, the Minister may exempt the proposal from conformity with the plan and will refer it to NIRB for screening.61 While the rule sounds simple, in application the NLCA provides next to no guidance as to who the “appropriate Minister” is at law. Similarly, there are no policies or policy papers, territorial, federal or otherwise, which provide any guidance on this issue. It is also not clear why or by what rationale Ministers were specifically named in Article 11.6.3, but not so in Article 11.5.11. Where the amendment process requires the amendment application to be submitted to enumerated Ministers of both the federal and territorial governments, there is no such requirement here in the exemption application process. A number of complicated legal questions are presented by this
58 Ibid at art 11.6.3.
59 Ibid at art 11.5.5 (“Upon completion of the process in Section 11.5.4, the NPC shall submit the draft plan as revised along with a written report of the public hearings to the Minister of Indian Affairs and Northern Development and the Territorial Government Minister responsible for Renewable Resources. The NPC shall also make the revised draft land use plan public.”).
60 Ibid at art 11.5.11.
61 Ibid at art 11.5.11 (Additionally, “Nonconforming project proposals shall not be sent to NIRB until such exemption is obtained or a variance has been approved.”).
2017] NUNAVUT’S ENVIRONMENTAL IMPACTS ASSESSMENT REGIME 215 interpretation issue, and in the next part I illustrate with examples the true extent of the problems these unclear answers create.
IV. IDENTIFICATION OF THE PROBLEM(S) BY EXAMPLE
Three different examples illustrate what I have asserted is, first, a complicated intersection of politics and law in Nunavut’s impacts assessment regime and, second, the ways in which—despite participation in the process—the rights of Inuit of Nunavut are fragmented by such a regime. The first example is Baffinland Iron Mines Corp.’s Mary River Iron Ore project; the second is Areva’s Kiggavik Project; and the third is the Clyde River case heard by the Supreme Court of Canada in November 2016, respecting seismic testing off the coast of Baffin Island, Nunavut.
A. BAFFINLAND IRON MINES CORP AND THE MARY RIVER IRON ORE PROJECT
Baffinland Iron Mines Corp. (Baffinland) was granted a project certificate for its 2017 CanLIIDocs 175 Mary River Iron Ore project (Mary River) on December 28, 2012.62 The project is situated on Crown land and mines iron ore from a site at Mary River, which is located on northern Baffin Island, an inland off of Baffin Bay. Baffinland’s initial project consisted of mining iron ore from the reserve at “Deposit No. 1,” at a production rate of eighteen million tonnes per year.63 There are over nine high-grade lump and fine iron ore deposits that can be mined, crushed and screened into marketable products and then shipped through a dedicated port facility at the project site—no processing is required before shipping the iron ore to markets in Europe.64 Baffinland has adopted a phased development strategy for the project, which Baffinland claims lowers the project’s overall environmental impact.65
The first major amendment to the project was the “Early Revenue Phase” (ERP). On April 29, 2014, the Minister of AAND (Aboriginal Affairs and Northern Development, now Indigenous and Northern Affairs Canada (INAC)), approved the positive recommendation made by NIRB following the Article 12, Part 5 process outlined above in Part II. The ERP involves the seasonal shipping of millions of tonnes of iron ore from Milne Inlet, which is a small body of water connected to
62 Nunavut Impact Review Board, Public Registry, “NIRB Project Certificate [No.: 005]”, (Iqaluit: Nunavut Impact Review Board, 2012), online:
63 Baffinland Iron Mines Corporation, “Location and Project History” (March 2017), Baffinland Iron Mines Corporation (blog), online:
64 Ibid.
65 Ibid.
216 UNBLJ RD UN-B [VOL/TOME 68
Eclipse Sound and ultimately Baffin Bay.66 The Minister also approved NPC’s recommendation to amend the North Baffin Regional Land Use Plan (NBRLUP) to accommodate the seasonal shipping.
The second (proposed) amendment to the project was year-round shipping from Milne Inlet, which would have involved ice-breaking the frozen parts of Baffin Bay and the Davis Strait during the winter months.67 The proposal was sent by Baffinland to NPC for land use conformity determination, and the NPC issued a negative conformity determination.68 In light of the negative conformity determination, Baffinland applied for a Ministerial exemption from the NBRLUP; however, that application to the INAC Minister was submitted before NUPPAA was declared in force, and was thus submitted under the NLCA.69 NPC issued a negative conformity determination because the effects of ice-breaking and year-round shipping did not conform to the regional land use plan.70 Stated another way, the NPC issued a negative conformity determination because the effects of year-round shipping would be harmful to wildlife.71 Prior to the exemption application, Baffinland had first sought an amendment to the NBRLUP, but the NPC declined to adjudicate upon the submission on the basis that it was inadequately funded by 2017 CanLIIDocs 175 Ottawa.72 NPC’s refusal to adjudicate upon the submission presents interesting administrative law questions, but is left for discussion elsewhere.73
66 Eclipse sound is connected to Baffin Bay, at the northern end of Baffin Island.
67 I emphasize “proposed” here because final approval of the amendment had not yet been perfected, but as I discuss later, Baffinland withdrew its year-round shipping proposal.
68 Nunavut Planning Commission, “Negative Conformity Determination Recommendation: Baffinland Iron Mines Corporation”, (Iqaluit: Nunavut Planning Commission, 2015), online:
69 It stands to reason that the proposal would have been “grandfathered” into the NLCA process rather than the NUPPAA process given that the project had started well before NUPPAA was declared in force—this, however, is another question to which the answer is provided elsewhere. See Letter from Erik Madsen, Vice President if Baffinland, to Bernard Valcourt (15 May 2015) online:
70 Nunavut, Nunavut Planning Commission, “North Baffin Regional Land Use Plan”, (Iqaluit, Nunavut Planning Commission, 2000) at arts 3.2.1., 3.3.1. online:
71 BIMC Negative Conformity Determination, supra note 68.
72 See Letter from Erik Madsen, Vice President of Baffinland, to Hunter Tootoo (28 April 2015) online:
2017] NUNAVUT’S ENVIRONMENTAL IMPACTS ASSESSMENT REGIME 217
As discussed earlier, NLCA Article 11.5.11 provides that exemption applications are to be submitted to the “appropriate Minister,” and thus the question of why Baffinland submitted the exemption request to the INAC Minister (and not some other Minister) arises.74 In other words, why is or would the INAC Minister be the “appropriate Minister” in respect of this exemption application? The answer to this question, if there is one, involves the statutory interpretation of a constitutionally protected treaty, as it relates to an administrative procedure prescribed within that treaty, which is what I undertake in what follows. I will discuss four possible interpretations below.
First, Article 1 of the NLCA defines Minister as “…a Minister of the Government of Canada or a member of the Executive Council appointed as [a Minister of the Government of Nunavut], as the context requires, responsible for the subject matter referred to…”75 This definition reveals that “the appropriate Minister” under Article 11.5.11 is determined by establishing who, as “the context requires,” the Minister “responsible for the subject matter referred to” is. But, this definition does not tell us what “subject matter” means.76 We are also uncertain which or what 2017 CanLIIDocs 175 At present the Commission is in receipt of other applications and is committed to processing them on a timely basis in the order in which they have been received. Consequently, the Commission is unable to undertake any new processes not planned, budgeted for, and approved by the federal government in the Commission’s previous fiscal year. The Commission would be pleased to work with BIMC if BIMC chooses to proceed with an amendment application. The Commission also would welcome further dialogue on the issues raised in your April 28, 2015 letter. However, please be aware that unless supplementary funding is advanced by Aboriginal Affairs & Northern Development Canada pursuant to our current dialogue with them on that topic, the Commission may be required to prepare a budget and work plan for BIMC’s amendment application and submit them to the federal government for the 2016/2017 fiscal year. See also, “Nunavut premier Peter Taptuna backs Baffinland in regulatory dispute: Taptuna concerned dispute over jurisdiction on shipping question puts jobs, spinoffs at risk”, CBC News (1 June 2015), online:
73 The dispute between Nunavut Planning Commission and The Minister of Indigenous and Northern Affairs resulted in an application for judicial review by NPC, which was later withdrawn by the Nunavut Planning Commission. See Federal Court Registry, Proceedings Queries, (Ottawa, Federal Court Registry, 2015), online:
74 NLCA, supra note 2 at art 11.5.11 (“Where the NPC has determined that a project proposal is not in conformity with the plan, the proponent may apply to the appropriate Minister for exemption. The Minister may exempt the project proposal from conformity with the plan and shall, subject to Sections 12.3.2 and 12.3.3, refer it to NIRB for screening.”).
75 NLCA, supra note 2 at art 1.
76 Cf Wheatland Industrial Park Inc, Re, 2013 BCSC 27: Subject matter jurisdiction refers to the power of a particular court to decide a particular type of case. The Ontario Superior Court, as a court of general jurisdiction, has the prima facie power to decide every type of case, provided the statement of claim discloses a reasonable cause of action. Only by clear and explicit
218 UNBLJ RD UN-B [VOL/TOME 68
“context” the definition is referring to, but it does not seem unreasonable—perhaps even necessary—to assume that “context” relates to “subject matter referred to.” Such a reading only moves us minimally ahead in our understanding of the problem.
Interestingly, Article 12 of the NLCA, devoted to Development Impact, provides its own definition of Minister. It provides as follows: “‘Minister’, unless otherwise specified, means the federal or territorial Minister having the jurisdictional responsibility for authorizing a project to proceed; however, the Government of Canada and Territorial Government may, within their respective jurisdictions, designate a single Minister to be responsible for NIRB and to perform all functions assigned to ‘the Minister’…”77 Neither government has assigned a Minister to perform this function and if they each did, any interpretation questions would ostensibly easily resolved by simply having all Article 12 matters submitted to that so-named Minister. That said, even if such a Minister had been appointed by either government (or was now appointed), it is not clear whether the Article 12 definition would apply in interpreting an Article 11 provision, given that the NLCA provides a global definition in Article 1 which ought to apply to Article 11 interpretation questions. 2017 CanLIIDocs 175
Nevertheless, applying the Article 12 definition, it might be presumed that the INAC Minister had jurisdiction to receive Baffinland’s exemption request because the existing project in respect of which the exemption application was made is situated on Crown land and therefore is a subject matter of federal jurisdiction (i.e. is “jurisdictional responsibility for authorizing a project to proceed”). In contrast, other projects in Nunavut have taken place on territorial Commissioner’s Lands, and therefore by this logic, it would seem that a territorial Minister would have jurisdiction to receive the exemption request if the project were situated on Commissioner’s Lands. As noted earlier, some projects take place on IOL. But it is still not clear how “jurisdictional responsibility” would be determined in such cases. Based on past practices, the most consistently applied and accepted rationale for determining “appropriate Minister” or “jurisdictional responsibility” is land tenure.78 While it is generally the most consistently applied rationale, as I hope to show it ought not to be seen as necessarily absolute, let alone correct.
Returning to NLCA Article 1’s definition of Minister, and reading it in conjunction with provision 11.5.11 to posit a second possible interpretation, the phrase “subject matter referred to” (in the Article 1 definition) can be interpreted to refer to the subject matter of the proposal—that is, what is this project proposal about, what is its subject matter? In the Baffinland example, the larger project is about mining iron ore at Mary River, but the specific part of the proposal which was determined by NPC not to conform to the applicable land use plan, had as its subject
limitation may the power of the Superior Court to decide a particular type of case be curtailed. But we are not dealing here with the question of a court’s competence based on subject matter jurisdiction.
77 NLCA, supra note 2 at art 12.
78 See e.g. Territorial Lands Act, RSC, 1985, c T–7 and Territorial Lands Regulations, CRC, c 1525.
2017] NUNAVUT’S ENVIRONMENTAL IMPACTS ASSESSMENT REGIME 219 matter, year-round shipping of iron more. That then would appear to make the impugned part of the proposal about shipping and navigation, which is, exclusively a federal matter, with jurisdictional responsibility falling to the federal Transportation Minister, not the INAC Minister.79 On this logic, it would seem that the federal Transportation Minister should receive the exemption application instead of the INAC Minister. A reviewing court, however, would likely give deference to the federal government and find that, from a federal perspective, several ministers, jointly and/or severally, constitute the “appropriate Minister” making the decision respecting the exemption application and thus no legal error would be extant despite submission of the exemption application to the INAC Minister.80 Even so, such a conclusion is nevertheless insufficient to justify, legally, why jurisdiction lies with the INAC Minister in this case.
As a third possible interpretation, if we note that the proponent, Baffinland, applied to the federal INAC Minister (and not to the territorial or another federal Minister), another way then, perhaps, to determine proper ministerial jurisdiction might be to look at the grounds upon which the proponent seeks the exemption, and deciding based on those specified grounds which Minister—federal or territorial— 2017 CanLIIDocs 175 has “jurisdictional responsibility” based on the “subject matter referred to…”. As we have seen, Article 11.5.11 states: “where the NPC has determined that a project proposal is not in conformity with the plan, the proponent may apply to the appropriate Minister for exemption.”81 If a territorial or federal Minister lacked the jurisdiction to decide the exemption application, it would be incumbent upon that Minister to decline adjudicating upon it. It is not clear if it would be incumbent upon the Minister rejecting the application on the basis of an absence of jurisdiction to forward the request to another Minister, but the Minister to whom the application is referred would then be forced to answer the question of who the “appropriate Minister” is. It might, however, also make sense, to some extent at least, in an example where a project that has already passed part 5 screening by NIRB, such as this example, that the same Minister which approved the NIRB’s recommendation that an initial project certificate should issue, should also be the Minister which allows or denies exemption applications made by proponents in subsequent project proposals respecting that project.82 In the Baffinland example, that then would be the INAC Minister. But the question of whether the Minister had the jurisdiction in the
79 Constitution Act, 1867 (UK), 30 & 31 Vict, c 3 at s 91(10), reprinted in RSC 1985, Appendix II, No 5.
80 See James WJ Bowden & Nicholas A. MacDonald, “Writing the Unwritten: The Officialization of Constitutional Convention in Canada, the United Kingdom, New Zealand and Australia” (2012) 6 J Parliamentary & Pol L 365. See also Mark Schacter & Philip Haid, “Cabinet Decision-Making in Canada: Lessons and Practices” (April 1999) Institute on Governance, online:
81 NLCA, supra note 2 at art 11.5.11 [emphasis added].
82 Ibid at art 11.5.8: “Upon accepting a plan, the Minister of Indian Affairs and Northern Development shall seek Cabinet approval and commitment, and the Territorial Government Minister responsible for Renewable Resources shall seek approval and commitment of the Executive Council.”
220 UNBLJ RD UN-B [VOL/TOME 68 first place still lingers because NIRB is faced with a similar problem: did it make its recommendation to the correct (or “appropriate”) Minister? In other words, did NIRB furnish its report to the INAC Minister, and not some other Minister, solely because the Mary River project takes place on Crown land, and if so, was that correct at law?
In this example, and proffering a fourth possible interpretation, the question also arises as to whether the Government of Nunavut could assert responsibility or jurisdiction as the “appropriate Minister” based on wildlife management, given that NPC found that the proposal did not conform to the applicable land use plan because of the deleterious effects and disturbance year-round shipping would have on and to wildlife. Yet, grounding an argument on the basis of wildlife would have to be based on jurisdiction over wildlife (which the Government of Nunavut has), and presumably not on the effects the proposal would have on wildlife. In other words, wildlife is not the subject matter of the proposal and so it seems that such an assertion by the Government of Nunavut could not be properly—or at least could only be tenuously—anchored at law.
2017 CanLIIDocs 175 As noted earlier, NUPPAA was designed and enacted to fill any purported gaps in the impact assessment regime created by the NLCA, but it too is not helpful in resolving the question of who the “appropriate Minister” is.
Returning, nonetheless, to NUPPAA which, again, was not in force at the time the exemption application was submitted to the INAC Minister, subsection 82(1) provides that if NPC “…determines that the project is not in conformity with an applicable land use plan, the proponent may request an exemption from the federal Minister or the territorial Minister, or both, taking into account their respective jurisdictions, within 60 days after: (a) that determination, if the land use plan does not authorize the granting of a minor variance or if it does and the conditions are not met; or (b) the Commission’s decision to refuse to grant a minor variance.”83
What is similarly problematic about subsection 82(1) is that it is equally unclear which Minister (federal or territorial), to the exclusion of the other, has the “respective jurisdiction” to receive the exemption request and upon what basis such jurisdiction is determined. As the four possible interpretations I provided above with respect to interpreting the NLCA provision showed, the ambiguity lies in what “respective jurisdictions” refers to, i.e., respective jurisdiction to what? The land tenure of the project? The project itself? The grounds or subject matter of the exemption application? Furthermore, it is not clear who is to take respective jurisdictions into account—the proponent? The Minister who receives the exemption request application, irrespective of whether that Minister enjoys jurisdiction or not? Given the lack of clarity as to who the “appropriate Minister” is, how would the proponent—Baffinland in this example—determine jurisdiction? By looking at which government holds land tenure? By the proposal’s subject matter? Which subject matter—the project as it already exists (as in this example), the subject
83 NUPPAA, supra note 3 at s 82(1) [emphasis added].
2017] NUNAVUT’S ENVIRONMENTAL IMPACTS ASSESSMENT REGIME 221 matter of the project amendment proposal, or the subject matter of the exemption application? Many of these questions have no answers, let alone clear ones.
Yet another problem exists within subsection 82(1). It provides three options to a proponent: apply to the federal minister, the territorial minister, or both—the NLCA does not provide this third option, and whether it provides the first two, at least in the same manner, is a matter of interpretation. If Baffinland had submitted the exemption request to both Ministers, it is not clear in subsection 82(1) what happens if there is disagreement between them as to whether the exemption should issue.84 Furthermore, no path to resolution in such a situation is provided in NUPPA. Again, it seems that the resolution to such a situation is left to be crafted in the political arena. A proponent’s safest bet, from a practical and legal point of view, seems to be to submit the exemption application to both Ministers, and to leave jurisdictional questions and whether they are ad idem to be sorted out between them. Any unfavourable decision by either Minister (or both) might then be amenable to judicial review under Article 12.10.5.85
Finally, the federal Interpretation Act is of little assistance here, as it does 2017 CanLIIDocs 175 not define “appropriate” or “Minister.”86 Moreover, the NLCA provides that there “… shall not be any presumption that doubtful expressions in the Agreement be resolved in favour of Government or Inuit.”87 We have seen that “Government” is defined as “…the Government of Canada or the Territorial Government or both, as the context requires, depending on their jurisdiction and the subject matter referred to…” That being the case, the Article 2.9.3 presumption is mutually exclusive to government—because it is not clear which level of government has jurisdiction in this case—leaving the only conclusion to be that doubtful expressions are not to be resolved in favour of Inuit. Article 1.1.6, however, enables both levels of government to designate a single Minister from either level of government to act on behalf of one government or both; however, neither government has taken this step, thus still leaving the mess of interpreting who the “appropriate Minister” is untangled.88
84 It is doubtful that the doctrine of constitutional paramountcy is of much help here given that NUPPAA is a federal statute and it is not in conflict with any provincial or territorial law. As noted elsewhere in this article, where NUPPAA conflicts with the NLCA, NUPPA yields to the NLCA.
85 NLCA, supra note 2 at art 12.10.5 (“…any person or body that is recognized by laws of general application as having standing to seek a court determination…shall have standing before an appropriate court…to seek judicial review of decisions and orders, whether interim or final, made pursuant to this Article.”).
86 Interpretation Act, RSC, 1985, c I–21.
87 NLCA, supra note 2 at art 2.9.3.
88 Ibid at art 1.1.6: Without diminishing or otherwise altering the responsibilities of Her Majesty The Queen in Right of Canada under the Agreement, where, in the Agreement, it is unclear from the context which Government is to perform a function or where the context indicates that both Governments are to perform a function, without abrogating or derogating from their obligations under the Agreement or altering
222 UNBLJ RD UN-B [VOL/TOME 68
Unfortunately, there are no clear answers to questions of which Minister has jurisdiction to issue an exemption application under the NLCA (or NUPPA). NIRB is in the essentially the same position when making recommendations to a/the Minister because it relies on the definition of Minister provided in Article 12 which, as I have shown, also has interpretation difficulties. Altogether, the absence of clarity in the NLCA and NUPPAA may be a reflection of s.91 and s.92 of the 1867 Act in the sense that there is no head of power clearly demarcating jurisdiction over the environment in Canada’s Constitution. It is doubtful that this ambiguity is a conscious design element of the Agreement, and perhaps is more likely a product of improvident drafting than anything else at the same time, however, it could be a conscious design element to the extent that the governments were at an impasse and could not resolve the issue at the time or because they wanted to provide maximum flexibility in the future. Nevertheless, until a court of law answers these questions, there will be legal uncertainty as to whether some of the decisions that are being rendered in Nunavut’s environmental impacts assessment regime, especially those pertaining to exemption applications, have been made in accordance with law. So far, no one has emerged to launch a judicial review application of these decisions.
However as NPC approaches completion of the Nunavut-wide land use plan, 2017 CanLIIDocs 175 development in Nunavut increases and more conflicts between economic development and wildlife protection emerge, it seems quite possible that a party will.89
On that note, a final illustration in this example of the intersection of politics and law in Nunavut’s impacts assessment regime is a letter sent by the Premier of Nunavut, Peter Taptuna, to the then INAC Minister urging his support for Baffinland’s exemption application.90 In a news article published by CBC News, Premier Taptuna was quoted as saying: “I gotta ensure the best interests of Nunavummiut…I am the Premier, indicating that our mandate is economic development and employment.”91 It is not clear whether the Executive Council of the Government of Nunavut countenanced the sending of such a letter, but it is clear in any event that there is no procedural basis for such a letter found in the NLCA, NUPPA, or the Constitution. Incidentally, when Baffinland’s amendment application was refused, prior to the submission of its exemption application, the Premier also then sent a letter to the INAC Minister urging his general support for the project.
their respective jurisdictions, the two Governments may designate one of them to perform that function on behalf of the other or both. The DIO shall be given notice of such designation.
89 It is important to note that the exemption process simply enables a project that would otherwise not reach the NIRB to reach it for either a Part 4 review and Part 5 or Part 6 screening.
90 “Nunavut premier stands by stance on Baffinland”, CBC News (2 June 2015), online:
91 Ibid. See also “Nunavut premier’s leaked letter to Ottawa disappoints board chair: Premier Peter Taptuna defends position as job protection strategy”, Nunatsiaq News (1 June, 2015), online:
2017] NUNAVUT’S ENVIRONMENTAL IMPACTS ASSESSMENT REGIME 223
The Qikiqtani Inuit Association (QIA), the birthright organization (and RIA) which represents the rights and interests of Nunavut Inuit in the Qikiqtani region of Nunavut, in which Baffin Island sits, and Nunavut Tunngavik Incorporated (NTI), the birthright organization which represents the interests of Nunavut Inuit intra-territorially, nationally, and internationally, were both opposed to the granting of an exemption request.92 In the absence of any legal or procedural basis for the Premier’s letter, the only conclusion to be drawn is that the Premier was simply engaging politics. A corollary question arises as to what the best interests of Nunavummiut are in this case, and whether they accord with his assertion that the Government of Nunavut’s mandate is exclusively “economic development and employment.” It is also necessary to wonder if the Premier meant—either intentionally or unintentially—to ignore protection of wildlife, ecosystems, and wildlife habitat, which is a part of the Government of Nunavut’s mandate as well.93 As any government official in the Premier’s position would know, a determination as to which controls or supersedes the other is very difficult, if not impossible, to make.
One might even be so bold as to say that the Premier’s actions are a rendering of regulatory capture in the landscape of Canadian environmental 2017 CanLIIDocs 175 regulation.94 The fact that the Premier of Nunavut would urge Ottawa to override a decision made by an administrative body established by the very same treaty which acts as his territory’s constitution, is indeed curious, even unusual. Given Ottawa, not Nunavut Inuit, makes final decisions which affect the future of wildlife and economic development in the territory, I argue, fragments the very rights that Inuit fought so long and hard to obtain. While we can reasonably presume that at the time the treaty was signed, Canada would not relinquish complete control of Crown land to the Inuit of Nunavut via the NLCA. The fact that the environmental impacts assessment regime culminates or crystalizes with final decision-making power vested in Ottawa shows that the NLCA in itself provides a marginal legal basis for Inuit to
92 “QIA, NTI team up against Baffinland's request for NPC exemption: Qikiqtani Inuit Association, Nunavut Tunngavik Inc. to oppose granting an exemption from NPC”, CBC News (9 June 2015), online:
93 NLCA, supra note 2 at art 11.2.1 (“…people are a functional part of a dynamic biophysical environment, and land use cannot be planned and managed without reference to the human community; accordingly, social, cultural and economic endeavours of the human community must be central to land use planning and implementation…”); see also the self-delineated mandate of the Government of Nunavut: Government of Nunavut, Sivumut Abluqta: Stepping Forward Together (Iqaluit: Government of Nunavut, 2014), online:
94 Jason MacLean, “Striking at the Root Problem of Canadian Environmental Law: Identifying and Escaping Regulatory Capture” (2016) 29 J Envtl L & Prac 111.
224 UNBLJ RD UN-B [VOL/TOME 68 determine their best interests. After all, the preamble to the NLCA merely refers to “participation” in the process and that, at least, is provided. But the apparent absence of any final procedural decision-making powers for Nunavut Inuit in the environmental impacts assessment regime is ultimately the genesis of any erosion and fragmenting of those very rights, because where Inuit are not ad idem with Ottawa, Ottawa makes the final decisions, not Inuit.
Not much came of the apparent illegitimacy of Premier Taptuna’s letters by way of public outrage, parliamentary censure or political injury, and in the end, the INAC Minister granted and issued the land use plan exemption to Baffinland.95 Baffinland was required to submit a revised impact assessment statement to NIRB in September 2016; however, on November 30, 2016 (coincidentally the same day that the Supreme Court of Canada heard the Clyde River appeal discussed below), Baffinland informed the NIRB that it intended to abandon its year-round shipping proposal.96 In a surprising twist of events, on December 19, 2016, NIRB nevertheless sent the remainder of the proposal to NPC for a new conformity screening, thus beginning the whole impacts assessment process anew.97 In other words, the
Baffinland proposal is back at square-one, seeking a conformity determination from 2017 CanLIIDocs 175 NPC.98
Ultimately, through the issuance of the exemption, and Baffinland’s withdrawal of that portion of the proposal proposing year-round shipping, the tough jurisdictional questions I have posed in this paper, in essence, would merely have been deferred to another day. That day would have been when NIRB completed its review and assessment of the proposed amendment impacts. If Baffinland had not withdrawn its proposal, and NIRB had recommended the proposal be accepted and the amended project proceed, the INAC Minister still would have had to decide
95 Letter from Bernard Valcourt to Elizabeth Copland, Chair of NIRB, Hunter Tootoo, Chair of NPC, and Erik Madsen, Vice President of BIMC (13 July 2015) online:
96 Baffinland Iron Mines Corporation, “Mary River Project, Project Update, Final Report” (30, November 2016) Nunavut Impact Review Board, online:
97 Letter from Elizabeth Copland, Chair of NIRB, to Todd Burlington, Vice President of Sustainable Development at BIMC (19 December 2016) online:
98 Ibid. The NIRB did, however, inform Baffinland that “…the NIRB emphasizes that it is committed to ensuring that all information received during the Board’s consideration of the original Phase 2 Proposal to date will be brought forward into any future assessment of the modified Phase 2 Proposal.”
2017] NUNAVUT’S ENVIRONMENTAL IMPACTS ASSESSMENT REGIME 225 whether to accept that recommendation, despite the opposition from QIA and NTI respecting year-round shipping, although—much like the proposal itself—now presumably having likely fallen away.99 This, again, however, is a political question, not a legal one. If NIRB recommended that the proposal be rejected and the amended project not proceed, the Minister would still have had to decide whether to accept that recommendation despite the Premier’s support (and ostensibly Nunavut since he arguably speaks for all of its constituents) for the exemption and the project writ- large. Indeed, it will be interesting to see what result the NPC produces after a review of the new proposal is completed, i.e. whether a positive or negative conformity determination is issued, and what result the NIRB produces if the proposal is forwarded to it by NIRB. Following that, it will be even more interesting how the INAC Minister will respond to either another exemption application should it be sought, a positive or negative recommendation from NIRB if a positive conformity decision is issued, whether the Premier of Nunavut again attempts to exert pressure on Ottawa in respect of making that decision, and whether Baffinland seeks judicial review of any of these decisions or simply decides to wind-up the project altogether.
2017 CanLIIDocs 175
B. AREVA AND THE KIGGAVIK PROJECT
Another illustrative example of the problematic relationship among politics, administrative and constitutional law in Nunavut’s environmental impacts assessment regime is found in another decision rendered by a federal Minister in respect of a project undertaken in Nunavut by the Areva Resources Canada Inc. (Areva), a company that produces nuclear energy. Areva’s proposed project, the Kiggavik Project, consisted of a planned uranium mining and milling operation located approximately 80 km west of Baker Lake, in the Kivalliq region of Nunavut.100 It was a contentious project, which bitterly divided the community. Some community members were in support of the project, because of the economic opportunities it would provide to them, while others were bitterly opposed to it because of the project’s impact on wildlife and the perceived unsafety of mining uranium, despite assurances from the Canadian Nuclear Safety Commission that the project posed little to no radiation risks and no uranium would be enriched on-site.101
99 Following Baffinland’s revision, tenuous support for the proposal emerged from the community closest to the project. See “Nunavut hamlet backs Baffinland’s new shipping plans: But Pond Inlet wants to see the review move forward without delay” Nunatsiaq News (16 December 2016) online: 100 Areva Resources, “Kiggavik Project” (10 March 2017), Areva Resources (blog), online: 101 Many intervenors were opposed to the project; see e.g. Baker Lake Hunters and Trappers Organization, “Baker Lake Hunter and Trappers’ Organization Motion to suspend the final hearing” (16 January 2015), Nunavut Impact Review Board, online: 226 UNBLJ RD UN-B [VOL/TOME 68 Although no specific support for this project was issued by the Government of Nunavut, the government did have a policy of supporting uranium mining in Nunavut.102 The proposal was submitted to NPC by Areva and underwent conformity review in accordance with the NLCA, as the project had also commenced before NUPPAA was declared in force. Following NPC’s issuance of a positive conformity determination, the proposal was sent to NIRB for review and screening. A Part 5 public hearing was held in Baker Lake, and in May 2016, the NIRB recommended to the INAC Minister that the project should not proceed. Over a year later, following a federal election in the fall of 2015, which saw a new government elected, the newly installed INAC Minister accepted NIRB’s recommendation, asserting, without any justification, her “jurisdictional authority for authorizing the project to proceed.”103 In this example, the Premier of Nunavut did not write any letter of support or opposition; however, in the year between NIRB furnishing its recommendation to the Minister and the Minister making the decision Areva wrote to the INAC Minister (without any procedural basis) urging her to reject the NIRB’s recommendation on the basis that the project should have been approved because of its importance in the 2017 CanLIIDocs 175 regional interest.104 Despite the many contentious issues respecting the project, NIRB thinly veiled the substantive concerns the community and many interveners expressed in the final hearing in the cloak of a concern respecting the lack of a firm starting date (blog), online: 102 Nunavut, Legislative Assembly, Hansard, 3rd Leg, 3rd Sess (May 31, 2012) at question 352-3(3): Uranium Mining Issues in Nunavut (Aupaluktuq); see also Government of Nunavut, “Government of Nunavut Uranium Mining Policy Statement” (10 March 2017) online: 103 Letter from Carolyn Bennet, Indigenous and North Affairs Canada Minister, Elizabeth Copland, Chair of Nunavut Impact Review Board (14 July 2016) online: 104 Letter from Vincent Martin, Present of Areva Resources Canada, to Bernard Valcourt, INAC Minister, (3 July 2015) online: 2017] NUNAVUT’S ENVIRONMENTAL IMPACTS ASSESSMENT REGIME 227 for the project. Areva objected to this concern, citing another project which had been approved by NIRB despite the lack of a firm start date.105 Had Areva even arbitrarily identified a start date for the project, NIRB and the Minister would have had to meet the substantive issues head-on instead of escaping them on the grounds—arguably a pretext—that an indefinite start date made it impossible to assess the impacts the project would have. In contrast, had NIRB recommended in its report that the project should have proceeded, Ottawa would have been faced with making sensitive political decision: approving a project which many Nunavummiut were bitterly opposed to, and justifying it on the basis that it was in the national or regional interest to approve it. While it is not necessarily unusual or problematic that Ottawa might have a final say in a project of this nature, it is conceivably and politically problematic that Ottawa could, while legal, override a decision made by a constitutionally created administrative tribunal which has as its aim the protection and promotion of the well-being of the environment and Nunavummiut through the environmental impacts assessment regime.106 Given Canada’s history of dispossessing Aboriginal Peoples from their lands, the potential of such an occurrence is unsettling. 2017 CanLIIDocs 175 C. CLYDE RIVER AND SEISMIC TESTING The final example is the appeal heard by Supreme Court of Canada in November 2016: Clyde River, 2015 FCA 179. This case does not arise out of the NLCA impacts assessment regime, and instead arises from a decision of the National Energy Board (NEB), but it is nevertheless illustrative of the complicated relationship of politics, administrative and constitutional law in Nunavut as it relates to Inuit interests in and participation in decision-making respecting the offshore. In May, 2011, the proponents applied to the NEB for a Geophysical Operations Authorization to undertake an offshore seismic survey program off of the coast of Clyde River, Nunavut, also located on North Baffin Island. The program would involve detonating air guns exponentially louder than a jet engine, every 13 to 15 seconds, for 24 hours a day, five months per year, for a period of five years.107 The residents of Clyde River, the Applicants for judicial review of the NEB decision, were opposed to the project prior to the proponents’ application to the NEB, and generally remained so during the NEB’s hearing and consultative process regarding the application because of the potential impact the project could have on marine mammals in the project area and the Applicants’ opportunities to harvest 105 Letter from J Duncan to Elizabeth Copland (29 May 2012), regarding Hope Bay Gold Project. 106 NLCA, supra note 2 at art 11.2.1(b). 107 Clyde River, supra note 5 (Factum of the Appellant at para 14) 228 UNBLJ RD UN-B [VOL/TOME 68 them. The Applicants were also concerned about the effects the disturbance on marine mammals would have to their traditional way of life. On June 26, 2014, the NEB issued to the project proponents authorization to conduct a seismic survey program. On July 28, 2014, the Applicants filed an application for judicial review in the Federal Court of Appeal (FCA) which has exclusive jurisdiction over the NEB. The application was heard by the FCA on April 20, 2015 and judgment dismissing the application was issued on August 17, 2015. On October 16, 2015, the Applicants filed an application for leave to appeal to the SCC. On March 10, 2016, the SCC granted leave to appeal. The Court also ordered that the appeal would be heard in conjunction with Chippewas of the Thames First Nation v. Enbridge Pipelines Inc., et al., another similar case.108 The Appeal was heard on November 30, 2016 and judgement was reserved. The Applicants argued in their leave to appeal factum that: the duty to consult with them was triggered by the NEB’s receipt of the proponents’ GOA application; given that they possessed treaty rights to harvest marine mammals in the project area and the potential impact on those rights, the duty to consult lied at the 2017 CanLIIDocs 175 high end of the consultative spectrum and therefore meaningful attempts to engage them in the decision-making process were required and if necessary, so too was accommodation; and, the Crown had done “virtually nothing” to discharge this duty.109 This case is important in the context of this paper and particularly to the Baffinland example, because it illustrates, at the intersection of politics and law, in making a decision to accept or reject a recommendation of the NIRB or the NPC, or to issue an exemption application request, the complex and necessary legal question—which the Supreme Court will answer—of whether it is incumbent upon the adjudicating Minister to fulfill the constitutional duty of consultation with Inuit affected by the decision being rendered by that Minister. If the Baffinland project amendment-cum-exemption application were still capable of being accepted by Ottawa, and thus year-round shipping through Baffin Bay and the Davis Strait would have resulted, and if the SCC ultimately finds that the NEB did not fail in its duty to consult to Inuit, and the proponents are free to proceed with their seismic survey testing in Baffin Bay and the Davis Strait, the combined or cumulative effects of these projects will be devastating to wildlife and the Inuit who rely upon for cultural and subsistence purposes. Altogether, these three examples illustrate that decision-making respecting resource extraction in Nunavut is anything but simple, and instead politically and legally complex. Moreover, we see how Inuit rights are fragmented by these complex legal proceedings and processes. 108 Chippewas of the Thames First Nation v Enbridge Pipelines Inc, NEB, 2015 FCA 222, [2016] 3 FCR 96. 109 Clyde River, supra note 5 (leave to appeal Factum of the Appellant at para 36–51). 2017] NUNAVUT’S ENVIRONMENTAL IMPACTS ASSESSMENT REGIME 229 V. WHY THESE ISSUES ARE IMPORTANT We need only look at the primary purpose of land use planning again to understand why these issues are important. As we saw, the NLCA provides that: …the primary purpose of land use planning in the Nunavut Settlement Area shall be to protect and promote the existing and future well being of those persons ordinarily resident and communities of the Nunavut Settlement Area taking into account the interests of all Canadians; special attention shall be devoted to protecting and promoting the existing and future well-being of Inuit and Inuit Owned Lands…110 We can also look to Article 12 again, which states that the NIRB’s mandate “shall be at all times to protect and promote the existing and future well-being of the residents and communities of the Nunavut Settlement Area, and to protect the ecosystemic integrity of the Nunavut Settlement Area.”111 2017 CanLIIDocs 175 If a federal Minister continues to issue land use plan exemptions in respect of projects which do not conform to an applicable land use plan because of the deleterious effects it would have, and in respect of proposals which the Inuit of Nunavut strongly object to, the impact on Inuit rights and interests contained in the NLCA such issuance would have is unclear. If my thesis is correct, the exercise and enjoyment of those rights, if nothing else, will be diluted. Similarly, if a federal Minister ultimately decides which projects go forward under the NIRB process, and which do not, any resulting impact on the rights and interests Inuit have in the NLCA will also probably be diluted. Although the final decision is by design meant to rest with Ottawa, a decision that is incongruous with Inuit desires and interests has the real possibility of thwarting the promotion and protection of rights which the NPC and the NIRB aim to ensure. As noted in the foregoing analysis, the NLCA only contemplates “participation” by Inuit in the regime discussed in this article, not “decision-making.” To the extent that Inuit are provided an opportunity to “participate” in the regimes’ administrative processes, such as a Part 5 review, perhaps it can be said that the spirit and intent of the Agreement is being lived up to. But with close to twenty years having lapsed since the Agreement was signed, we can see how much the Agreement’s design is skewed in the federal government’s favour. Such a result is troubling because Nunavut’s future is at stake, the future of Inuit in Nunavut is at stake, land and resources are at stake as is land and resource development, wildlife and marine mammals, and Nunavut’s society, economy and posterity. The Agreement was supposed to place in the hands of the Inuit some, if not entire, control over their destiny. 110 NLCA, supra note 2, at art 11.2.1(b). 111 Ibid at art 12.2.5. 230 UNBLJ RD UN-B [VOL/TOME 68 As experiences in the Baffinland and other projects (not discussed here) illustrate, the tendency for proponents to propose one project and slowly, over time, erode the protective mechanisms contained in a project certificate through numerous minor variances, amendments, and applications for exemptions seems to be an emerging trend. Being such a young jurisdiction, it is difficult to test what the cumulative effect of such practices across varying projects would be in Nunavut. Part of the problem with respect to predicting the effects such decisions might have, apart from the inherent impossibility of accurately predicting the future, is embodied in the nature of defining the problem itself. While I have illustrated the problems associated with interpreting these provisions, the consequences have yet to manifest in legal challenges. Ultimately, in Nunavut’s environmental impacts assessment regime the question of whether politics, and not law, are deciding the fate of Inuit and Nunavut arises. So too does the question of what the impact on Inuit society, culture and way of life is once projects which are approved in the name of economic development go forward at the expense of wildlife and environmental protection. One might also fear that Inuit Qaujimajatuqangit (traditional knowledge) is also slowly eroded by the 2017 CanLIIDocs 175 regime and the increasing prevalence in Nunavut of the cash economy. Related to that are other important social and legal concerns that exacerbate the issues discussed here, such as (in particular order): climate change, eroding seasons, poverty, homelessness, education, unemployment, crime (murder, sexual assault), suicide, alcohol and substance abuse, and re-colonization, which are all intricately interwoven in Nunavut, and simultaneously too complex to discuss here. CONCLUSION This article has raised many more questions than it has answered—but many of these questions can only be answered by the courts. As such, they remain emerging issues with little territorial jurisprudence to guide lawyers and the courts. While Nunavut’s environmental impacts assessment regime is governed by administrative procedures established in a constitutionally protected treaty, there are competing interests, and the issues, it seems, are resolved politically as much as they are legally, if not more so. My analysis is important because NUPPAA is what now governs project planning at the statutory level in Nunavut, and the statute’s durability will be put to the test by future projects; however, we are reminded that, in any event, inconsistencies between NUPPPAA and the NLCA are resolved in favour of the NLCA, leaving the questions I have raised here to still be answered. Even so, the absence of clarity as to which Minister—federal or territorial—and by what jurisdiction or legal authority he or she authorizes a project to proceed or issues a land use exemption remains highly problematic. The problems associated with establishing or determining Ministerial jurisdiction, or who the “appropriate Minister” is in land use exemption applications, is only further complicated when politics joins the fray, sometimes bordering on or even fully venturing into regulatory capture, serving mostly, if not exclusively, to fragment the rights and interests of Nunavut Inuit under the Agreement when final-decision making power is vested in Ottawa. The problems remain, and it seems, for the time being at least, will 2017] NUNAVUT’S ENVIRONMENTAL IMPACTS ASSESSMENT REGIME 231 only be resolved where politics and administrative and constitutional law intersect in Nunavut’s environmental impacts assessment regime. 2017 CanLIIDocs 175 PERFECTLY LEGAL, BUT STILL BAD: LESSONS FOR SEX WORK FROM THE DECRIMINALIZATION OF ABORTION Jula Hughes Abstract Sex workers’ rights advocates and prostitution abolitionists have argued for a causal link between decriminalization and destigmatization while criminal law theory and jurisprudence similarly suggest a close link between criminalization and stigma. Based on social stigma theory and an observational study of abortion laws in the Maritime provinces, this paper argues that this link is overdrawn. Following the complete decriminalization of abortion in 1988, destigmatization did not follow decriminalization. Instead, abortion stigma continued as all three Maritime provinces adopted restrictive regulatory frameworks. Social stigma theory supports the idea 2017 CanLIIDocs 175 that criminal law plays a labelling function for stigma, but contends that rather than creating stigma, it tends to be responsive to pre-existing stereotypes. Experience with abortion law reform supports the view that public stigma is not very sensitive to changes in structural stigma such as criminal legislation and that the interaction between criminal law and social stigma is complex. Introduction 1. Social Stigma 235 2. Criminal Law and Stigma 237 3. Abortion Stigma 244 4. Implications for Sex Work 250 Introduction Following the Supreme Court of Canada decision in Bedford,1 advocates of both abolition and decriminalization anticipated a legislative outcome that would lessen the stigma of sex work. Abolitionists feared that destigmatization, often referred to as ‘normalization’, would endanger the project of women’s equality and render women’s victimization in the context of prostitution invisible.2 Decriminalization Associate Professor of Law, University of New Brunswick. Thank you to Leah Ferguson and Ashley Godfrey for unfailingly thorough and helpful research assistance, the organizers of the conference Controlling Sexuality and Reproduction, Past and Present, at the University of Lethbridge, Alberta for providing the opportunity to present an earlier version of this paper, to colleagues at the conference including Dr. Julie Kaye for their feedback, and to the Faculty of Law at the University of New Brunswick for a grant to attend the conference. 1 Canada (AG) v Bedford, 2013 SCC 72, [2013] 3 SCR 1101. 2017] PERFECTLY LEGAL, BUT STILL BAD 233 advocates predicted that destigmatization would lead to improved working conditions and occupational health protections for sex workers.3 Both sides of the debate have tended to assume that decriminalization would necessarily be attended by destigmatization.4 This view is contested in the social science literature5 but is consistent with the criminal law jurisprudence of the Supreme Court of Canada.6 The first part of this paper summarizes the social science literature on stigma production. Social science theories of stigma generally accept that stigma is produced through the interactions of the identification of difference, stereotyping, labelling and discrimination. The application of stigma tends to be spontaneous and attaches to diffuse targets. It is not typically susceptible to reflection or direct control. The second part considers the doctrinal treatment of stigma by the Supreme Court. The Court has suggested that there is a close relationship between criminal law and stigma.7 Criminal law is understood as stigmatizing and the production of stigma is controlled by the criminal justice system. This close relationship helps to delineate 2017 CanLIIDocs 175 2 Suzanne Jay, Submission to the Standing Senate Committee on Legal and Constitutional Affairs regarding Bill C-36 on behalf of the Asian Women Coalition (1 September 2014), online: 3 Cactus Montreal, Brief to the Standing Senate Committee on Legal and Constitutional Affairs regarding its study of Bill C-36, the Protection of Communities and Exploited Persons Act (5 September 2014), online: 4 Parliament received some documentation to the contrary. See Department of Justice Canada, “Technical Paper: Bill C-36, An Act to amend the Criminal Code in response to the Supreme Court of Canada decision in Attorney General of Canada v. Bedford and to make consequential amendments to other Acts (Protection of Communities and Exploited Persons Act) at note 40. 5 Bos et al argue that public stigma lies at the root of all forms of stigma including self-stigma, stigma by association and structural stigma and urge more research be conducted to evaluate whether efforts to decrease structural stigma, including changing discriminatory laws, operate to lower public stigma. This chapter is a contribution to this line of inquiry. Arjan ER Bos et al, “Stigma: Advances in Theory and Research” (2013) 35:1 Basic & Applied Social Psychology 1 at 6. 6 The jurisprudence on this point is discussed in Part 2 of this paper. Important cases include R v Pierce Fisheries Ltd (1970), [1971] SCR 5, 3 NSR (2d) 1 [Pierce Fisheries]; R v Sault Ste Marie, [1978] 2 SCR 1299, 85 DLR (3d) 161 [Sault Ste Marie]; R v Vaillancourt, [1987] 2 SCR 636, 68 Nfld & PEIR 281 [Vaillancourt]; R v Chaulk, [1990] 3 SCR 1303, 69 Man R (2d) 161 [Chaulk]; R v Swain, [1991] 1 SCR 933, 4 OR (3d) 383 [Swain]; R v DeSousa, [1992] 2 SCR 944, 95 DLR (4th) 595 [DeSousa]; R v Mabior, [2012] 2 SCR 584, [2012] 2 SCR 584 [Mabior]. 7 Ibid. 234 UNBLJ RD UN-B [VOL/TOME 68 the scope of criminal law.8 Federal laws seeking to stigmatize certain behaviours as immoral are more likely to be found to be valid exercises of the criminal law power, though this is not an absolute requirement.9 Conversely, provincial laws that stigmatize conduct by labeling it as immoral are likely to be found ultra vires intrusions into the federal sphere.10 The relationship of criminal law to stigma also attracts interpretive presumptions and even Charter11 attention. The stigmatizing effect of criminal law helps to explain the interpretive preference for subjective fault12 and the Charter may even require full mens rea as a constitutional matter.13 The validity of the sociological and legal models of stigma production is then tested in the context of abortion decriminalization. Based on a case study of abortion law in the Maritime provinces, I argue that criminal law and stigma are less directly connected than either decriminalization advocacy or the Supreme Court’s doctrine suggests. Drawing on sociological insights on stigma production, I show that while criminal law can certainly play a role in stigmatizing conduct, stigmatization is not limited to criminalized conduct nor is it the case that (non- 2017 CanLIIDocs 175 criminal) regulatory law necessarily avoids stigmatization. Conversely, decriminalization may be attended by destigmatization, but not necessarily so. For example, the international experience with the decriminalization of sex work has been mixed.14 Destigmatization may follow decriminalization as was arguably the case in the context of homosexuality,15 or precede it, as in the case of 8 David W Ball, “The Civil Case at the Heart of Criminal Procedure: In re Winship, Stigma, and the Civil- Criminal Distinction” (2010) 38:2 Am J Crim L 117 at 137. 9 Reference re Firearms Act (Can), 2000 SCC 31 at para 54, [2000] 1 SCR 783. For the connection between morality and stigma see: R v Creighton, [1993] 3 SCR 3 at 46 (McLachlin J (as she then was) for the majority and at 19 per Lamer CJ, dissenting), 105 DLR (4th) 632. 10 R v Morgentaler, [1993] 3 SCR 463, 125 NSR (2d) 81 [Morgentaler 1993] 11 Canadian Charter of Rights and Freedoms, Part I of the Constitution Act, 1982, being Schedule B to the Canada Act 1982 (UK), 1982, c 11 (Charter). 12 R v H(AD), 2013 SCC 28, [2013] 2 SCR 269. 13 Vaillancourt, supra note 6; R v Martineau, [1990] 2 SCR 633, 76 Alta LR (2d); DeSousa, supra note 6. 14 Chris Bruckert & Stacey Hannem, “Rethinking the Prostitution Debates: Transcending Structural Stigma in Systemic Responses to Sex Work” (2013) 28:1 CJLS 43, commenting on the continued stigmatization of sex work in New Zealand; Germany, Federal Ministry for Family Affairs, Senior Citizens, Women and Youth, Report by the Federal Government on the Impact of the Act Regulating the Legal Situation of Prostitutes (Prostitution Act), (Berlin: Federal Ministry for Family Affairs, Senior Citizens, Women and Youth, 2007) acknowledging the limited impacts of the Prostitution Act on the perceived immorality of sex work in Germany. This report is noteworthy because it shows the very limited overall effects, positive or negative, of the German Prostitution Act in any dimension including law enforcement, public perceptions or working conditions. Similarly, in New Zealand decriminalization did not effect normalization. Gillian Abel & Lisa Fitzgerald, “Decriminalisation and stigma: Taking the crime out of sex work. New Zealand sex workers' fight for decriminalization” in Gillian Abel et al, eds, Taking the crime out of sex work (Bristol UK: Policy Press, 2010) 239 at 241. 2017] PERFECTLY LEGAL, BUT STILL BAD 235 marijuana use.16 It tends, in any event, to be incomplete.17 Social stigma theory can assist us in predicting when decriminalization will have destigmatizing effects and when stigma is likely to persist. In the fourth and final part of the paper, the implications of abortion decriminalization for future decriminalization efforts in the context of sex work are addressed. This is of broader interest in areas of the law targeting conduct that is tied up in public morality discourses, including laws that implicate the sexual and bodily autonomy of women and sexual minorities. It may also require us to rethink the boundaries between criminal and regulatory law. In the next part, the social science literature on the production and maintenance of stigma is discussed. Social Stigma It has been over 50 years since Goffman’s seminal monograph conceptualized stigma as a deeply discrediting attribute that tainted or discounted someone’s identity.18 Subsequent literature has continued to support a central role of stigma in the 2017 CanLIIDocs 175 construction of social identity, but has drawn increased attention to the social (rather than individual) location of stigma,19 emphasized its close relationship to existing power imbalances20 and refined our understanding of the process of stigma production. Stigma theory has widely accepted two fundamental components: 15 Mariana Valverde, “A New Entity in the History of Sexuality: The Respectable Same-Sex Couple” (2006) 32:1 Feminist Studies 155. Note that destigmatization was attended by a change in label. Valverde rightly points out that the respectable same-sex couple is not merely two homosexuals added together (at 156). 16 Rebecca J Haines-Saah et al, “The privileged normalization of marijuana use – an analysis of Canadian newspaper reporting, 1997–2007” (2014) 24:1 Critical Public Health 47. For the usefulness of media studies as a measure for systemic stigma see Patrick W Corrigan et al, “Newspaper Stories as Measures of Structural Stigma” (2005) 56:5 Psychiatric Services 551. 17 Haines-Saah et al, supra note 16 at 53–54. See also Brenda Cossman, “Lesbians, gay men, and the Canadian Charter of Rights and Freedoms” (2002) 40 Osgoode Hall LJ 223 at 248. 18 Erving Goffman, Stigma: Notes on the Management of Spoiled Identity (London: Penguin, 1990). For a useful overview of the recent theoretical literature on stigma see Bos et al, supra note 5. 19 Gillian Abel & Lisa Fitzgerald, “Decriminalisation and stigma” in Abel et al, eds, supra note 14; Catherine Kohler Riessman, “Stigma and Everyday Resistance Practices: Childless Women in South India” (2000) 14:1 Gender & Society 111–35; Richard Parker & Peter Aggleton, “HIV and AIDS-related stigma and discrimination: a conceptual framework and implications for action” (2003) 57:1 Social Science & Medicine 13; Bruce G Link & Jo C Phelan, “Conceptualizing Stigma” (2001) 27 Annual Rev of Sociology 363; Graham Scambler, “Sex Work Stigma: Opportunist Migrants in London” (2007) 41:6 Sociology 1079; Graham Scambler & Frederique Paoli, “Health work, female sex workers and HIV/ AIDS: Global and local dimensions of stigma and deviance as barriers to effective interventions” (2008) 66:8 Social Science & Medicine 1848. 20 Andrea Krüsi et al, “‘They won’t change it back in their heads that we’re trash’: the intersection of sex work-related stigma and evolving policing strategies” (2016) 38:7 Sociology of Health & Illness 1137. 236 UNBLJ RD UN-B [VOL/TOME 68 recognition of difference and devaluation.21 In a highly influential paper, Link and Phelan proposed an elaborated definition that understands stigma as a process of social exclusion.22 They argue that stigma arises in four distinct steps: 1) A labeling of human difference; 2) A dominant cultural beliefs links a labeled person to an undesirable characteristic (negative stereotype); 3) The labeled person is placed into a distinct category marking them as other; and 4) The labeled person experiences a loss of status and discrimination leading to unequal outcome.23 For Link and Phelan, all four steps are contingent on access to (differential) social, economic and political power.24 They critique an individualized notion of stigma and emphasize its essentially social nature. Link and Phelan review a body of social psychology literature that strongly suggests that these beliefs are applied in preconscious and near automatic ways.25 2017 CanLIIDocs 175 This model of stigma production is helpful in the contexts under consideration, abortion and sex work. In the context of abortion, Kumar et al have proposed a definition of abortion stigma … as a negative attribute ascribed to women who seek to terminate a pregnancy that marks them, internally or externally, as inferior to ideals of womanhood. While definitions of womanhood vary depending on local cultures and histories, a woman who seeks an abortion is inadvertently challenging widely-held assumptions about the ‘essential nature’ of women. 26 A woman who terminates a pregnancy is the cultural target of stigma because she embodies opposition to deeply held cultural beliefs about female sexuality, motherhood and the nurturing nature of women: We hypothesise that there are at least three archetypal constructs of the ‘feminine’ that can be transgressed through an abortion experience: female sexuality solely for procreation, the inevitability of motherhood and instinctual nurturance of the vulnerable.27 21 John F Dovidio, Brenda Major & Jennifer Crocker, “Stigma: Introduction and overview” in Todd F Heatherton et al, eds, The Social Psychology of Stigma (New York: Guilford Press) 1. 22 Link & Phelan, supra note 19. 23 Ibid at 367. 24 Ibid. 25 Ibid at 369. 26 Anuradha Kumar, Leila Hessini & Ellen MH Mitchell “Conceptualising abortion stigma” (2009) 11:6 Culture, Health & Sexuality 625 at 628. 27 Ibid. 2017] PERFECTLY LEGAL, BUT STILL BAD 237 Similarly, sex workers are cultural targets of stigma because they transgress deeply held beliefs about female sexuality, the connection between love and sex, and heterosexual monogamy as the proper (and private) location of sexual relations.28 This is not to suggest that the criminalization of abortion is indistinct from the criminalization of sex work. The criminalization of sex work has been justified on the basis of public nuisance, and sometimes the protection of children and/or public morality. By contrast, the criminalization of abortion has its origins in a desire to expand the criminal law generally, in the protection of the professional monopoly of doctors, and religiously motivated concerns about the soul of the unborn child, particularly at later stages of the pregnancy.29 Interestingly, more recently both abortion restrictions and sex work have been the subject of a women protective discourse.30 The justification for treating abortion and sex work as analogous cases is not, then, that they are closely related offences. Rather, the comparison is justified because in both cases, the law has sought to control, through criminalization, the sexual conduct and bodily autonomy of women based on similar stereotypical beliefs about female sexuality, and the role of women in families and society. One important implication of this focus on the bodies and lives of women is that in both contexts, 2017 CanLIIDocs 175 non-female bodies and identities have often been ignored.31 Criminal Law and Stigma What would happen if we applied the sociological insight into the production of stigma to criminal law? I will argue that there are several legally important implications of the sociological perspective. In this next part, I will discuss three propositions: Firstly, criminal law as much responds to preexisting stigmatizing contexts as it acts as a primary producer of stigma. Secondly, criminal law operates to both support and constrain social stigma. Thirdly, there likely exists a complex set of interactions between social stigma and criminal law stigma.32 28 Helga Kristin Hallgrimsdottir, Rachel Phillips & Cecilia Benoit, “Fallen Women and Rescued Girls: Social Stigma and Media Narratives of the Sex Industry in Victoria, BC, from 1980 to 2005” (2006) 43:3 Can Rev Sociology 265 at 270, 29 Constance B Backhouse, “Involuntary Motherhood: Abortion, Birth Control and the Law in Nineteenth Century Canada” (1983) 3 Windsor Yearbook of Access to Justice 61 at 66, 71. 30 Jula Hughes, Vanessa MacDonnell & Karen Pearlston, “Equality & Incrementalism: The Role of Common Law Reasoning in Constitutional Rights Cases after Bedford (ONCA)” (2013) 44:3 Ottawa L Rev 467; Jay Levy & Pye Jakobsson, “Abolitionist feminism as patriarchal control: Swedish understandings of prostitution and trafficking” (2013) 37:2 Dialectical Anthropology 333 at 337. 31 Trans men have long been absent in abortion discourse, though this may be changing. Michelle Goldberg, “What is a Woman?”, The New Yorker (4 August 2014) 24 at 28, online: 238 UNBLJ RD UN-B [VOL/TOME 68 Applying Link and Phelan’s process theory of stigma production, the creation of a criminal offence could be understood as generating a label that can be attributed to individuals. Criminalizing labels are applied to people who are thought to be deviant.33 A person committing an intentional killing is labeled ‘murderer’, a person committing rape is labeled ‘rapist’. These examples implicitly demonstrate that not all criminal offences effectively perform this labeling function. Despite decades of rape law reform, there is no stigmatizing label of ‘sexual assailant’. Other examples might include negligence-based offences that would require labels like ‘negligent gun-storer’. Decriminalization advocates often seek to detach the previously criminalized conduct from its criminal label. ‘Abortionists’ seek to become abortion providers,34 ‘prostitutes’ become sex workers.35 Some newly minted stigmatizing labels stick (drunk driver), others do not (street racer).36 This is consistent with social stigma theory that in order to produce stigma, the label itself must be associated with negative group characteristics. The offence- creating language of a statute does not produce these group characteristics. Instead, they draw on deeply embedded cultural beliefs. This is borne out by the high variability of stigma associated with offences compared to their seriousness and is 2017 CanLIIDocs 175 true whether we consider maximum or typical punishment or degree or magnitude of harm caused. For example, environmental crimes, cartel and occupational offences may carry significant punishments and can cause widespread physical or economic harms, but perpetrators are not associated with negative stereotypes and the offences do not carry significant social stigma.37 By contrast, possession of child pornography may 32 Link et al note that there are two distinct sociological literatures on labeling. The first addresses the question who gets labeled. The second considers the effect of labeling on individuals. See Bruce G Link et al, “The Social Rejection of Former Mental Patients: Understanding Why Labels Matter” (1987) 92:6 American Journal of Sociology 1461 at 1462. There is a body of criminological research flowing from this latter literature that inquires into possible criminogenic effects of labeling referred to as “labeling theory”. I am not concerned with this literature here. Instead, I consider the implications of the first body of literature. Link et al describe it as involving “questions about whether individuals with varying status characteristics (male vs. female, black vs. white, etc.) are exposed to different societal reactions and thus to very different labeling experience.” 33 In its Greek root, the meaning of stigma was a tattoo, a mark used for criminals, slaves and soldiers to declare them publicly to be under the direction or property of others. In Christian legends, the stigmata were the wounds inflicted on the crucified Christ and replicated in ardent followers as spontaneous bleedings of the hands and feet. Stigma has also been identified with the mark of Cain in the Book of Genesis. See Schlomo Shoham, The Mark of Cain: The Stigma Theory of Crime and Social Deviation (Jerusalem: Israel Universities Press, 1970). 34 Carole Joffe, Doctors of Conscience: The struggle to provide abortion before and after Roe v. Wade (Boston: Beacon Press, 1995) at 158. 35 Noah D Zatz, “Sex Work/Sex Act: Law, Labor, and Desire in Constructions of Prostitution” (1997) 22:2 Signs: J Women in Culture & Society 277 at 300. 36 Fraser McGuire et al note that drunk driving carries extremely high social stigma in Canada. “Driving under the Influence of Cannabis or Alcohol in a Cohort of High-frequency Cannabis Users: Prevalence and Reflections on Current Interventions” (2011) 53:2 Can J Criminology and Criminal Justice 247 at 252; Andrew Leigh, “Youth and street racing” (1995) 7:3 Current Issues Criminal Justice 388 at 390. 2017] PERFECTLY LEGAL, BUT STILL BAD 239 cause only remote harms and may attract only modest punishment, but is associated with extremely negative beliefs and highly stigmatized.38 Therefore, the degree of stigma does not merely depend on the severity or label of a criminal offence, but on the nature of the group characteristics with which the label is associated. Criminal offences such as loitering or prostitution are associated with marginalized groups like low-income people, homeless people or people with drug addictions. These offences tend to appear more severe as a result of social stigma. Negative stereotypes about gender, race and other marginalizations operate to amplify or even exaggerate criminal law stigma by bringing social stigma into play. Kumar et al consider the example of abortion stigma and its relationship to social constructs and negative stereotypes. They note: The power dynamics that underline abortion are part of an ideological struggle about the meaning of family, motherhood and sexuality (Petchesky 1990). … Sexual activity, specifically female sexuality, is at 2017 CanLIIDocs 175 the core of abortion stigma because it may amplify transgressions of stated norms about who, when, why and how to have sex (Bleek 1981, Gilmore and Somerville 1994). Similarly, suitability for motherhood and acceptability of pregnancy termination is determined by a host of individual characteristics including socio-economic status, occupation, race or ethnicity and age. 39 The social control exerted over female sexuality is an important contributor to, and informs the content of, abortion stigma. This is similar to stigmatization related to sexual assault victims. Rape stigma continues to draw on deeply embedded constructs of protecting good feminine sexuality related to virginity and procreation, while not protecting bad feminine sexuality enacted outside of adulthood, marriage and motherhood.40 As a result of stigma adhering to these negative stereotypes, 37 Girard et al aptly describe the dominant discourse on environmental crimes as constituting “the unfortunate but inevitable ‘price of prosperity’” April L Girard, Suzanne Day & Laureen Snider, “Tracking Environmental Crime through CEPA: Canada‘s Environment Cops or Industry’s Best Friend?” (2010) 35:2 Can J Sociology 219 at 237. Interestingly, being a victim of industrial pollution has been documented to be stigmatizing. This is because living in polluted areas is an incident of poverty, a highly stigmatizing context. See Judith Bush, Suzanne Moffatt & Christine Dunn, “‘Even the birds round here cough’: stigma, air pollution and health in Teesside” (2001) 7:1 Health & Place 47. For similar concerns regarding corporate criminal liability for occupational health and safety and cartel crimes see: Steven Bittle & Laureen Snider, “‘Moral Panics’ Deflected: The Failed Legislative Response to Canada’s Safety Crimes and Markets Fraud Legislation” (2011) 56:4 Crime L & Social Change 373. 38 Laura J Zilney & Lisa Anne Zilney, Perverts and Predators: The Making of Sexual Offending Laws (Lanham, MD: Rowman & Littlefield Publishers, 2009) at 53. 39 Kumar et al, supra note 26 at 628. 240 UNBLJ RD UN-B [VOL/TOME 68 sexual assault labeling and criminalization is often ineffective in seeking to extend the protective cloak of the law to bad feminine sexuality. Some stigmatizing labels find their sources in laws outside of the Criminal Code41 and related penal laws. For example, divorce laws used adultery as a legal marker, a label that was effective in producing stigma despite not being located in a penal law. This suggests that as long as the constructs supporting stigmatization are sufficiently negative, strong and pervasive, penal laws are not required to create stigma. Conversely, if there are no such constructs, the use of penal laws, without more, is unlikely to lead to stigmatization. The third step in the production of stigma, the marking of members of the stigmatized group as ‘other’ can readily be theorized in law as a commonly desired function of criminal law: the symbolic and bodily separation of offenders from the community. However, as we will see, there is a mismatch between the criminal law theory of a procedurally careful and substantively rational allocation of blame in a criminal trial and the social othering through stigmatization as a preconscious and instantaneous act posited by social stigma theory. 2017 CanLIIDocs 175 For the final step, the accrual of detriment to stigmatized individuals, criminal law theorizes this as the rational application of sentencing laws and the stigmatization of the convict as a criminal, ultimately controlled by legal and constitutional requirements of proportionality between the severity of the offence, the circumstances of the offender and the quantum of punishment. Again, there is somewhat of a mismatch between the criminal law theory and stigma theory. The criminal law theory contemplates a careful weighing of relevant criteria, stigma theory suggests that stigmatization will not be the result of a careful and rational process but rather driven by snap judgment and stereotypes. The application of concepts developed outside of law to legal doctrine requires some methodological caution. In this case, it appears justified however because stigma has long played an important role in the criminal law jurisprudence of the Supreme Court of Canada. Stigma is a core concept that performs two related functions: it guides the interpretation of offences; and it determines the scope of true criminal law. The starting point was a significant expansion in regulatory law. This was driven by the use of the vehicle of criminal or quasi-criminal prohibition for enforcing administrative regimes. These regimes tended to impose fines and occasionally even imprisonment for violations in the absence of any statutory language directing a fault requirement. 40 Dominic Abrams et al, "Perceptions of Stranger and Acquaintance Rape: The Role of Benevolent and Hostile Sexism in Victim Blame and Rape Proclivity"(2003) 84:1 J Personality & Social Psychology 111 at 113. 41 RSC, 1985, c C-46. 2017] PERFECTLY LEGAL, BUT STILL BAD 241 In the 1970s, the Court began to explore the role of stigma for the distinction between regulatory and truly criminal prohibitions. From the beginning, stigma was an important criterion for distinguishing between true crimes and regulatory offences. In Pierce Fisheries, a case dealing with the offence of possession of undersized lobsters, the Court had to determine whether a company employee had to be aware of the presence of undersized lobsters in order to convict.42 The Court relied on the 1895 ruling in Sherras v. De Rutzen to draw a distinction between quasi-criminal acts (what we might today call regulatory offences) and acts of a truly criminal character.43 The majority concluded that knowledge was not required for regulatory offences while affirming that proof of knowledge or intention continued to be a requirement for true crimes and that such a requirement would be read into the offence where the language of statute was silent.44 The key reason for the distinction was that stigma attached to a conviction for a true crime and greater stigma yet to serious crimes. But when one comes to acts of a truly criminal character, it appears to me that there are at least two other factors which any reasonable legislator would have in mind. In the first place a stigma still attaches to any person 2017 CanLIIDocs 175 convicted of a truly criminal offence, and the more serious or more disgraceful the offence the greater the stigma.45 It turned out that any categorical distinction was difficult to sustain. In the seminal case of Sault Ste Marie, the municipality was charged with an environmental offence. Justice Dickson (as he then was) rejected the Crown's argument that the absence of stigma justified the imposition of an absolute liability offence. He noted that not only had the defendant been put to significant expense, but also to the “opprobrium” of a criminal conviction.46 With the advent of the Charter, stigma took on an even more central role. It became the deciding factor not only in interpreting criminal offences for which Parliament had not identified the requisite mental element, it also gave rise to a series of constitutional demands for fault requirements for some offences that the Court described as exceptionally stigmatizing including murder and theft. In these areas, the Court’s theory of stigma first moved from the generic idea that criminal conviction is stigmatizing to a more elaborate theory which particularized some offences as more stigmatizing than others. 42 Pierce Fisheries, supra note 6. 43 Ibid, citing Sherras v De Rutzen, [1895] 1 QB 918. 44 Ibid at 15–17. 45 Ibid at 15, citing Sweet v Parsley, [1969] 2 WLR 470 at 474. 46 Sault Ste Marie, supra note 6 at 1311–1312. 242 UNBLJ RD UN-B [VOL/TOME 68 Such is theft, where, in my view, a conviction requires proof of some dishonesty. Murder is another such offence. The punishment for murder is the most severe in our society and the stigma that attaches to a conviction for murder is similarly extreme.47 The Court did not explain why dishonesty and murder were singled out, nor did the Court suggest what other offences would attract this kind of enhanced stigma. What is clear, however is that the Court seems to be of the view that criminal law stigma is finely calibrated. Some offences carry great stigma (and thus require the Crown to prove subjective fault as a matter of constitutional law), while most others carry less stigma and may only require proof of negligence. This point is made succinctly in DeSousa, a case involving negligent injury to an innocent bystander: As this Court has not indicated that fundamental justice requires fault based on a subjective standard for all offences, the mental element required by s. 269 [unlawfully causing bodily harm] passes constitutional muster unless s. 269 is one of those few offences which due to its stigma and penalty require fault based on a subjective standard. I agree with the respondent and interveners that s. 269 has neither the stigma nor criminal 2017 CanLIIDocs 175 sanction to require a more demanding mental element than it already has.48 The stigma associated with conviction will generally reflect the degree of opprobrium attached to the underlying offence. The stigma attached to an offence will in turn influence the minimum fault requirement for that offence.49 Sometimes, the Court seems to anticipate a careful calibration of stigma. In other cases, the fact of criminal conviction and punishment is a generically stigmatizing event and no resort is made to degrees of offence-specific stigma.50 To complicate matters further, the Supreme Court has at times had to contend with intersecting sources of stigma. For example, in Chaulk51 and Swain,52 two cases involving special legal regimes for people who cannot be convicted because they suffered from serious mental illness at the time of the offence, the Court had to confront the intricacies of mental illness stigma and its interaction with criminal conviction stigma. The cases demonstrate the difficulty of integrating criminal law and social forms of stigmatization into a coherent factual and doctrinal scheme. 47 Vaillancourt, supra note 6 at 653–654. Vaillancourt successfully challenged the constitutionality of felony murder. The Court concluded that “stigmatizing the crime as murder unnecessarily impairs the Charter right” under s 7 (at 660). 48 DeSousa, supra note 6 at para 962. 49 Ibid. 50 Mabior, supra note 6 at para 15. 51 Chaulk, supra note 6. 52 Swain, supra note 6. 2017] PERFECTLY LEGAL, BUT STILL BAD 243 In Chaulk, the majority rejected the argument that imposing an onus on the accused to prove his or her own insanity furthered the objective of avoiding “stigmatizing and punishing, as criminals, people who are ‘sick’ as opposed to ‘bad’.”53 Instead, the majority preferred to see the provision as merely responsive to the evidentiary difficulties of proving someone’s sanity beyond a reasonable doubt. In so doing, the majority remained focused on the stigma arising from criminal conviction and punishment to the complete exclusion of any consideration of the social stigma of mental illness. Only one year later, a challenge was raised to the automatic detention scheme for not criminally responsible by reasons of mental disorder. This time, the majority stayed focused on the stigma of mental illness, to the complete exclusion of the stigma of criminal conviction and punishment. It is only in the reasons of Justice Wilson, in a concurring opinion, that the two forms of stigma were briefly addressed together. She noted: An insane acquittee is detained at the pleasure of the Lieutenant Governor, often for a period exceeding that which would have been possible upon conviction. He must also live with the stigma of being held to be both a criminal and insane and may face 2017 CanLIIDocs 175 conditions worse than those obtaining in prison.54 A similar difficulty occurs in cases where the Court has had to acknowledge that stigma can attach to interactions with the criminal justice system prior to or in the absence of ‘just desert.’ For example, stigmatization is acknowledged in Kang- Brown in the context of an investigation,55 in Stone as arising from a charge56 and arising from being remanded into custody before trial in Morales.57 The right to a speedy trial was supported by the stigma that accrues when a person is subject to criminal proceedings in Morin.58 What is interesting about these cases is that they acknowledge that the aura of stigma resulting from criminalization reaches far beyond the criminal law stigma that results from a valid criminal conviction for wrongful conduct that has been proven beyond a reasonable doubt. Finally and returning briefly to the distinction between criminal and regulatory offences, the Supreme Court has consistently held that being subject to administrative law mechanisms such as human rights complaints and remedies does not invoke (much) stigma,59 though it has acknowledged that some of the underlying 53 Chaulk, supra note 6 at 1336. 54 Swain, supra note 6 at 1027–1028. 55 R v Brown, 2008 SCC 18, [2008] 1 SCR 456. 56 R v Stone, [1999] 2 SCR 290, 173 DLR (4th) 66. 57 R v Morales, [1992] 3 SCR 711, 17 CR (4th) 74. 58 R v Morin, [1992] 1 SCR 771, 12 CR (4th) 1. 59 Canada (Human Rights Commission) v Taylor, [1990] 3 SCR 892 at 932–933, 75 DLR (4th) 577. 244 UNBLJ RD UN-B [VOL/TOME 68 regulated conduct may well be stigmatizing.60 The Court seems to assume that there is a connection between the underlying conduct and/or its becoming known in the community, and the regulatory or punitive response, while at the same time maintaining a conceptual difference between the two forms of stigmatization. To summarize, stigma is a core concept in Canadian criminal law. Offences that carry stigma are characterized as being true crimes, while non-stigmatizing offences are held to be merely regulatory in nature. Because of their stigmatizing nature, true crimes require proof of subjective fault such as knowledge of the relevant facts or the intentional commission of the crime even where the language of the statute does not appear to require such proof. Parliament can normally create offences that require mere proof of negligence, but the Charter may require subjective fault for high stigma offences regardless of Parliamentary intent. The Court has developed a theory of stigma production that attaches stigma to individuals for particularized conduct through a procedurally controlled and substantively fair process. Despite this central role, criminal law stigma remains undertheorized by the Court. This is particularly apparent in areas of intersectional stigma such as when mental illness stigma intersects with criminal law stigma, in areas where the timing 2017 CanLIIDocs 175 of the stigmatization precedes the conviction and in areas where penal law is not implicated, but the underlying conduct is highly stigmatized. However, even outside these areas, the theoretical foundations for criminal law stigma remain uncertain. It is unclear why some offences such as theft are considered high stigma and others including sexual assault and genocide are not. The Court’s theory of stigma production is difficult to reconcile with social stigma theory because it locates stigma in the individual and posits that stigmatization is the result of a carefully calibrated, rational and procedurally controlled process, while social stigma theory emphasizes the diffuseness of stigmatization, its spontaneous and pre-rational elements and its basis in preexisting negative stereotypes. Abortion Stigma In this part of the paper, I consider stigma production through observation of a natural experiment, viz. the transformation of abortion law in the Maritime provinces following the complete decriminalization of abortion in 1988. Since that time, all three Maritime provinces (i.e. New Brunswick, Nova Scotia, and Prince Edward Island), have adopted more or less restrictive regulatory regimes governing abortion access relying on provincial jurisdiction over health. If the Supreme Court’s distinction between the effects of regulatory and true criminal law is meaningful, it would be reasonable to predict that decriminalization of abortion should have been attended by some degree of destigmatization even if regulatory law continues to constrain previously criminalized conduct. 60 An example of the latter is Blencoe where the Court was urged to analogize delay in human rights proceedings to violations of speedy trial rights because of the stigma attached to allegations of sexual misconduct. Blencoe v British Columbia (Human Rights Commission), 2000 SCC 44, [2000] 2 SCR 307. 2017] PERFECTLY LEGAL, BUT STILL BAD 245 If abolitionists are correct that decriminalization leads to normalization, this should be observable in the abortion context since analogous arguments have animated anti-choice advocacy for a long time. The argument is that by permitting women to choose whether and when to carry a pregnancy to term, the fundamental dignity of all human life is threatened and the taking of a human life is normalized.61 And finally, if decriminalization advocates are right to hope that decriminalization will lead to access to protective regulatory regimes like labour relations and workplace health and safety laws, we should be able to see this in abortion access. Medicare is the abortion analogue to occupational health and safety, a state-funded mechanism for the protection of the health and wellbeing of patients and workers respectively. The issue seems timely. There is an active discussion about decriminalization of sex work and some other criminalized conduct, and neither patients accessing abortion nor abortion providers have been the subject of criminal prosecution since 1988. It is important to take somewhat of a long view because criminalization histories are often long, but the concurrent decriminalization of 2017 CanLIIDocs 175 homosexuality suggests that such a process could have substantially run its course by now. It appears reasonable to assume that enough time has passed to permit even complex social processes such as destigmatization to occur.62 In 1988, when the Supreme Court of Canada ruled in R v Morgentaler,63 abortion had been part of New Brunswick criminal law for almost 180 years. In 1810, New Brunswick had been a leader in criminalizing abortion by passing a law modeled on an infamous omnibus bill known as Lord Ellenborough's Act of 1803 which prohibited the procurement of a miscarriage, though not by the pregnant woman herself.64 The offence applied to established pregnancies after foetal movement could be detected by the pregnant woman (quickening). In 1842, New Brunswick abolished the requirement that quickening had occurred, making the procuring of a miscarriage an offence from the beginning of a pregnancy, including criminalizing attempts when the woman was not in fact pregnant. In another amendment New Brunswick also introduced a minimum sentence of three years for the offence.65 Until this time, criminalization exclusively targeted the abortion provider. In 1849, however, New Brunswick criminalized abortion by the pregnant woman herself. In 1869, Parliament consolidated the 61 Johanna Schoen, Abortion after Roe (Chapel Hill: The University of North Carolina Press, 2015) at 89. 62 A history of Canadian abortion law is recounted by the Supreme Court of Canada in Tremblay v Daigle, [1989] 2 SCR 530, 62 DLR (4th) 634 [Tremblay] where the Court treats it as the law of fetal rights. 63 R v Morgentaler, [1988] 1 SCR 30, 44 DLR (4th) 385 [Morgentaler 1988]. 64 Tremblay, supra note 62 at 566. 65 Backhouse, supra note 29 at 70. 246 UNBLJ RD UN-B [VOL/TOME 68 criminal law applicable to all the provinces and adopted abortion provisions identical to New Brunswick law, with a penalty of life imprisonment.66 Upon codification in 1892, s. 272 of the Criminal Code made it a crime punishable by life imprisonment to attempt to procure a woman’s miscarriage whether or not she was with child. This remained the law until 1969 when the Trudeau government decriminalized abortions approved by a therapeutic abortion committee of at least three doctors.67 The Criminal Code remained the vehicle for this essentially regulatory scheme. As is well known, the 1969 scheme did not survive constitutional review under the Charter and in 1988, it was struck down.68 From there, the stage was set for provincial governments to either treat abortion as a health care service like any other, create special assistive regimes for access,69 or adopt restrictive laws, regulations and policies. Maritime provinces all went the latter route, but not in an identical manner. Triggered by Dr. Morgentaler’s announcement of a plan to open clinics in Nova Scotia and New Brunswick, the Nova Scotia government promulgated regulations and subsequently passed legislation prohibiting abortions outside of 2017 CanLIIDocs 175 hospitals in March of 1989.70 This constituted a dual attack on the newly won right of Nova Scotians to access legal abortions without administrative strictures. The Medical Services Act prohibited clinic abortions while regulations under the Act defunded them. Despite this, Dr. Morgentaler opened the Morgentaler Clinic in Halifax in 1990 and was promptly prosecuted in 14 cases. In defending the prosecutions, Dr. Morgentaler successfully challenged the constitutionality of the Act and the regulations and in October of 1990, the Medical Services Act provisions and the regulation were struck down by the Nova Scotia Supreme Court.71 That decision would eventually be upheld by the Supreme Court of Canada in 1993.72 The Halifax Morgentaler clinic operated for 13 years and closed in 2003. The vast majority of abortions are now provided at a single hospital in Halifax. Stigma continues to 66 Ibid at 75. 67 Tremblay, supra note 62 at 567. 68 Morgentaler 1988, supra note 63. For an analysis of Canadian as well as some comparator abortion decisions see: Vanessa MacDonnell & Jula Hughes, “The German Abortion Decisions and the Protective Function in German and Canadian Constitutional Law” (2013) 50 OHLJ 999. 69 An example is Ontario where the province used to cover travel costs for Northern patients under s. 3 of Northern Health Travel Grant, O Reg 20/94, repealed in 2000. 70 Joanna Erdman noted that this provincial response was by no means restricted to the Maritimes but also reached to BC and Manitoba. She commented that “many of these laws and regulations were challenged on jurisdictional grounds. Some survived scrutiny, while others were defeated. In response to invalidation, some provinces enacted amended versions of laws and regulations to overcome courts' objections.” [footnotes omitted] Joanna N Erdman, “In the Back Alleys of Health Care: Abortion, Equality, and Community in Canada” (2006-2007) 56 Emory LJ 1093 at 1094. 71 Nova Scotia (AG) v Morgentaler, [1990] 96 NSR (2d) 54, 253 APR 54 (CA) 72 Morgentaler 1993, supra note 10. 2017] PERFECTLY LEGAL, BUT STILL BAD 247 surround the procedure and access remains constrained, particularly in rural areas and in Cape Breton.73 Despite the legal successes of their Nova Scotia neighbours in the Supreme Court of Canada, New Brunswickers and Prince Edward Islanders did not experience any practical or even discourse effects of either the 1988 or the 1993 Morgentaler cases.74 Prior to 1988, the difficulty with abortion access in many parts of Canada had been documented by a federal report on the operation of the 1969 law. The 1977 Badgley Report had commented on patchy access following the 1969 amendments: Coupled with the personal decisions of obstetricians-gynaecologists, half of whom (48.9 percent) in eight provinces did not do the abortion procedure in 1974-75, the combined effects of the distribution of eligible hospitals, the location of hospitals with therapeutic abortion committees, the use of residency and patient quota requirements, the provincial distribution of obstetricians and gynaecologists, and the fact that the abortion procedure was done primarily by this medical specialty resulted in sharp regional disparities in the accessibility of the abortion procedure. … What this means is that the procedure in the Criminal Code for 2017 CanLIIDocs 175 obtaining abortion is in practice illusory for many Canadian women.”75 It is noteworthy that the Badgley Report found that at least five hospitals (or maybe more, the Report is not entirely clear on this point) in New Brunswick performed 440 funded hospital abortions in 1974.76 The situation did not shift appreciably following the 1988 and 1993 Morgentaler cases. In 2011, there were only two hospitals in New Brunswick, which provided 414 funded abortions.77 The legal treatment of abortion changed significantly between 1974 and 2011, but neither access nor stigma were transformed. The New Brunswick regulatory regime enacted following the 1988 Morgentaler decision warrants a closer look. The New Brunswick Liberal government under Premier Frank McKenna promulgated a regulation disentitling abortion from Medicare funding unless performed in a hospital by a specialist after two doctors certified that the procedure was medically 73 Cape Breton does not have an abortion clinic. Residents have to travel to the QEII Health Sciences Centre in Halifax to access abortion services. (Action Canada for Sexual Health & Rights, “Service Providers”, online: 74 Immediately following the 1988 decision, provincial resistance was more widespread. For similar observations about Alberta, see: Ian Urquhart, “Federalism, Ideology, and Charter Review: Alberta’s Response to Morgentaler” (1989) 4 Can JL Soc 157 at 160. However, Alberta liberalized abortion access in the wake of R v Morgentaler, [1993] 1 SCR 462, 1993 CanLII 158. 75 Canada, Department of Justice, “Report of the Committee on the Operation of the Abortion Law (Badgley Report)”, (Ottawa: Minister of Supply and Services, 1977) at 140–141. 76 Ibid at 112. 77 Canadian Institute for Health Information, “Induced Abortions Quick Stats, 2011” (Ottawa: Canadian Institute for Health Information 2014) at 1. 248 UNBLJ RD UN-B [VOL/TOME 68 required.78 A prohibition of clinic abortions was already on the books under a private act, the New Brunswick Medical Act, which regulates the medical profession.79 When Dr. Morgentaler bought clinic property in Fredericton in 1992 and opened an abortion clinic in 1994, Premier McKenna vowed to give Dr. Morgentaler the fight of his life.80 Immediately upon opening the clinic on July 5, 1994, Dr. Morgentaler was restrained by Order of the Council of the College of Physicians and Surgeons from performing abortions outside a hospital relying on sections 56(b.1) and 56.2 of the Medical Act.81 On September 14, 1994, a judge of the Court of Queen’s Bench declared the legislative provisions upon which the College relied unconstitutional.82 One week later, the Fredericton Morgentaler Clinic reopened. The exclusion of clinic-based abortions from Medicare continued for the entire time the Fredericton Morgentaler was in operation. In 2014, Liberal Premier Brian Gallant promised to remove all barriers to abortion in New Brunswick by amending Regulation 84-20, and removing the requirement that the procedure be conducted by a specialist after the certification by two doctors that the procedure was medically necessary.83 As of April 2017, the hospital requirement for funded abortion care continues, excluding the successor Clinic 554 from Medicare funding for abortion 2017 CanLIIDocs 175 care.84 The path to abortion access has been even more fraught in Prince Edward Island. In the wake of Morgentaler 1988, PEI adopted an unwritten and secret policy “that it will only pay for an abortion deemed to be a medical necessity provided that the abortion is performed at a hospital. The determination of whether an abortion is a medical necessity for payment purposes, is determined by a Medical Advisory Committee of five (5) doctors, provided for under the Health Services Payment Act, and appointed by the… Commission”.85 The terms of the policy were only disclosed to Dr. Morgentaler after litigation was commenced. On April 22, 1993, Dr. Morgentaler applied for a declaration that the abortion policy of the Government's Agency was ultra vires. The application was scheduled for hearing on April 28, 1994. On April 26, 1994, the 78 NB Reg 84-20, Schedule 2, (a.1). 79 SNB 1981, c 87. 80 Ottawa Citizen (18 February 1998) A3. 81 Morgentaler v New Brunswick (AG), [1994] 152 NBR (2d) 200 at para 25, 117 DLR (4th) 753, aff’d [1995] 156 NBR (2d) 205, 121 DLR (4th) 431, leave to appeal ref’d [1995] 124 DLR (4th) vi (note), 164 NBR (2d) 320 (note). 82 Ibid. 83 CBC News, “New Brunswick abortion restrictions lifted by Premier Brian Gallant”, CBC News (26 November 2014), online: 84 NB Reg 84-20, Schedule 2, (a.1). 85 Morgentaler v Prince Edward Island (Minister of Health & Social Services), [1994] 117 Nfld & PEIR 181, 112 DLR (4th) 756 at 756 (PEISC(TD)). 2017] PERFECTLY LEGAL, BUT STILL BAD 249 Executive Council issued Order in Council No. EC220/94 regularizing the policy. The Regulation was struck down in 1995,86 but an appeal was allowed by the PEI Supreme Court Appeal Division in 1996, permitting legal impediments to funded abortions to persist.87 In 2014, McQuarrie et al published a report on the experiences of PEI women trying to navigate reproductive health care in the province.88 Their research documented what abortion rights advocates in the Maritime provinces had known anecdotally for decades: Maritime patients routinely experience insurmountable barriers to accessing reproductive health care, leading to coerced continued pregnancies, at times dangerous attempts at self-managed abortions, and desperate travels on and off the island in search of care. Her report also speaks poignantly to the persistence of abortion stigma.89 In all three Maritime provinces, both people accessing abortions and abortion providers remain heavily stigmatized despite over two decades of complete decriminalization.90 The persistence of social stigma in the wake of decriminalization calls into question the Supreme Court’s assumptions about the production of stigma 2017 CanLIIDocs 175 related to criminal behaviour as arising from and being peculiar to criminal law. It also casts doubt on the Court’s assertion that regulated behaviour is intrinsically less stigmatized than criminalized behaviour. Focusing on decriminalization and its aftermath renders visible that the interaction between criminal law and stigma is more complex than is suggested by the Supreme Court. 86 Morgentaler v Prince Edward Island (Minister of Health & Social Services), [1995] 126 Nfld & PEIR 240 (SC), 122 DLR (4th) 728 (PEISC(TD)). 87 PEI (Minister of Health and Social Services) v Morgentaler, [1996] 144 Nfld & PEIR 263; 139 DLR (4th) 603; 45 Admin LR (2d) 245 (PEISC(AD)). 88 Colleen MacQuarrie, Cathrine Chambers & Jo-Ann MacDonald, Trials and Trails of Accessing Abortion in PEI: Reporting on the Impact of PEI's Abortion Policies on Women (Charlottetown, PEI: University of Prince Edward Island, 2014). 89 Based on the MacQuarrie report, a group of advocates supported by LEAF filed notice of a constitutional challenge to the policy in January of 2016. Premier MacKay conceded the suit and promised abortion access in PEI by the end of 2016. 90 My claim for Nova Scotia and New Brunswick is anecdotal and observational. It is based on a decade of working with abortion rights advocates, abortion providers, clinic managers as well as being a visible figure in the struggle for abortion access in New Brunswick. There is a live debate about measuring abortion stigma. Bos et al, supra note 5 have recommended continued exploration of qualitative methods in stigma research, and MacQuarrie, supra note 88 is a recent example, while others have experimented with quantitative methods including Lisa A Martin et al, “Measuring Stigma Among Abortion Providers: Assessing the Abortion Provider Stigma Survey Instrument” (2014) 54:7 Women & Health 641; Annik M Sorhaindo et al, “Constructing a validated scale to measure community-level abortion stigma in Mexico” (2016) 93:5 Contraception 421. 250 UNBLJ RD UN-B [VOL/TOME 68 Implications for sex work The basic premise of abolitionists and decriminalization advocates that decriminalization leads to destigmatization and normalization is doubtful in light of the experience of decriminalization of abortion. This is because both abortion and sex work stigma are tied up in stereotypical beliefs about the role of women and female sexuality. These stereotypical beliefs are likely to sustain stigmatization in a decriminalized legal context. This is not to suggest that decriminalization has no effects, far from it. There were real gains in terms of health outcomes for women following the decriminalization of abortion and at least for a time, there was a considerable amount of public debate. However, we can expect that if sex work were to be decriminalized, at least some provinces and municipalities would use their legislative and regulatory powers to enact or expand existing regimes that seek to replicate criminalization, including the creation or use of provincial offences that threaten imprisonment.91 These regulatory laws are likely to be effective in performing the labeling function required for continued stigma production. At the same time, advocacy for improvements would be more difficult as 2017 CanLIIDocs 175 the social movements supporting change will be split into winners and losers.92 Sex workers living and working in provinces that choose not to regulate or regulate in a protective or supportive manner will be less likely to advocate strongly for sex workers in provinces and municipalities that use their jurisdiction over community safety, zoning or licensing for the purpose of restricting or minimizing prostitution. As the group of advocates shrinks, targets for advocacy will become more diffuse and more difficult to identify. Rather than asking a single federal government to change the law or challenging a Criminal Code prohibition in court, sex work advocates would need to tackle various provincial and municipal regimes. This would decrease the ability to mobilize at a national level. From an abolitionist perspective and drawing on the experience with the decriminalization of abortion, it seems that the concern about normalization is not well founded. Abolitionists should find comfort in the fact that decriminalization does not, without more, cause normalization. That said, the abortion analogue also does support the related claim that asymmetrical criminalization of clients will be effective in redirecting stigma to new targets. In other words, it is unlikely that criminalizing clients will lead to stigmatized clients and destigmatized sex workers. Instead, the abortion experience suggests that stigma targets are both diffuse (anyone 91 Provinces already use their jurisdiction over traffic, community safety and child protection to regulate prostitution. See Canada, Library of Parliament, “Prostitution in Canada: International Obligations, Federal Law, and Provincial and Municipal Jurisdiction”, by Laura Barnett & Julia Nicol, Legal and Legislative Affairs Division, Parliamentary Information and Research Service (Ottawa: Library of Parliament, 2012), online: 92 Jula Hughes, Vanessa MacDonnell & Karen Pearlston, “Equality & Incrementalism: The Role of Common Law Reasoning in Constitutional Rights Cases after Bedford (ONCA)” (2013) 44:3 Ottawa L Rev 467 at 473–474. 2017] PERFECTLY LEGAL, BUT STILL BAD 251 with a connection to the issue is stigmatized) and less than rational. This is supported by social stigma theory where variations of this kind of broad stigmatization of associated identities are known as courtesy stigma,93 contamination stigma or vicarious stigma.94 The abortion experience further supports the view that social stigma is not very sensitive to changes in structural stigma such as criminal legislation. This is because social stigma is produced in a complex process that includes the identification of difference, stereotypical beliefs, labeling and discrimination. In the context of abortion, people experiencing unintended pregnancy are identified as different through community expectations that pregnancy should be planned and occur in the context of monogamous relationships, in which both partners are mature, educated and financially stable.95 In the context of prostitution, difference is identified in the role of sex (private vs. public, for fun/procreation vs for money). As discussed above, stereotypical beliefs about people with unintended pregnancy include various transgressions of feminine ideals such as caring, sexual restraint/purity and mothering. Similarly, in the context of sex work, positive ideals 2017 CanLIIDocs 175 of womanhood such as cleanliness, propriety and family orientation collide with perceptions of sex work as dirty, perverse or undermining family values. Law plays a role in labeling actors and conduct, but it is not apparent in the abortion context that regulatory laws are any less effective in performing labeling functions than criminal legislation. Changes in the criminal law applicable to sex work may lessen stigma associated with prostitution if no special laws are enacted federally or provincially, but even protective and benevolent legislation can have labeling effects as is evident in the context of social benefit legislation.96 Finally, discriminatory effects of stigmatization may persist even if the stigmatizing law is removed or altered. The ‘stickiness’ of discriminatory effects is evident in the marginalization of people in contexts where laws specifically limit or prohibit discrimination. The lesson from abortion law, then, is that stigma production is a complex process that is very imperfectly controlled by law. Decriminalization, without more, 93 Goffman, supra note 18. 94 Patrick W Corrigan & Frederick E Miller, “Shame, blame, and contamination: A review of the impact of mental illness stigma on family members” (2009) 13:6 J Mental Health 537. 95 Whitney Smith et al, “Social Norms and Stigma Regarding Unintended Pregnancy and Pregnancy Decisions: A Qualitative Study of Young Women in Alabama” (2016) 48:2 Perspectives on Sexual & Reproductive Health 73 at 75–76. 96 Michael Bratton, “Welfare Discourse and the Subjectivity of Social Assistance Recipients: Understanding Classism as a Barrier to Justice” (2015) 17:1 Can Social Work 40 at 44–45. 252 UNBLJ RD UN-B [VOL/TOME 68 will not lead to destigmatization. This is important for proponents of decriminalization who should be ready to address all aspects of stigma production rather than relying on law reform as a cure-all. Conversely, abolitionists have less reason to fear decriminalization than abolitionist advocacy suggests. At the same time, abolitionists have likely exaggerated the ability of law to shape and specifically direct social stigma formation. The broader questions of best policy to ensure the safety of sex workers, promote sexual autonomy and general equality rights for women remain the subject of important debate. Appreciating the limited and complex role of criminal law in the production and maintenance of social stigma associated with sex work should assist in advancing rational policy development in this area. 2017 CanLIIDocs 175 THE RIGHT TOOL FOR THE JOB?: FREEDOM OF ASSOCIATION UNDER PROVINCIAL HUMAN RIGHTS CODES Keir Vallance* “It is a vain thing to imagine a right without a remedy.”1 1. Introduction Over the last decade, the Supreme Court of Canada has provided an expanding framework of labour rights that are protected by the constitutional guarantee of freedom of association2 under the Canadian Charter of Rights and Freedoms.3 It has now been recognized that freedom of association in the labour context protects the rights to form, join, and maintain trade unions, including a right to choice of 2017 CanLIIDocs 175 bargaining agent which is independent from one’s employer; to collectively bargain, including a duty on employers to bargain in good faith; and to strike. But despite the, unquestionably profound, changes in s. 2(d) jurisprudence, the reach of freedom of association under the Charter remains limited. This is because the Charter applies only to government action or government actors, not to private parties or the relationships between private parties.4 Hence, while the Charter will apply directly to labour relations between a government and its employees,5 that is because of the government’s role as legislator, not as an employer qua employer. The Charter does not directly protect workers within the private sector.6 In other * Assistant Professor, College of Law, University of Saskatchewan. I would like to thank my anonymous peer reviewers, particularly the inimitable Peer Reviewer #1, whose detailed notes left this paper, I think, much-improved; and thanks to the student editors of the UNB Law Journal for their unfailing courtesy. 1 Ashby v White (1703), 92 ER 126 [Ashby]. 2 See in particular Health Services and Support – Facilities Subsector Bargaining Assn v British Columbia, 2007 SCC 27, [2007] 2 SCR 391 [BC Health Services]; Ontario (AG) v Fraser, 2011 SCC 20, [2011] 2 SCR 3 [Fraser]; Mounted Police Assn of Ontario v Canada, 2015 SCC 1, [2015] 1 SCR 3 [MPAO]; Meredith v Canada, 2015 SCC 2, [2015] 1 SCR [Meredith]; and Saskatchewan Federation of Labour v Saskatchewan, 2015 SCC 4, [2015] 1 SCR 245 [SFL]. 3 Canadian Charter of Rights and Freedoms, Part I of the Constitution Act, 1982, being Schedule B to the Canada Act 1982 (UK), 1982, c 11 [Charter]. 4 Ibid at s 32(1). See also RWDSU v Dolphin Delivery Ltd, [1986] 2 SCR 573, 33 DLR (4th) 174 and Hill v Church of Scientology of Toronto, [1995] 2 SCR 1130 at para 95, 126 DLR (4th) 129. 5 BC Health Services, supra note 2 at para 88. 6 Though the constitutional protections would seem to indirectly protect private sector workers by preventing governments from eliminating existing statutory protections of section 2(d) rights; or even, in some cases, by placing a positive obligation upon governments to provide statutory protections to those 254 UNBLJ RD UN-B [VOL/TOME 68 words, the Charter does not itself require private sector employers to collectively bargain with their employees; it does not relieve workers or unions of liability in tort for strike activity;7 it does not even render collective agreements enforceable as contracts at common law.8 Such rights and obligations arise under labour relations legislation, not from the Charter directly. That said, freedom of association is also protected under human rights legislation.9 In some cases, application of the freedom is specifically limited to review of other legislation; the human right statute’s purpose is to ensure that the rights and freedoms protected therein are respected within other legislation. 10 More germane to this article, however, is where a province or territory has included freedom of association within its human rights code. Such inclusion is admittedly uncommon – only Yukon, Saskatchewan, and Québec have done so to date.11 However such inclusion, when coupled with the Supreme Court’s purposive approach to freedom of association, arguably places the same obligations that are currently upon public employers – a duty to collectively bargain, a duty not to retaliate against employees who strike – directly upon private employers.12 And this may be true even – or especially – for those workers who are not “unionized” 13 2017 CanLIIDocs 175 currently excluded from, for instance, trade union legislation: see Dunmore v Ontario, 2001 SCC 94, [2001] 3 SCR 1016 [Dunmore]. 7 While strikes are no longer forbidden under the common law (Canadian Pacific Railway Co v Zambri, [1962] SCR 609, 34 DLR (2d) 654, strikers could be and can still be liable for numerous torts. For a general and still relevant review, see IM Christie’s seminal The Liability of Strikers in the Law of Tort: A Comparative Study of the Law in England and Canada (Kingston: Queen’s University Press, 1967) and Harry Arthurs, “Tort Liability for Strikes in Canada; Some Problems in Judicial Workmanship” (1960) 38 Can Bar Rev 346. See also RWDSU Local 558 v Pepsi-Cola Canada Beverages (West) Ltd, [2002] 1 SCR 156 at para 73, 208 DLR (4th) 385 (continuing applicability of tort law in restraining picketing). 8 Young v CNR, [1931] 1 DLR 645, [1931] AC 83 [Young], (collective agreements are not enforceable at common law). While unions have often been granted “personhood” under labour legislation (for example, s 6-3 of The Saskatchewan Employment Act, SS 2013, c S-15.1 (SEA), states that “for the purposes of this Act, every union is deemed to be a person”), such status is often limited to the scope of the statute in question. And while unions also have sufficient legal status to, for instance, be found liable for contempt, as in United Nurses of Alberta v Alberta (AG), [1992] 1 SCR 901, 89 DLR (4th) 609, that does not mean that there will be the necessary privity of contract to convert individual contracts of employment to a legally binding collective agreement in the absence of a statutory declaration to that effect. 9 The Charter is of course also a “human rights” statute, and in using the term “human rights legislation” I am not suggesting that the Charter is something other than what it is. For the purposes of this article, however, I will use the term “human rights legislation” or “human rights code” to distinguish “mere” statutes from the constitutional human rights protections found in the Charter. 10 The Alberta Bill of Rights, RSA 2000, c A-14, and the Canadian Bill of Rights, SC 1960, c 44. 11 The Saskatchewan Human Rights Code, SS 1979, c S-24.1, s 6 [SHRC]; the Yukon Human Rights Act, RSY 2002, c 116, s 5 [YHRA]; The Charter of Human Rights and Freedoms, RSQ c C-12 at s 3 [Québec Charter]. 12 As I will explore further below, for the purposes of enforcement, the Yukon and Saskatchewan statutes, at least, do not distinguish between fundamental freedoms and anti-discrimination provisions. 13 I recognize here that Professor Roy Adams, for one, takes issue with the Canadian tendency to equate “represented by a certified bargaining agent” to “unionized”, as there are other ways to “unionize” outside of the Wagner Act model of labour relations: Roy Adams, “Fraser v Ontario and International Human 2017] THE RIGHT TOOL FOR THE JOB? 255 under the relevant labour relations legislation, and who therefore cannot take advantage of the statutory protections that currently exist for workers who are covered, or who are seeking to be covered, by that legislation. This presents a conundrum. On the one hand, if freedom of association is guaranteed within a provincial statute, why cannot workers (or indeed anyone) exercise that freedom on the terms set out by the Supreme Court under the Charter? On the other hand, is it a given that the two freedoms of association are equivalent? And might enforcement of a provincial freedom of association undermine or destabilize the existing labour relations legislation that the Supreme Court has taken pains to reinforce in its section 2(d) jurisprudence? What I explore in this article, therefore, is the potential of provincial human rights codes in promoting and protecting freedom of association and labour rights for non-unionized workers.14 I am less concerned with the use of freedom of association to challenge legislative provisions; that can also be done using the Charter, and while it may be arguable that at least some elements of Canadian labour relations 15 regimes are unconstitutional, that is not my focus here, though that aspect of the 2017 CanLIIDocs 175 inquiry will become more relevant when dealing with the right to strike. It is hardly self-evident, however, that human rights commissions are a possible (or appropriate) venue to pursue associational rights. There are numerous issues that must be addressed in making this argument. The issues I will address therefore are, in order: first, whether human rights commissions can, or should, be used to enforce freedom of association; second, whether freedom of association under human rights codes is the equivalent to freedom of association under the Charter; and third, what freedom of association under human rights codes might look like in application. In terms of application, I will examine the less controversial applications – freedom of association as an aspirational principle, as a principle used in interpreting other legislation, and as a complementary remedy within labour arbitrations – and then the more controversial – specifically the duty to recognize Rights: A Comment” (2009) 14 CLELJ 378 at 387 and “A Pernicious Euphoria: 50 years of Wagnerism in Canada” (1995) 3 CLELJ 321. However, for convenience’s sake, I will uphold the general practice and refer to workers who are represented by a certified bargaining agent as “unionized”, for the most part, and those who are not as “non-unionized”. There is some ambiguity here, as workers who are represented by an exclusive bargaining agent that is uncertified, but voluntarily recognized by the employer, would also usually be seen as “unionized”; where such distinctions become relevant I will be more precise in my nomenclature. 14 As discussed later in the paper, where workers fall under the existing protections of labour relations legislation, the necessity and desirability of human rights commissions entertaining associational rights complaints are much reduced. 15 For instance, I have previously argued that the statutory restrictions on sympathetic action, such as workers refusing to cross another union’s picket line, are unconstitutional: Keir Vallance, “’Lest You Undermine Our Struggle: Sympathetic Action and the Canadian Charter of Rights and Freedoms” (2015) 53:1 Alta L Rev 139 [Vallance, “Lest You Undermine”]. 256 UNBLJ RD UN-B [VOL/TOME 68 employee associations (especially those without majority support within the workplace), the duty to bargain in good faith, and the right to strike. I should add that nothing I say here should be taken to suggest that governments cannot or should not amend existing labour relations legislation to expand the range of protections or the range of workers who are covered. Of course, governments can do so; and perhaps should, given the limited success of the existing regime in promoting collective bargaining.16 For that matter, governments in Saskatchewan, Yukon, and Québec could remove freedom of association from their human rights codes or restrict the application of fundamental freedoms to review of other legislation. However, in the absence of legislative will one way or the other, it may be that the courts (or as I will argue, human rights commissions) may provide an alternative avenue for labour rights.17 And at a more basic level, my focus is, simply put, on the implications that arise from the existing legislative guarantees of freedom of association. We already have the right; what, then, is our remedy?18 A final introductory point: While the focus of this paper is labour law, the interaction between Charter rights and provincial human rights codes, and labour 2017 CanLIIDocs 175 rights and human rights, may be of more general interest. In the vein of this issue of the University of New Brunswick Law Journal, the issues raised herein have constitutional and administrative law implications. The interaction between constitutional Charter freedoms and the equivalent quasi-constitutional freedoms within human rights legislation is not limited to freedom of association; though association is perhaps the most complicated in application, given the positive obligations inherent in the section 2(d) jurisprudence. The use of human rights commissions to enforce fundamental freedoms also raises significant administrative law implications, with an administrative tribunal potentially being asked to deal with issues outside of its institutional competence; and with two competing tribunals (human rights commissions and labour relations boards) potentially dealing with identical or similar issues. 16 In 2011, for example, just over 31 per cent of Canadian employees were covered by a collective agreement, but this number is misleading because while almost 75 per cent of public sector workers were covered, only 17.5% of private sector workers were: Canada, Statistics Canada, Unionization 2011, by Sharanjit Uppal, Catalogue no. 75-001-x (Ottawa: Statistics Canada, 2011), online: StatsCan 17 In a similar vein, see Roy Adams, “Bringing Canada’s Wagner Act Regime into Compliance with International Human Rights Law and the Charter” (2016) 19:2 CLELJ 365 [Adams, “Compliance”]. 18 As stated by McLachlin J (as she then was) in Watkins v Olafson, [1989] 2 SCR 750 at 767, 61 DLR (4th) 577, paraphrasing Ashby, supra note 1: “Where there is a right, there must also be a remedy.” 2017] THE RIGHT TOOL FOR THE JOB? 257 2. Non-Wagnerist Labour Rights for the Non-Unionized Canadian labour relations legislation is based on the American “Wagner Act”.19 Union representation of workers under Wagner Act-style (or “Wagnerist”) legislation is premised on two key principles: majoritarianism and exclusivity. If workers wish to collectively bargain with their employer, they must first become members of a bargaining unit certified by the relevant Labour Relations Board. Within the proposed bargaining unit, a majority of the workers must vote to be represented by the applicant union – majoritarianism – and then only the applicant union may represent those workers – exclusivity – unless and until the union is displaced by another union or the workers choose to decertify. If the union is successful, then it becomes the bargaining agent for all workers in the bargaining unit and the employer and union both have a statutory duty to bargain in good faith and make efforts to conclude a collective agreement. 20 If the applicant union is unsuccessful in becoming the bargaining agent for the bargaining unit, however – if it fails to obtain majority support within the unit – then the employer is under no legal obligation whatsoever to collectively bargain with either its employees, or any union. 2017 CanLIIDocs 175 Canadian labour relations legislation is also characterized by strict control of strike activity. Workers who are not represented by a certified bargaining agent cannot, generally speaking, strike. 21 Workers who undertake such action will be seen to have breached, or even repudiated, their contracts of employment.22 And workers who are represented by a certified bargaining agent are governed by the so-called “peace obligation” which prohibits any strike during the term of a collective agreement.23 Where workers historically used the strike to force an employer to recognize and bargain with their union, they now use the certification process and 19 National Labor Relations Act, 29 USC §§ 151-169 (Wagner Act). I will also use the terms “Wagnerist” or “Wagnerism” when discussing labour relations statutes or principles based on the Wagner Act. 20 See e.g. the SEA, supra note 8 at s 6-7 (duty to bargain in good faith) and 6-62(1)(d) and 6-63(1)(c) (unfair labour practice for employers or unions, respectively, to fail to bargain in good faith); Canada Labour Code, RSC 1985, c L-2, Part 1, Div IV. 21 For example, s 6-63(1)(d) of the SEA sets out that it is an unfair labour practice for an employee to, inter alia, take part in a strike unless a “strike vote is taken, and a majority of employees who vote do vote in favour of a strike”, while s 6-32 provides that “no employee shall strike before a vote has been taken by the employees in the bargaining unit affected and the majority of those employees who vote have voted for a strike”. The prohibition here is both explicit (no employee shall strike) and implicit (by definition, unorganized workers do not have a “bargaining unit” as defined under the SEA). The interaction between such provisions and situations where an employer has voluntarily recognized an uncertified bargaining agent is not always clear. See also Roy Adams, Industrial Relations Under Liberal Democracy (Columbia: University of South Carolina Press, 1995) at 89; Peter Barnacle, Roderick Wood, Geoffrey England & Innis Christie, Employment Law in Canada, 4th ed (Markham: LexisNexis Canada, 2005) at 11-10 22 See e.g. Reference Re Public Service Employees Relations Act (Alta), [1987] 1 SCR 313, 38 DLR (4th) 161. [Alberta Reference] at paras 176-77, per McIntyre J. 23 See e.g. SEA, s 6-30; Canada Labour Code, s 88.1. See also generally Geoffrey England, “Some Thoughts on the Peace Obligation” (1980) 12:52 Ottawa Law Rev 521. 258 UNBLJ RD UN-B [VOL/TOME 68 the statutory “duty to bargain in good faith”. Where strikes were once the means by which collective agreements were enforced,24 disputes between employers and their unionized employees are now handled through statutorily-mandated labour arbitration.25 Outside of labour relations legislation, labour rights such as collective bargaining and the right to strike are fragile. Consider non-unionized workers undertaking strike action. This could include activity such as the “fast food strikes” or the “Fight for 15”26 in the United States, where low-wage workers walked off the job en masse to militate for a higher minimum wage and more effective protection of union organizing. Or it could include less dramatic action such as Icelandic women collectively leaving work two hours early to protest the 30 per cent wage gap between women and men.27 Such action would not be protected – indeed, would be illegal – under Canadian labour relations statutes. The workers would be liable for discipline or dismissal owing to their individual breaches of contract. Consider also a situation where workers wish to collectively bargain with their employer, but they either cannot obtain the majority support necessary to certify 2017 CanLIIDocs 175 a bargaining agent under the labour relations legislation in their jurisdiction, or they do not wish to avail themselves of the rights, protections, and obligations that labour relations legislation would provide. Their employer has no duty to negotiate with them; Wagnerism is, as Professor David Doorey has described it, an “all or nothing” affair.28 A helpful approach to labour rights, and one that I will adopt in my discussion, is Professor Doorey’s characterization of labour rights as “thick” and “thin”.29 The “thinnest” right is what the Supreme Court has referred to as “constitutive” freedom of association – the right to form, belong, and maintain worker associations.30 “Thicker” rights include the right to exercise a “right of 24 See Young, supra note 8. 25 See SEA, s 6-45ff; Labour Relations Code, RSA 2000, c L-1, s 135. 26 E.g. Adam Edelman, “Thousands of airport, fast food workers across U.S. plan strike”, New York Daily News (21 November 2016), online: 27 Uri Friedman, “Why Thousands of Women in Iceland Left Work Two Hours Early This Week” The Atlantic (27 October 2016), online: 28 David Doorey, "Graduated Freedom of Association: Worker Voice Beyond the Wagner Model" (2012), 38:2 Queen's LJ. 29 Ibid. 2017] THE RIGHT TOOL FOR THE JOB? 259 association” without being “punished, terminated or interfered with” by one’s employer; the right to make “collective representations” to an employer through an association; and an obligation on an employer to “receive collective employee representations, and to engage in ‘meaningful dialogue’ and consider representations ‘in good faith’”.31 The “thickest” rights are those found under Wagnerist labour relations legislation – a right to strike (on the terms allowed under the legislation) without penalty; access to grievance arbitration, interest arbitration, and/or mediation to assist with collective bargaining; and a statutory duty to bargain in good faith.32 Some of the rights that he categorized as “thickest” (the right to strike, to full collective bargaining, and to mediation and arbitration; as well as a duty to bargain in good faith)33 have arguably now been constitutionalized in SFL (the right to strike)34 and BC Health Services and Fraser (a constitutional duty to bargain in good faith).35 To further apply Professor Doorey's categorization, my focus in this article is on the "thinner" and "thicker" labour rights – but not the "thickest" (i.e. Wagnerist) rights.36 I am not suggesting that human rights commissions be called upon to create and adjudicate a comprehensive labour code, nor that they insert themselves 37 wholesale into matters properly before a labour relations board. 2017 CanLIIDocs 175 In addressing labour rights outside of existing labour statutes, the Charter is of limited help. The Supreme Court has made it quite clear that Wagnerism is, in 30 Ibid at 533; MPAO, supra note 2 at para 52. 31 Ibid. 32 Ibid. 33 Ibid. 34 SFL, supra note 2 at paras 3, 24, 77. 35 BC Health Services, supra note 2 at para 90; Fraser, supra note 2 at paras 40ff. See also Justice Rothstein’s dissent in MPAO, supra note 2 at para 169. 36 Note however that I do consider the right to strike later in the paper. 37 Practically speaking, human rights legislation generally grants its Commission the discretion to refuse to accept a complaint where the subject matter of the complaint has been adequately dealt with elsewhere: see YHRA, s 20(1)(h) and (i): the commission shall investigate a complaint unless “… the complainant has not exhausted grievance or review procedures which are otherwise reasonably available or procedures provided for under another Act…”; or “the substance of the complaint has already been dealt with in another proceeding”; SHRC, supra note 11 at s 27(1)(d): the commission may dismiss a complaint if “the substance of the complaint has been appropriately dealt with pursuant to another Act or proceeding”. Where a worker or a group of workers has availed, or attempted to avail, themselves of Wagner Act-style legislation, they gain access to the rights, protections, and obligations of that regime. Again, such rights and protections may not extend beyond that specific statute or regime. See e.g. SEA, supra note 10, s 6- 62(1)(a) (an employer may not “interfere with, restrain, intimidate, threaten, or coerce an employee in the exercise of any right conferred by this Part” (emphasis added); similar restrictions appear under 6-62(1)(g) (discrimination in terms and conditions of employment), while the duty to bargain is limited to bargaining with a “bargaining unit” (6-62(1)(d)). The implication seems to be that labour rights outside of the Act are not so protected. Notably the Canada Labour Code’s protections (s 94) are not so limited, though strike activity is only protected insofar as it is permitted by the Code (s 94(3)(a)(vi)). 260 UNBLJ RD UN-B [VOL/TOME 68 general, constitutional.38 While exclusion of a group of workers from a Wagnerist statute, or at least from protections comparable to those within a Wagnerist statute, might be unconstitutional,39 the Court has shown little appetite to overturn the Wagnerist model. But the Charter is not entirely without application. If an employer were to pursue an unfair labour practice against non-unionized employees who went on strike, an argument could be made that the prohibition on strikes is unconstitutional.40 It is more likely, however, that an employer would simply either dismiss striking employees, or accept their repudiation of their contracts of employment, rather than pursue an ULP – though it is perhaps also arguable that the values underlying the Charter41 would dictate that such strikes would not be considered breaches, or at least not repudiation, of contract by employees.42 I would argue, however, that freedom of association under human rights legislation might provide meaningful protection of labour rights. In particular, in relying upon a human rights code, a claimant is not dependent upon “Charter values”; the human rights protection will apply to private employers and to the common law directly. The procedural protections under human rights legislation may also provide greater access to labour rights for those workers not represented by an 2017 CanLIIDocs 175 established union. And if nothing else, placing freedom of association within human rights legislation can serve as an aspirational statement about the value of labour rights.43 38 See MPAO, supra note 2 at paras 94–95 (Wagner Act model is one possible model); SFL, supra note 2 at para 101-102 (legislative amendments making it more difficult to unionize under Wagner Act model nonetheless constitutional). 39 See MPAO, supra note 2; Dunmore, supra note 6. 40 Though equally arguably justified under section 1. See Brian Etherington, “The Right to Strike Under the Charter after Saskatchewan Federation of Labour: Applying the New Standard to Existing Regulation of Strike Activity” (2016) 19 CLELJ 492. 41 See e.g. Hill, supra note 4 at para 91. 42 See again Vallance, “Lest you Undermine”, supra note 15. 43 That said, Professor Adams made a similar argument over ten years ago about the aspirational/educational value of including freedom of association in human rights legislation: Adams, Labour Left Out, supra note 16 at 39–40. The argument has merit, but I cannot help but note that Saskatchewan has had freedom of association in its human rights legislation since 1947 (The Saskatchewan Bill of Rights Act, 1947, SS 1947, c 35, s 5), with little or no observable impact on the state of the law in that province. Perhaps the newly-expanded scope of freedom of association that has now been recognized by the Supreme Court will revitalize this argument. 2017] THE RIGHT TOOL FOR THE JOB? 261 3. Human Rights Commissions and Fundamental Freedoms i. Human Rights Commissions as an Appropriate Venue It may at first blush seem absurd to suggest that workers pursue freedom of association complaints before human rights commissions and tribunals.44 For reasons that follow, I suggest the concept is not as far-fetched as it may appear and, indeed, the Supreme Court's freedom of association jurisprudence may require us to consider the possibility. At the most basic level, freedom of association does, after all, appear within the human rights codes in question. That suggests on the face of it that it should be taken as seriously as other provisions within the statute. Further, the Supreme Court has again invited exploration of these issues in its administrative law jurisprudence. The traditional hostility of courts to the ability of human rights commissions and tribunals45 to interpret their “home statutes” has faded, and a more deferential standard of review regarding decisions of administrative tribunals has become the norm.46 Commensurate with that relaxed standard is, perhaps, a greater ability of commissions and tribunals to engage with and interpret previously 2017 CanLIIDocs 175 unexplored territory within the wording of their statutes, particularly when it is serving to extend the reach of a right also protected under the Charter.47 I will note here also that as far as enforcement is concerned, the Yukon and Saskatchewan statutes do not distinguish between bill of rights portions and non- discrimination portions of the statute. There is therefore no statutory bar to proceeding with a human rights complaint, including investigation by the Commission or subsequent litigation, regarding violation or infringement of a 44 I will note at the outset that in Yukon, Saskatchewan, and Québec, it falls upon the human rights commission to accept, investigate, mediate, and carry forward human rights complaints. However, in Saskatchewan, it now falls upon the Saskatchewan Court of Queen’s Bench, not a human rights tribunal, to adjudicate any complaints that proceed to a hearing: SHRC, supra note 11 at s 29.6. Yukon and Québec retain administrative tribunals – a board of adjudication and a human rights tribunal, respectively – as the adjudicative body: see YHRA, supra note 11 at s 22 and Québec Charter, supra note 11 at s 78-82. 45 Canada (CHRC) v Canada (AG), [2011] 3 SCR 471 [Mowat] at paras 19-20. 46 E.g. Edmonton (City) v Edmonton East (Capilano) Shopping Centres Ltd, 2016 SCC 47 at para 22, [2016] 2 SCR 293. See also McLean v British Columbia (Securities Commission), 2013 SCC 67, [2013] 3 SCR 895; Alberta (Information and Privacy Commissioner) v Alberta Teachers’ Association, 2011 SCC 61, [2011] 3 SCR 654; Dunsmuir v New Brunswick, 2008 SCC 9, [2008] 1 SCR 190 [Dunsmuir]; R v Conway, 2010 SCC 22 at paras 81-84, [2010] 1 SCR 765 [Conway] (administrative tribunals’ ability to apply the Charter). See also Paul Daly, “Canada’s Bipolar Administrative Law: Time for Fusion” (2014) 40:1 Queen’s LJ 213 at 218. 47 Indeed, I would suggest that it would be an error if a human rights commission were to refuse to take on a freedom of association complaint based only on the fact that workers could pursue certification under another Act. After all, the Supreme Court has stated that the Wagner Act model has not been constitutionalized. To interpret freedom of association to mean simply compliance with a Wagner Act– style labour relations statute seems not merely to limit but outright ignore a right given by a quasi- constitutional statute. 262 UNBLJ RD UN-B [VOL/TOME 68 person’s freedom of association.48 The Québec Charter differs in that its investigative functions seem limited to the statute’s anti-discrimination provisions,49 and human rights complaints must be within the “sphere of investigation of the commission.”50 The Québec Charter does, however, have a “catch-all” investigative provision which allows the Commission to investigate “any act of reprisal or attempted reprisals and into any other act or omission which, in the opinion of the commission, constitutes an offence under this Charter…”51 Nonetheless, the Québec commission’s mandate to accept and investigate human rights complaints regarding freedom of association is more ambiguous than under the Yukon and Saskatchewan statutes. It should be noted, though, that the Québec Charter does provide the Commission with the ability to identify where legislation has violated the Québec Charter, though its power is only to recommend changes to the government.52 Both the YHRA and the SHRC provide that each Act takes precedence over other legislation.53 In the SHRC, statutory provisions that are inconsistent with the Code are “inoperative”, while under the YHRA the Act “supersedes” other legislation. Neither statute expressly provides the power to strike down legislation. However, the 2017 CanLIIDocs 175 Charter can be used for that purpose insofar as freedom of association is statutorily infringed. The Supreme Court’s constitutional and administrative law jurisprudence invite exploration. But furthermore, the purposes of the various statutes, regimes, and areas of law at play here – the Charter, human rights legislation, and labour law - are also eminently compatible. The purposes of both freedom of association and of human rights legislation are arguably furthered by enforcement of labour rights under human rights statutes. The Supreme Court has confirmed that freedom of association should be interpreted generously and purposively;54 that it is “aimed at reducing social imbalances, not enhancing them;”55 that it “prevents individuals from being overwhelmed by the powerful while at the same time providing collective strength.”56 While s. 2(d)’s purpose may be to protect individuals from “state- enforced isolation”,57 freedom of association under human rights legislation is not so 48 YHRA, supra note 11, s 20 (“a contravention of this Act”); SHRC, supra note 11 at s 27 (“contravened a provision of this Act”). 49 Québec Charter, supra note 11 at s 71(1). 50 Ibid at s 74. 51 Ibid at s 71(9). 52 Ibid at s 71(6) 53 SHRC, supra note 11 at s 44; YHRA, supra note 11 at s 39. 54 MPAO, supra note 2, para 5. 55 Ibid at para 59 56 Ibid at para 70. 57 Ibid at 58, citing the Alberta Reference at 365 [emphasis added]. 2017] THE RIGHT TOOL FOR THE JOB? 263 restricted. Using human rights codes to extend the reach of a now-recognized constitutional right seems compatible with, albeit not a direct application of, the Supreme Court’s expansion of administrative tribunals’ ability to apply Charter remedies.58 Similarly, human rights legislation has a special, quasi-constitutional59 status and should be interpreted liberally and purposively.60 And certainly a number of the types of complaints that could involve freedom of association, at least, are in essence discrimination cases (i.e. reprisal for union involvement or union or strike activity). There seems no reason why the Commission’s expertise could not be turned to preventing discrimination on the basis of union activity; and indeed the existence of freedom of association within the Code suggests that workers have a right to freedom from anti-union discrimination – “right of non-reprisal”, in Professor Doorey’s words.61 When applying freedom of association to labour law, it is also important to consider the underlying purposes of labour law. Professor Doorey has framed the 62 goal of labour law as the “facilitation of employee voice,” while Professor Beth 2017 CanLIIDocs 175 Bilson has suggested that worker participation and self-determination are rights in themselves, not merely means to an end.63 The value of freedom of association in the workplace was recognized by the majority in MPAO, citing Dickson CJ in the Alberta Reference – that association was of most value to those who were “liable to be prejudiced by the actions of some larger and more powerful entity, like the government or an employer.”64 In the Alberta Reference, Dickson CJ also noted that collective bargaining enhanced the “human dignity, liberty and autonomy of works by giving them the opportunity to influence the establishment of workplace rules and thereby gain some control over a major aspect of their lives, namely their work…”65 Finally, the purpose and values underlying the Charter – “human dignity, equality, liberty, respect for the autonomy of the person and the enhancement of 58 See, e.g., Conway, supra note 46. 59 ICBC v Heerspink, [1982] 2 SCR 145, 137 DLR (3d) 219; Ontario Human Rights Commission v Simpson– Sears, [1985] 2 SCR 536, 52 OR (2d) 799. 60 Gould v Yukon Order of Pioneers, [1996] 1 SCR 571 at para 5, 133 DLR (4th) 449 [Gould] (“a fair, large and liberal interpretation with a view to advancing its objects”); Ruth Sullivan, Sullivan on the Construction of Statutes, 5th ed (Markham: LexisNexis Canada Inc, 2008) at 497. 61 Doorey, supra note 28 at 528–529. 62 Ibid at 520. See also MPAO, supra note 2 at paras 55, 87. 63 Beth Bilson, “Future Tense: Some Thoughts About Labour Law Reform” (2005) 12 CLELJ 233 at 256, 259. 64 MPAO, supra note 2 at para 57. 65 Alberta Reference, supra note 22 at para 82. 264 UNBLJ RD UN-B [VOL/TOME 68 democracy”66 – are close to the stated purposes of human rights legislation. The SHRC, for example, provides that the Code is meant to “promote recognition of the inherent dignity and the equal inalienable rights of all members of the human family” and to “further public policy in Saskatchewan that every person is free and equal in dignity and rights and to discourage and eliminate discrimination”.67 To this I will add one more related point: access to justice. Unlike Charter litigation, human rights legislation places emphasis on conciliation and settlement prior to recourse to an adversarial hearing process.68 There is a greater potential, perhaps for parties to mediate or otherwise resolve their dispute within the human rights framework - the process is arguably less adversarial than that within the Wagnerist model of labour relations. If the goal of labour rights is employee "voice", then resolution by the parties themselves would continue to be a priority within the human rights process (as it ostensibly is within labour relations legislation). Human rights commissions potentially have a greater emphasis on the good of society generally – the social good – rather than merely justice between the parties – workplace justice.69 This, at least in theory, places greater emphasis on social relationships and harmony over individualistic conflict. It is when a dispute is “closed to conciliation and mediation” that “human rights law and procedures [enter] the realm of the semijudicial.”70 In 2017 CanLIIDocs 175 situations where individuals are attempting to exercise their labour rights, possibly without the assistance of an established union and the accompanying resources and expertise, such access to justice concerns seem relevant. At the same time, the carriage of a complaint by a human rights commission may by its very nature disempower or delegitimize employee voice. A human rights Commission may have a final say on whether a settlement negotiated between the parties is acceptable71 and may refuse to proceed with a complaint if a complainant has refused an offer of settlement that the Commission views as fair.72 Employee voice or employee interests may be subordinated to the Commission's view of what is fair, or what is in the broader interest. While the means of enforcement of "thinner" labour rights may be available under human rights legislation, there is by no means a guarantee that they will be exercised in a manner that respects or promotes the principle (per Professor Bilson) of "self-determination" which, at least ostensibly, underlies collective bargaining under labour relations legislation. 66 Ibid at para 81. 67 SHRC, supra note 11 at s 3. 68 Brian R Howe & David Johnson, Restraining Equality: Human Rights Commissions in Canada (Toronto: University of Toronto Press, 2000) at 11, 43; see SHRC, supra note 11 at s 28. See also Scowby v Glindenning, 148 DLR (3d) 55, 1983 CanLII 2065 (SKCA) (rev'd on other grounds [1986] 2 SCR 226) at paras 19, 22 (emphasis on amicable resolution). 69 However, this also raises the uncomfortable question of to what extent labour rights should be subordinate to the "greater good," though Wagnerism already unquestionably does so in many ways (e.g. emphasizing industrial peace via the peace obligation). 70 Ibid at 46. 71 SHRC, supra note 11 at s 28(2.1). 72 Ibid at s 29.5(2); YHRA, supra note 11 at s 20(1)(g). 2017] THE RIGHT TOOL FOR THE JOB? 265 ii. The Risks of “Labour Rights as Human Rights” In following this line of reasoning, it is also important to note that pursuing freedom of association under human rights codes – of characterizing labour rights as human rights – is not free of risk or controversy. The arguments invoke the long- standing debate over labour rights as human rights; an ongoing debate among those sympathetic to unions since the advent of the Charter – one between, in Professor Brian Etherington's terms, Charter “romantics,” “realists,” and “pragmatists.”73 There are certainly those who have seen Charter rights (and the use of the language of human rights generally) as a welcome addition to the labour field74 and caution that exclusion from constitutional discourse leaves labour vulnerable to government action,75 or who at least view a human-rights based rhetoric as having potential.76 But there are also ample and valid criticisms of viewing labour rights through the lens of human rights discourse. There has been and is significant hostility within the labour relations community to rights- or Charter-based analysis of labour law. Such hostility is perfectly understandable, whether it is based on distrust of the judiciary77 after a “century of bruising encounters with judges determined to manipulate the 2017 CanLIIDocs 175 73 Brian Etherington, “An Assessment of Judicial Review of Labour Laws Under the Charter: Of Realists, Romantics and Pragmatists” (1992) 24 Ottawa L Rev 685. 74 E.g. Roy Adams, “The Revolutionary Potential of Dunmore” (2003), 10 CLELJ 117 [Adams, “Dunmore”], Roy Adams, “From Statutory Right to Human Right: The Evolution and Current Status of Collective Bargaining” (2008) 12 Just Labour 48 [Adams, “From Statutory Right to Human Right”], and Adams, Labour Left Out, supra note 16; Ken Norman, “What’s Right is Right: The Supreme Court Gets It” (2008) 12 Just Labour 16; Guy Davidov, “Judicial Development of Labour Rights – Contextually” (2010) 15(2) CLELJ 235; David Beatty, Putting the Charter to Work: Designing a Constitutional Labour Code (Kingston: McGill Queen’s University Press, 1987); Canadian Foundation for Labour Rights website, online: 75 E.g. David Beatty, “Labouring Outside the Charter” (1991) 29 Osgoode Hall LJ 839; Alan Hyde, “Exclusion is Forever: How Keeping Labour Rights Separate from Constitutional Rights Has Proven to Be a Bad Deal for American Trade Unions and Constitutional Law” (2010) 15(2) CLELJ 251. 76 Judy Fudge, “Labour Rights as Human Rights: Turning Slogans into Legal Claims” (2014) 37 Dal LJ 601; Bradley Walchuk, “The Best of Both Worlds: A Pragmatic Approach to the Construction of Labour Rights as Human Rights” (2009) 14 Just Labour 75 [Fudge, “Labour Rights as Human Rights”]. Both Fudge and Walchuk have also expressed skepticism, or at least caution, about the value of rights discourse or rights litigation in the labour field: Judy Fudge, “Labour, Courts, and the Cunning of History” (2010) 16 Just Labour 1 and Judy Fudge, “The Supreme Court of Canada and the Right to Bargain Collectively: The Implications of the Health Services and Support case in Canada and Beyond” (2008) 37 Indus LJ 25; Bradley Walchuk, “Union Democracy and Labour Rights: A Cautionary Tale” (2011) 2:2 Global Labour J 106. 77 E.g. Eric Tucker, “The Constitutional Right to Bargain Collectively: The Ironies of Labour History in the Supreme Court of Canada” (2008) 61 Labour/Le Travail 151; Judy Fudge; Michael Mandel, The Charter of Rights and the Legalization of Politics in Canada (Toronto: Thomson Educational Publishing, Inc, 1994) at ch 5; Harry Arthurs, “‘The Right to Golf’: Reflections on the Future of Workers, Unions and the Rest of Us Under the Charter” (Labour Law Under the Charter, delivered at the School of Industrial Relations and Faculty of Law, Queen’s University, 24-26 September 1987) [Labour Law Under the Charter]. 266 UNBLJ RD UN-B [VOL/TOME 68 common law and legislation to extinguish rights of workers”;78 scepticism over the efficacy or practicality of rights litigation as a means of protecting labour rights;79 or concerns regarding rights-based rhetoric or legalism in labour relations or more generally.80 While the Supreme Court has recognized, in what could almost be described as a throw-away line, that the Charter protects collective rights as well as individual rights,81 that cannot be a full answer to the skeptics. A more substantive response, at least within the scope of this paper, may be found in the nature of human rights commissions, which I canvassed earlier. Human rights complaints are relatively inexpensive to pursue, compared to constitutional litigation. And the ostensibly less adversarial nature of processes under human rights legislation may mean that labour rights under human rights legislation would not necessarily fall into a rights model that is individualistic, atomistic, or inherently corrosive to collective labour rights. But, on the other hand, the guiding (or, less charitably, paternalistic) hand of the Commission in its carriage of complaints may also serve to stifle class consciousness and radicalism, just as (in some skeptics’ view) does the Charter. 2017 CanLIIDocs 175 I am not convinced that the skeptics are entirely wrong. Just as with rights discourse, applying human rights legislation to labour rights has its risks. While, in 78 Sandra Freedman, “Scepticism Under Scrutiny: Labour Law and Human Rights” in Tom Campbell et al, eds, Sceptical Essays on Human Rights (Oxford: Oxford University Press, 2002) 197. 79 E.g. Harry Arthurs, “The Constitutionalization of Employment Relations: Multiple Models, Pernicious Problems” (2010) 19 Soc & Leg Stud 43, and Harry Arthurs “Constitutionalizing the Right of Workers to Organize, Bargain and Strike: The Sight of One Shoulder Shrugging” (2010) 15(2) CLELJ 373 [Arthurs, “Constitutionalizing the Right of Workers”]; Bob Hepple, “The Right to Strike in an International Context” (2010) 15(2) CLELJ 133; Harry Arthurs & Brent Arnold, “Does the Charter Matter?” (2005) 11 Rev Const Stud 37; Bryan Palmer, “What’s Law Got to Do With It?: Historical Considerations on Class Struggle, Boundaries of Constraint, and Capitalist Authority” (2003) 41 Osgoode Hall LJ 465; Paul JJ Cavalluzzo, “Freedom of Association – its Effect upon Collective Bargaining and Trade Unions” in Labour Law Under the Charter, supra note 77 at 267 [Freedom of Association]; Charles W Smith, “Labour, Courts and the Erosion of Workers' Rights in Canada” in Stephanie Ross & Larry Savage, eds, Rethinking the Politics of Labour in Canada (Winnipeg: Fernwood Publishing, 2012) 184. See also Larry Savage & Charles W Smith, Unions in Court: Organized Labour and the Charter of Rights and Freedoms (Vancouver: UBC Press, 2017). For an interesting view on the interaction between human rights and labour rights, specifically regarding the agendas and perspectives of organizations and activists within the labour movement and human rights movement, see Kevin Kolben, “Labor Rights as Human Rights?” (2009–2010) 50 Va J Int’l L 449. 80 E.g. Palmer, supra note 79; Susannah Quail, “Labour Rights and Labour Politics under the Charter (2013-2014), 45 Ottawa L Rev 343 (specifically regarding the “de-radicalizing” effect of Charter litigation at p 359) Eric Tucker, “Labor's Many Constitutions (And Capital's Too)” (2012) 33 Comp Lab L & Pol'y J 355; Harry Arthurs, “Labour Law Without the State?” (1996) 46 UTLJ 1 at 40 (legalistic approaches as “corrosive” to collective bargaining) and Harry Arthurs, “The New Economy and the New Legality: Industrial Citizenship and the Future of Labour Arbitration” (1999) 7 CLELJ 45; Larry Savage, “Labour Rights as Human Rights? A Response to Roy Adams” (2008) 12 Just Labour 68. More generally, see Allan Hutchinson and Andrew Petter, “Private Rights/Public Wrongs: The Liberal Lie of the Charter” (1988) 38 UTLJ 278; Allan Hutchinson, Dwelling on the Threshold: Critical Essays on Modern Legal Thought (Toronto: Carswell, 1988) at 14 and 21 (the power of rights discourse and the risks of being co- opted thereby). 81 MPAO, supra note 2 at paras 64–65. 2017] THE RIGHT TOOL FOR THE JOB? 267 the end, I keep returning to the old maxim in Ashby – the right already exists, and I am simply exploring the remedy – this is not a complete answer either (after all, s. 2(d) plainly exists and the Charter debate remains). At the same time, the potential of human rights legislation to over-run or corrode labour rights seems much less than the potential of the Charter, for instance, given the existence and prevalence of Wagnerist labour legislation. That is, regardless of how interesting the legal questions may be, a freedom of association guarantee within human rights legislation will likely to be seen as a "poor cousin" of more robust Wagner Act model protections. However, depending on the content of freedom of association, a human rights guarantee may even strengthen non-legalistic approaches. For instance, if freedom of association protects the right to strike, it may provide for greater use of the strike outside of labour relations legislation.82 In the end, I would argue that “labour rights as human rights” in the context of human rights codes might be less problematic than within rights discourse generally, but like rights discourse remains one of many options and alternatives open to workers and to labour. 4. What is Freedom of Association? 2017 CanLIIDocs 175 Of course, much of the foregoing depends on exactly what freedom of association means. The definition of “freedom of association” provided by the Supreme Court now provides a relatively expansive web of rights and obligations. This new formulation has not been without criticism.83 However, for the purposes of this article I intend to take the Supreme Court at its word, and apply freedom of association as the Supreme Court has framed it. But it does not automatically follow that freedom of association under human rights codes will include the same rights as freedom of association under the Charter. Freedom of association under provincial codes may provide the full range of rights and protections that we see under s. 2(d); or only some of them; or may provide different rights and protections entirely. While I will argue that freedom of association under s. 2(d) and under provincial human rights legislation are more or less synonymous, that conclusion is not a given and in some ways, treating freedom of association as a single unified concept creates greater challenges that must be addressed. 82 This is itself problematic as it may undermine the existing labour relations regime, including the peace obligation. This will be examined further below. 83 I do not intend to engage in a fulsome review or critique of the Supreme Court's section 2(d) jurisprudence; that has been done elsewhere. Among many others, see Brian Langille, “The Condescending Constitution (or, The Purpose of Freedom of Association is Freedom of Association)” (2016) 19 CLELJ 335 [Langille, “Condescending Constitution”]; Brian Langille, “The Freedom of Association Mess: How We Got into It and How We Can Get out of It” (2009) 54 McGill LJ 177 [Langille, “Freedom of Association Mess”]; Brian Etherington, “The Right to Strike Under the Charter after Saskatchewan Federation of Labour: Applying the New Standard to Existing Regulation of Strike Activity” (2016) 19 CLELJ 492; Keir Vallance, “Developments in Labour Law – Freedom of Association and Labour Law at the Supreme Court: The 2014-2015 Term” (2016) 72 SCLR 227; Arthurs, “Constitutionalizing the Right of Workers”, supra note 79. 268 UNBLJ RD UN-B [VOL/TOME 68 i. Freedom of Association under s. 2(d) Freedom of association most obviously includes the right to form, join, and maintain associations - the "thinnest" labour rights. The Supreme Court recognized this even in the notorious Alberta Reference.84 The Court has characterized this approach as the “constitutive” approach to freedom of association, and while it continues to be protected by section 2(d), the “derivative” approach (associational activity that “specifically relates to other constitutional freedoms”) and the “purposive” approach (which looks at the overall purpose of the constitutional guarantee) are also protected under the Charter.85 Indeed the Court has now fully endorsed the “generous and purposive” approach to freedom of association.86 The arguments put forward in this paper are not, of course, inherently tied to ‘state-enforced isolation’ or other criteria put forward by the Supreme Court that are clearly aimed at governmental action. But as referenced previously, the Court also noted that a purpose of freedom of association is to prevent individuals from being overwhelmed by the powerful (such as a government or employer) and that it is aimed at “reducing social imbalances”. The reach of freedom of association in 2017 CanLIIDocs 175 principle seems to extend beyond governmental action, even though the Charter itself may not. Within the labour context specifically, the heart of the “new” freedom of association is the right to “meaningful collective bargaining.”87 It is from this right that other labour rights appear to emanate. In BC Health Services, it was defined “the ability of workers to engage in associational activities, and their capacity to act in common to reach shared goals related to workplace issues and terms of employment. In brief, the protected activity might be described as employees banding together to achieve particular work-related objectives.”88 In Fraser, it was described as “meaningful association in pursuit of workplace goals.”89 In MPAO, the Court further defined “meaningful collective bargaining” to mean “a process that gives employees meaningful input into the selection of their collective goals, and a degree of independence from management sufficient to allow members to control the activities of the association, having regard to the industry and workplace in question.”90 The right to “meaningful collective bargaining” contains numerous sub- rights. From MPAO, we know that it includes the right of workers to be represented 84 Alberta Reference, supra note 22 at para 143 per Le Dain J. 85 MPAO, supra note 2 at paras 52-56. 86 Ibid at para 46. 87 MPAO, supra note 2 at para 5. 88 BC Health Services, supra note 2 at para 89. 89 Fraser, supra note 2 at para 42. 90 MPAO, supra note 2 at para 99. 2017] THE RIGHT TOOL FOR THE JOB? 269 by a bargaining agent where workers have "a degree of choice and independence sufficient to enable them to determine their collective interests and meaningfully pursue them."91 In the context of MPAO, “choice” and “independence” were analyzed in relation to a government-imposed labour relations scheme that did not, in the majority’s view, adequately protect RCMP members’ freedom of association.92 This is obviously distinct from a private sector scenario; non-governmental employers do not have the luxury of using statutory muscle to impose a labour relations regime upon their employees. However, the basic principle that workers have a right to determine their collective interests and pursue them would be relevant in any workplace. The right to meaningful collective bargaining also appears to include a duty on employers (and, one assumes, workers) to bargain in good faith. In BC Health Services, the Court held that s. 2(d)’s guarantee included a requirement that “both employer and employees…meet and…bargain in good faith, in the pursuit of a common goal of peaceful and productive accommodation.”93 The majority in Fraser endorsed this duty94 – certainly to the basic level that “employers and trade unions 95 should negotiate in good faith and endeavour to reach an agreement – though the 2017 CanLIIDocs 175 value of Fraser as a precedent is admittedly questionable at this point.96 And, indeed, the Court’s ruling in Fraser ultimately upheld a labour relations statute that arguably did not include an enforceable right to bargain, but rather an obligation merely to “listen in good faith.”97 Meredith further muddied the waters in refusing to require governments to consult with the workers affected (or their bargaining agents) prior to rolling back bargained-for wage increases.98 The characterization of the duty as a “duty to listen in good faith” rather than the Wagnerist “duty to bargain in good faith” was arguably strengthened by the 91 Ibid at para 81. 92 Ibid at para 5. 93 BC Health Services, supra note 2 at para 90. See also Justice Rothstein’s dissent in MPAO at para 169. 94 Fraser, supra note 2 at paras 40ff. 95 Ibid at para 95 96 Professor Etherington has suggested MPAO has relegated Fraser to “rulings that shall be ignored”: Etherington, supra note 40 at 447. 97 Quail, supra note 80 at 358. 98 Meredith, supra note 2 at paras 25ff. The majority held that the government employer had not precluded consultation on other compensation issues and did not prevent future consultation on wages, and that the impact of the wage rollbacks were consistent with those in other government departments with which the government had consulted; therefore, the government had not infringed s. 2(d). This certainly suggests that even if there is a “duty to bargain” generally, a government need not bargain with every group of workers affected. This is in contrast with Justice Abella, who in her dissent argued that the lack of consultation with affected workers, in itself, would render the wage rollbacks unconstitutional (at paras 62ff). 270 UNBLJ RD UN-B [VOL/TOME 68 Supreme Court’s recent endorsement of Donald JA’s dissent in BCTF v BC.99 Justice Donald suggested that, at least for governments, the duty to bargain under s. 2(d) was explicitly not the Wagnerist duty to bargain in good faith100 but rather a more general duty: Parties are required to meet and engage in meaningful dialogue where positions are explained and each party reads, listens to, and considers representations made by the other party. Parties’ positions must not be inflexible and intransigent, and parties must honestly strive to find a middle ground. In order to determine whether the government is bargaining in good faith, it may sometimes be necessary to probe and consider the government’s substantive negotiating position.101 Later in BCTF, Justice Donald chided the B.C. government for not “listening in good faith” and framed the duty to listen in good faith as the “absolute minimum that is required from the employer.”102 This is arguably more general than the Wagnerist duty to bargain in good faith, where parties are explicitly legally obliged to attempt to finalize a collective 2017 CanLIIDocs 175 agreement as defined by labour relations legislation.103 Nonetheless, Fraser’s exhortation for governments and their employees to “endeavour to reach an agreement”, and BCTF’s reference to finding “middle ground”, suggest that while a collective agreement – in the sense of a legally binding agreement between the parties104 – may not be the ultimate goal under the constitutional duty to bargain, there must at least be an attempt by both (or all) sides to resolve the parties’ differences. Regardless of the characterization, however, it seems clear that a government employer has a duty to at least consider proposals from its unionized 99 British Columbia Teacher’s Federation v BC, 2016 SCC 49, [2016] 2 SCR 407, rev’g British Columbia Teachers’ Federation v British Columbia, 2015 BCCA 184 [BCTF BCCA]. Notably the Court did not fully canvass the issue, stating only that the majority was allowing the appeal “substantially for the reasons of Justice Donald.” It is therefore not certain that the Supreme Court has fully embraced Justice Donald’s reasons nor that it will ultimately give a full-throated endorsement to the “duty to listen in good faith” should it have the opportunity. 100 BCTF BCCA at paras 337–338. 101 Ibid at para 348. 102 Ibid at para 362. 103 For example, SEA, supra note 8 at s 6-1(e), defines “collective bargaining” as, inter alia, “negotiating in good faith with a view to the conclusion of a collective agreement or its renewal or revision…” But see Rothstein J’s dissent in SFL, supra note 2 at para 131, questioning whether a right to strike is necessary in light of a constitutional duty to bargain in good faith. It is not clear, however, by what means a union or workers would enforce the constitutional duty in the absence of statutory unfair labour practice protections. 104 As I will explore below, a requirement to come to a collective agreement outside of a Wagnerist labour statute is complicated by the common law refusal to recognize collective agreements as binding: see Young, supra note 8. 2017] THE RIGHT TOOL FOR THE JOB? 271 employees – it cannot simply refuse to listen and legislate the result it wishes. The application of the duty to bargain (or to listen) on non-governmental employers is perhaps less clear. Freedom of association also includes the right to strike. The strike is – to use the Supreme Court’s wording – the “powerhouse of collective bargaining”105, an “essential component of the process through which workers pursue collective workplace goals”106 – “unique and fundamental” to effective collective bargaining.107 But the right to strike’s status within s. 2(d) also remains unclear. The language used in SFL appears to relegate the right to strike as protected only insofar as that right promotes “meaningful collective bargaining.”108 However Professor Etherington, for one, argues that the Supreme Court’s language in SFL is more consistent with the right to strike as a “stand-alone right” than as a “derivative” right.109 The characterization of the right to strike may have implications on whether protest strikes, for instance, are protected strike activity. ii. Freedom of Association under Provincial Codes 2017 CanLIIDocs 175 Section 5 of the YHRA reads: Every individual and every group shall, in accordance with the law, enjoy the right to peaceable assembly with others and the right to form with others associations of any character. The SHRC reads, at s. 6: Every person and every class of persons shall enjoy the right to peaceable assembly with others and to form with others associations of any character under the law. And in the Québec Charter, at s. 3, the freedom is phrased thusly: 105 SFL, supra note 2 at para 55 106 Ibid at para 46. 107 Ibid at para 51. 108 SFL, supra note 2 at paras 24–25: Along with their right to associate, speak through a bargaining representative of their choice, and bargain collectively with their employer through their representative, the right of employees to strike is vital to protecting the meaningful process of collective bargaining within s. 2(d)…Where strike action is limited in a way that substantially interferes with a meaningful process of collective bargaining, it must be replaced by one of the meaningful dispute resolution mechanisms commonly used in labour relations. [Emphasis added.] See also Langille, “Condescending Constitution”, supra note 83 at 351. 109 Etherington, supra note 40 at 450. 272 UNBLJ RD UN-B [VOL/TOME 68 Every person is the possessor of the fundamental freedoms, including freedom of conscience, freedom of religion, freedom of opinion, freedom of expression, freedom of peaceful assembly and freedom of association. Historically, courts and tribunals have tended to treat fundamental freedoms under provincial codes as synonymous with the comparable freedoms under the Charter.110 For example, in Syndicat Northcrest111 the Supreme Court applied the same analysis to freedom of religion under the Québec Charter as it did to freedom of religion under the Charter. Freedom of expression under the Saskatchewan Code has been interpreted as being much the same freedom as that in s. 2(b) of the Charter,112 and freedom of association has been treated as a single concept regardless of whether it appears in provincial legislation or under the Charter.113 The SHRC guarantee, notably, does not reference “freedom of association,” but rather the “right...to form with others associations...”114 The wording might be more suggestive of the “constitutive” model of freedom of association. Nonetheless in Dunmore v Ontario, the right to “form and maintain associations” was held to be 2017 CanLIIDocs 175 the interest at stake in s. 2(d), and the definition of s. 2(d) has of course expanded since 2001.115 On the face of it, it would seem that s. 6 of the SHRC protects the same fundamental freedom as is found in s. 2(d) of the Charter and that similar interpretive principles would apply.116 110 The parallels between anti-discrimination provisions within provincial codes, on the one hand, and section 15 equality rights under the Charter, on the other, are not always so clearly and easily drawn. However, in that regard see e.g. British Columbia (Public Service Employee Relations Commission) v BCGSEU, [1999] 3 SCR 3, 176 DLR (4th) 1 [Meiorin] at para 17ff; Quebec (Commission des droits de la personne et des droits de la jeunesse) v Montréal (City); Quebec (Commission des droits de la personne et des droits de la jeunesse) v Boisbriand (City), [2000] 1 SCR 665 at para 34ff, 185 DLR (4th) 385 (applying s 15 jurisprudence in interpreting anti-discrimination provisions under the Québec Charter). The difference is in application (the Charter does not apply to private entities) and in exhaustive (human rights codes) versus non-exhaustive (open-ended analogous grounds in s. 15) grounds of discrimination: McIntyre J (dissenting in the result) in Law Society of British Columbia et al v Andrews et al (1989), 56 DLR (4th) 1 at 18. 111 Syndicat Northcrest v Amselem, 2004 SCC 47, [2004] 2 SCR 551. Many other cases effectively treat the various freedoms as synonymous; see for example Proulx c Québec (PG), 2015 QCCS 1042 (expression and association). 112 Whatcott v the Queen and the University of Regina, 2002 SKQB 399 at para 29, 201 CRR (2d). 113 See e.g. Gould, supra note 60; Syndicat des employées et employés professionnels et de bureau, section locale 573 (CTC-FTQ) c Commission de la construction du Québec, 2014 QCCA 368 [Syndicat des employées]; Johner’s Homestyle Catering v RWDSU Loc 568 (2 November 2010), Regina, unreported, aff'd (in this regard) 2012 SKQB 539 [Johner's]. 114 This is no doubt due to its vintage; it predated the Charter by some 35 years, the Québec Charter by over 25, and even the Universal Declaration of Human Rights, (GA Res 217A (III), UNGAOR, 3rd Sess, Supp No 13, UN Doc A/810 (1948) 71) by a year. 115 Dunmore, supra note 6 at para 17. 116 Arbitrator Norman so found in Johner’s, supra note 113. The fact that it is phrased as a “right” in the Code rather than a “freedom” would not appear to make any difference to the scope of the provision, as 2017] THE RIGHT TOOL FOR THE JOB? 273 However, while courts and tribunals have more or less treated “freedom of association” as a single concept, with much the same meaning under provincial legislation as it would have under the Charter, that is not an end to the inquiry. Courts and tribunals have not had to interpret the application of the purposive model of freedom of association in the labour context except when dealing with already- unionized employees.117 All of the cases thus far have viewed freedom of association through the lens of the Wagner Act model. As noted above, the Supreme Court in MPAO held the fundamental purpose of s. 2(d) to be “to protect the individual from ‘state-enforced isolation in the pursuit of his or her ends’”; this clearly is targeted at governments. But s. 2(d) also “prevents individuals from being overwhelmed by the powerful while at the same time providing collective strength” – a more general statement that arguably has application beyond actions by the state. Furthermore, Justice Donald in BCTF noted the particular concerns that animate the government’s constitutional duty to bargain. The relationship between a government and its employees is not the same as the relationship between a private employer and its employees for at least two reasons. First, government must represent the public, which includes the workers it 2017 CanLIIDocs 175 employs.118 And second, a government has the power to “unilaterally resolve impasse through legislation, or force workers to end a strike through constitutionally compliant back-to-work legislation.”119 In other words, the government has a broad public duty, but also holds a legislative trump card in its labour relations. A government must listen in good faith, perhaps, but once it has done so, it is arguably free to legislate as it sees fit. We therefore see three employee rights, and three corresponding employer duties. First, workers120 have the right to organize into independent employee associations, and the corresponding duty upon employers to recognize those associations. Outside of a Wagnerist model, this right and duty may have troubling implications, including a duty to recognize non-majoritarian employee associations. Second, workers have a right to collectively bargain; employers have a corresponding duty to bargain (or at least listen) in good faith. Third, workers have a right to strike; employers may have a corresponding duty not to discipline or dismiss workers who strike - a duty of non-retaliation. the Supreme Court has rejected a strict division between positive and negative rights: Fraser, supra note 2 at para 69. 117 For example, both Johner’s, supra note 113, and Syndicat des employées, supra note 113. Syndicat des employées involved employees who were already operating under a Wagnerist labour relations model. 118 BCTF BCCA, supra note 99 at para 338. 119 Ibid at para 339. 120 An open question, to which I do not have a satisfactory answer, would be whether associational rights would also accrue to those workers who are excluded from Wagnerist legislation – for instance managerial employees, employees who operate in a confidential capacity, etc. – but there seems no principled reason why these rights would not apply. 274 UNBLJ RD UN-B [VOL/TOME 68 Of course, the correspondence between governments and private employers is not exact. A private employer does not have the duty to represent the public at large; nor, obviously, does it have the power to impose a contract upon its workers. This may suggest that a provincial freedom of association guarantee is not necessarily the same as the constitutional guarantee, but there are counter-arguments in this regard. First, human rights codes apply to both the private sector and the public sector.121 It is possible that a provincial code would apply differently based on whether the act complained of was performed by a private employer or a public employer, but such a distinction does not appear within the legislation itself. Second, while a private employer does not have the power to unilaterally resolve an impasse, employees remain vulnerable in many if not most employment situations.122 An employer may, for instance, absent an attempt by employees to exercise collective rights under a Wagnerist statute,123 certainly take action to dismiss (or explicitly or implicitly threaten to dismiss) employees who are attempting to collectively bargain, and may dismiss (or accept the repudiation of 2017 CanLIIDocs 175 employment contracts by) employees who dare to strike. This dismissal may not be for just cause,124 and the employer may therefore be required to provide notice of dismissal or pay in lieu, but the employees will still be dismissed and out of the workplace. While by no means does a private employer hold the same powers as a government, it may still hold considerable power over those workers whom they employ, and the animating principles of freedom of association would seem to be applicable. Nonetheless, it is possible that freedom of association under human rights codes simply does not include the same web of rights and obligations that is included in the Charter as it applies within a Wagnerist regime (or exclusion from such a regime). Perhaps it is limited to the "thinnest" labour rights, as were recognized in the Alberta Reference, and does not include “thicker” rights such as a right to collective bargaining or a right to strike. Or perhaps the freedom is limited to the right to do collectively what one has the right to do individually, as Professor Langille has argued.125 While, for reasons given, I do not believe freedom of 121 SHRC, supra note 11, s 43; YHRA, supra note 11, s 38. 122 See e.g. Slaight Communications Inc v Davidson, [1989] 1 SCR 1038, 59 DLR (4th) 416. 123 As discussed previously, ULP protections are often limited to the exercise of rights under a particular labour relations statute – so attempts to unionize or collectively bargain outside of that scheme would not necessarily have the same protections. 124 Traditionally it likely would have been. In more recent times, where workplace discipline must be proportionate to the offence (McKinley v BC Tel, 2001 SCC 38, [2001] 2 SCR 161) and where the right to strike has been constitutionally affirmed, I would tentatively suggest that striking may be grounds for discipline but not dismissal. 125 See Langille, “Condescending Constitution”, supra note 83 and Langille, “Freedom of Association Mess”, supra note 73. 2017] THE RIGHT TOOL FOR THE JOB? 275 association is necessarily so limited within human rights legislation, it is worth addressing these alternative conceptions of freedom of association. The first alternative – the "constituitive" model of freedom of association – was roundly rejected by the Supreme Court in MPAO. The same model may be rejected regarding human rights codes for the same reasons. A general and purposive interpretation of the governing legislation arguably leads to the same conclusion regarding human rights codes as it does regarding the Charter (as was discussed earlier in this article). A right to meaningful collective bargaining is just as justified under human rights codes as under the Charter, and for similar reasons. The difficulty with the second alternative is that if freedom of association is merely the right to do collectively what one could do individually, any “right to collective bargaining” becomes difficult to justify. At common law, collective agreements are not enforceable, and while strikes are not necessarily illegal, they are invariably justification for dismissal, and may give rise to liability in tort. As Professor Judy Fudge has noted,126 emphasizing symmetry between an individual right and a collective analogue (for example, the right to strike) conceals the inherent 2017 CanLIIDocs 175 legal inequality that continues to suffuse employment law. In the present context, unless certain duties are recognized upon employers – at the very least, the duty of non-retaliation – a human rights code freedom of association guarantee remains more or less mute within the labour context. 5. Application Of course, even if freedom of association under human rights codes is more or less the same as freedom of association under the Charter, how is it to be applied? The Devil, as they say, is in the details. In the sections that follow I examine three uncontroversial applications of freedom of association under human rights legislation – aspirational, interpretive, and complementary – and three applications that I would imagine are much more controversial – the duty upon an employer to, first, recognize non-Wagnerist employee associations; second, to bargain; and third, to refrain from retaliating against workers who strike. i. Freedom of Association as Aspirational In addition to their investigative role, human rights commissions have an educational and aspirational role,127 to promote, and educate the public about, the rights protected 126 See Fudge, “Labour Rights as Human Rights”, supra note 76 at 613–615. 127 YHRA, supra note 11, s 16 (though the YHRA in this regard seems more specifically targeted at anti– discrimination measures); SHRC, supra note 11 at s 25 – though, again, the emphasis for much of s 25 276 UNBLJ RD UN-B [VOL/TOME 68 by human rights legislation. As noted previously, it was for this reason that Professor Adams argued that governments should add freedom of association under Canadian human rights legislation. Professor Adams viewed such a move as aspirational and educational – “proclaiming loudly that [labour rights] are rights equivalent to employment equity”.128 Professor Adams viewed this as part of a general strategy to bring about a “tripartite consensus” between government, employers, and unions (and the public at large) to create a culture friendlier to collective bargaining.129 Freedom of association has been present in human rights legislation for some time – in Saskatchewan, since 1947 – and it appears to have received little attention to date. But there is nothing in principle now preventing human rights commissions in Yukon, Saskatchewan, or Québec – or indeed elsewhere – from disseminating information on, and promoting, the new, purposive model of freedom of association. ii. Freedom of Association as Interpretive 2017 CanLIIDocs 175 Freedom of association within human rights codes can also serve as an interpretive tool, particularly when there are competing rights or freedoms – such as the potential conflict between freedoms of expression and association, on the one hand, and non- discrimination provisions on the other. For example, in one notable decision under an earlier version of the YHRA, Gould v Yukon Order of Pioneers, both the Yukon Supreme Court130 and the Yukon Court of Appeal131 considered freedom of association under the Act as a right that had to be balanced against the right of the applicant to demand non-discriminatory treatment. However when the case reached the Supreme Court of Canada,132 the majority was able to rule on the matter without considering freedom of association at all,133 though La Forest J argued in his concurring reasons that forcing association – in that case, forcing the respondent organization to accept female members – would act to "paralyze" the freedoms of expression and association.134 Similar “balancing” measures may take place under relates to anti-discrimination provisions rather than the Bill of Rights. Nonetheless, s 25(b) (“The commission shall…promote an understanding and acceptance of, and compliance with, this Act”), at least, seems to apply. See also Québec Charter, supra note 11. 128 Adams, Labour Left Out, supra note 16. 129 Adams, “From Statutory Right to Human Right”, supra note 74 at 63. 130 Gould v Yukon Order of Pioneers (1991), 87 DLR (4th) 618, 1991 CanLII 8298 (YK SC). 131 Yukon (Human Rights Commission) v Yukon Order of Pioneers, Dawson Lodge #1 (1993), 100 DLR (4th) 596, 1993 CanLII 3415 (YK CA) 132 Gould, supra note 60. 133 Ibid at para 18. 134 Ibid at para 74ff. McLachlin J (as she then was) strongly disagreed, noting that the very nature of anti- discrimination law that it “limit the freedom of those who are in a position to discriminate” (at para 145). 2017] THE RIGHT TOOL FOR THE JOB? 277 the common law – though of course Charter freedoms, and Charter values, may serve the same purpose.135 iii. Freedom of Association as a Complementary Remedy Another option is to apply freedom of association as a remedial ground within other proceedings. For example, the majority of the arbitration board in Johner’s Homestyle Catering v RWDSU Local 568136 found that the grievor had been harassed and demoted due to her activity on the part of the union. In addition to ordering that the grievor be compensated for lost wages, the board also ordered $10,000 in damages under The Saskatchewan Human Rights Code on the basis that the grievor's right to freedom of association – as defined by (at the time) BC Health Services – had been violated. This particular application of freedom of association does not seem to be controversial; collective agreements have been widely held to incorporate human rights and other employment-related statutes as implied terms.137 In labour arbitrations, then, it seems possible to use freedom of association as a basis for additional compensation for situations where an employee's freedom of association 2017 CanLIIDocs 175 has been infringed: anti-union discrimination or animus, for instance, as in Johner's. The value and availability of such awards may be limited by the terms of the legislation.138 But such application of freedom of association benefits those who are already unionized – who already have access to Wagnerist rights under labour legislation. It is less easy to apply such a remedy in individual wrongful dismissal claims. Canadian courts have been reluctant to imply statutory rights, particularly human rights codes, into individual contracts of employment.139 For the non- 135 See e.g. Wakeling JA’s dissent in Wall v Judicial Committee of the Highwood Congregation of Jehovah’s Witnesses, 2016 ABCA 255 at para 130ff (freedom of religion and freedom of association dictate that the courts should not intervene in religious organizations’ membership decisions). 136 Gould, supra note 60; Johners, supra note 113. 137 See Parry Sound (District) Social Services Administration Board v OPSEU, Local 324, [2003] 2 SCR 157, 2003 SCC 42 [Parry Sound]; Government of Alberta v Alberta Union of Provincial Employees, 2013 CanLII 29734 (AB GAA). Bisaillon v Concordia University, 2006 SCC 19 at para 33, [2006] 1 SCR 666, summarized it thusly: “This Court has considered the subject-matter jurisdiction of grievance arbitrators on several occasions, and it has clearly adopted a liberal position according to which grievance arbitrators have a broad exclusive jurisdiction over issues relating to conditions of employment, provided that those conditions can be shown to have an express or implicit connection to the collective agreement...” 138 SHRC, supra note 11 at s 31.4, limits compensation to a maximum of $20,000.00 and to situations where the party who has contravened the Code has acted "wilfully and recklessly" or where the injured party has “suffered with respect to feeling, dignity, or self-respect as a result of the contravention.” YHRA, supra note 11, however, contains no such limitations, requiring simply that the Act be contravened (s 20), and placing no restriction on the quantum of damages (s 24). 139 See e.g., Seneca College v Bhadauria, [1981] 2 SCR 181, 124 DLR (3d) 193 (refusal to recognize a tort of discrimination); Honda Canada Inc v Keays, 2008 SCC 39 at para 63, [2008] 2 SCR 362 (plaintiff 278 UNBLJ RD UN-B [VOL/TOME 68 unionized, then, we must explore the possibility of direct application of freedom of association under human rights legislation. iv. The Right to Organize / The Duty to Recognize Many workers will attempt to achieve, or will achieve, certification under Wagner Act style legislation; they may take advantage of their "thickest" Wagner Act model rights. My argument herein does not involve them directly. I am, rather, examining workers who seek non-Wagner Act bargaining; who eschew, either by choice or necessity, the Wagner Act model.140 This is likely through a form of minority, i.e. non-majoritarian, unionism, but it may also be for groups of workers who for whatever reason wish to collectively bargain but do not wish to assume the full panoply of Wagner Act rights.141 This has not been an option commonly pursued by Canadian workers, and the suggestion that non-Wagner unionism be legitimized has not gained widespread traction in the labour movement or elsewhere, not least because minority unionism, at least, carries real risks to labour rights in Canada.142 However having a non-majoritarian alternative in addition to the option of the 2017 CanLIIDocs 175 Wagnerist model has been long-championed by Professor Roy Adams and is perhaps more in keeping with international labour standards than is the existing single model.143 The argument by Adams and other proponents of minority unionism is premised on statutory reform144 – that governments should implement legislation that must pursue discrimination claim under human rights regime); Macaraeg v E Care Contact Centers Ltd, 2008 BCCA 182, 295 DLR (4th) 358 (overtime); Schulz v Beacon Roofing Supply Canada Company, 2016 BCSC 1475 (discrimination). Situations where the courts have been more open to pursuing statutory remedies in common law wrongful dismissal suits seem limited to where the statute lacks a comprehensive remedial scheme or where the statute does not clearly eliminate the common law remedy; see e.g. McCracken v Canadian National Railway Company, 2010 ONSC 4520, 3 CPC (7th) 81 (overtime under the Canada Labour Code). See also Evangelista v Number 7 Sales Limited, 2008 ONCA 599, 240 OAC 389, where the Ontario Court of Appeal – without explicitly addressing this issue – upheld a lower court's award for statutory vacation and public holiday pay within a wrongful dismissal claim. 140 One example of a statutory regime that is not reliant on majoritarianism or “trade unionism” as such is the Ontario Agricultural Employees Protection Act: Doorey, supra note 28 at 537. 141 One side-issue is that non-Wagner Act unions and associations would arguably not be bound by the "duty of fair representation" towards their members, premised as that duty is on majoritarian exclusivity: Steele v Louisville & Nashville Railroad Co, 323 US 192 (1944); Canadian Merchant Service Guild v Gagnon et al, [1984] 1 SCR 509, 9 DLR (4th) 641. 142 For concerns about minority unionism, see e.g. Brad Walchuk, "The Pitfalls of Embracing Minority Unionism" (2016) 6:3 J Workplace Rights 1. 143 See Adams, “Dunmore”, supra note 74; Adams, “Compliance”, supra note 17. Professor Adams does not suggest replacing the Wagner Act model but rather introducing non-majoritarian options in addition to the Wagner Act model. 144 Professor Adams argued that “the Charter protects the right of farm workers to organize, to bargain collectively in any format with which both they and their employers are comfortable, and to strike without putting their jobs in jeopardy or being punished for doing so – and that the onus is on the Ontario 2017] THE RIGHT TOOL FOR THE JOB? 279 will allow and protect workers’ right to organize into non-majoritarian unions. In keeping with the theme of this paper, however, I will examine the possibility of pursuing minority unionism through human rights codes.145 When considering risks, it is important to keep in mind that freedom of association under human rights codes has no greater ability to challenge Canadian Wagner Act model statutes than does the Charter. In other words, the Supreme Court appears satisfied for now that Wagnerism is constitutional. This means that for so long as there is a Wagnerist option open to workers, the risks of minority unionism through human rights legislation are fundamentally limited. Unless and until Canadian courts rule that the existing labour relations model is unconstitutional, workers will still be able to organize under the auspices of Wagnerist statutes. That means that they may take advantage of the “thickest” labour rights, including exclusivity (an employer must bargain with the certified bargaining agent, and only the certified bargaining agent) and unfair labour practice protections that work to prevent employers from undermining the collective bargaining process. Human rights-driven minority unionism does not directly interfere with those rights and protections; rather, it will almost inevitably operate on the periphery of the Wagner 2017 CanLIIDocs 175 Act model. It is possible that a greater emphasis on, or an enforceable right of, non- Wagnerist unionism would cause employers to promote such unions in bad faith, in order to dissuade workers from seeking majoritarian representation. But voluntary recognition under the Wagner Act model can be used in a similar manner.146 On the other hand, the very fact that human rights code-driven labour rights will operate outside of the Wagner Act model may demonstrate their value. The reach of Wagner Act unionism is short, and getting shorter, at least in the private sector. A duty upon employers to recognize employee associations even when those associations do not pursue certification may encourage recognition of labour rights among both employers and employees. It may legitimize labour rights; perhaps even serve as a "gateway" or half-measure which employees may pursue prior to then pursuing certification. All of that said, the main goal of recognition is to pursue some form of collective bargaining, and so it is impossible to fully canvass the implications of a government to formulate a policy that will effectively protect and promote those rights.” Adams, “Dunmore”, supra note 74. 145 In so doing I remain all too aware of the perils that minority unionism might bring, especially without being part of a broader effort in statutory reform. It is tempting to let sleeping freedoms lie. However, for reasons that follow I suggest the risks in the context suggested in this paper are limited. 146 E.g. an employer may voluntarily recognize a union perceived to be employer-friendly so as to dissuade employees from seeking representation by another, perhaps more assertive, certified bargaining agent. 280 UNBLJ RD UN-B [VOL/TOME 68 duty to recognize an employee association without also considering the extent of the employer's duty to bargain. v. The Right to Collectively Bargain / The Duty to Bargain in Good Faith It is possible to have collective agreements that exist independently of a Labour Relations Board order. Voluntary recognition, where an employer voluntarily agrees to accept a labour organization as the bargaining agent for a group of employees, is already recognized under labour statutes.147 In such situations, the employer and union are generally held to much the same standards as where a group of workers are represented by a certified bargaining agent. Generally, proof of majority support of a collective agreement, at least, is required in order to sustain a voluntarily recognized bargaining unit in the face of a certification application.148 In other words, voluntary recognition mimes Wagner Act majoritarianism.149 Non-majoritarian collective agreements seem to remain legally unenforceable.150 2017 CanLIIDocs 175 Earlier, in the discussion regarding BCTF, I (and Donald JA, indeed) suggested that an employer's duty to bargain (or "listen") in good faith falls short of a duty to conclude a collective agreement. I will not repeat that analysis, but I will add that outside of a Wagnerist statute, an employer's duty perhaps cannot include a duty to conclude a collective agreement. I would suggest that recognition of freedom of association under human rights codes does not necessarily make non-Wagnerist collective agreements enforceable. Unlike under a Wagnerist statute,151 "collective bargaining" under a human rights code would not necessarily displace the individual contract of employment. Rather, any “collective bargaining” that would occur would in effect be an employer considering in good faith workplace policies of general application, or identical or nearly identical terms and conditions of employment among a group of employees' contracts of employment.152 Again, this is not an unreservedly good thing for employees. The Supreme Court has made it clear that an employer is not required to negotiate with all groups 147 E.g. Labour Relations Code, RSA 2000, c L-1, ss 42–44; BC Labour Relations Code, RSBC 1996, c 244, s 34. 148 See e.g. SGEU v SIAST, [1989] SLRBD No 37 at 14–15. 149 See e.g. Ontario Workers’ Union v Service Employees International Union, Local 1, 2012 CanLII 71553 (ON LRB) at para 18. 150 See e.g. Ontario Secondary School Teachers Federation v Greater Essex District School Board, 2015 CanLII 38721 (ON LA) (MacDowell) at para 100; Northstar Lumber v United Steelworkers of America, Local No 1-424, 2009 BCCA 173 at para 105, 308 DLR (4th) 22. 151 McGavin Toastmaster Ltd v Ainscough, [1976] 1 SCR 718, 54 DLR (3d) 1. 152 It is also possible that freedom of association, as set out by the Supreme Court, would require that the common law revisit Young, supra note 8, and effectively bestow “personhood” by judicial fiat upon trade unions, but I do not believe that to be a necessary precondition to some form of meaningful employee voice via some form of collective or quasi-collective bargaining. 2017] THE RIGHT TOOL FOR THE JOB? 281 of employees who seek to collectively bargain153 but it also does not prevent an employer from doing so. It does not in itself prevent an employer from, for instance, playing two employee associations against one another, or paying higher wages to those employees who choose not to engage in this form of quasi-collective bargaining. But the Supreme Court's jurisprudence does suggest that an employer cannot simply dismiss employee collective representations out of hand. Where an employer does so, or engages in anti-associational discrimination in the manner described, it may be a matter for a human rights complaint. Similarly, if an employer chooses to exercise its common law right to dismiss employees who attempt to engage in collective bargaining, that too would potentially fall within the scope of a human rights complaint. In the result, a human rights commission would be empowered to bring an employer into mediation to attempt to resolve the complaint. That in itself may be sufficient to encourage good faith consideration of employee representations, and may further serve the aspirational goals of human rights legislation. If that were to fail, however, the remedies available would be much more limited than those available under labour relations legislation. A human rights tribunal (or in 2017 CanLIIDocs 175 Saskatchewan, the Court of Queen’s Bench) would be empowered to order the reinstatement of dismissed employees,154 or to order compensation,155 but it would not seem to have the authority to force the parties to bargain in good faith, for instance, or to table or accept particular bargaining proposals. This remedial shortfall in itself will limit the scope of an employer's duty to bargain – but given the nature of the duty as set out by the Supreme Court, where it is a duty to consider but not to agree, perhaps that is as it should and must be. vi. The Right to Strike / The Duty of Non-Retaliation I touched on the right of reinstatement above in the context of collective bargaining, but it has more significance when considering the right to strike. At the most basic level, a human rights complaint relating to freedom of association would allow for a right of workers not to suffer retaliation for attempting to exercise labour rights.156 153 MPAO, supra note 2 at para 83ff. 154 YHRA, supra note 11, s 21; SHRC, supra note 11, s 31.3. While workers in the federal jurisdiction, it has now been confirmed, have the right of reinstatement in unjust dismissal complaints under the Canada Labour Code (Wilson v Atomic Energy Commission, 2016 SCC 29, [2016] 1 SCR 770), workers in most provincial jurisdictions do not. For workers who are fired for attempting to unionize under the existing labour relations regime, there are Unfair Labour Practices to protect their rights. But for workers who attempt to seek voluntary recognition of a bargaining agent, perhaps, or who seek to bargain as a group without seeking certification, or who undertake strike action, the right of reinstatement -not normally available at common law - may be of some significance. 155 YHRA, supra note 11; SHRC, supra note 11, s 31.4; Québec Charter, supra note 11. 282 UNBLJ RD UN-B [VOL/TOME 68 That relatively straightforward principle presents challenges in the context of the right to strike. Those workers who strike outside of the protections of labour relations legislation have, at common law, breached or repudiated their contracts of employment and also face the possibility of torts brought against them – inducement of breach of contract if they encourage other workers to join the strike; interference with economic relations by unlawful means (i.e. threat of breach of contract or tortious behaviour); or the torts of trespass and nuisance that are already used as a basis for injunctions when picketing is undertaken. And if the strikers are successful in achieving their goals, having achieved their ends through threat of illegal action, i.e. either a strike or another economic tort, they may be liable in intimidation.157 The strike, however, is – to use the Supreme Court’s wording – the “powerhouse of collective bargaining”,158 an “essential component of the process through which workers pursue collective workplace goals”159 – “unique and fundamental” to effective collective bargaining.160 Professor Doorey noted that “thin” rights may be of limited benefit without the right to strike.161 A human rights 2017 CanLIIDocs 175 code that includes freedom of association may provide an answer. Unlike the more nebulous Charter values,162 freedom of association within human rights codes may allow workers to directly challenge the common law, limiting the power of employers to discipline or discharge workers who strike or otherwise seek to exercise labour rights; as well as to challenge tort liability that may accrue.163 That said, protecting a right to strike outside of the Wagner Act model is problematic. It would potentially herald a return to “recognition strikes”, where workers would use the strike, and not a certification process, to force an employer to collectively bargain. It could give rise to walkouts being used, as they were prior to the Wagner Act, to resolve workplace disputes, rather than resorting to arbitration. It would weaken, perhaps to absurdity, the restriction on strikes under labour legislation. These are significant challenges. If freedom of association includes the 156 See also YHRA, supra note 11, ss 29–30; SHRC, supra note 11, ss 35, 45. 157 See e.g. Christie, supra note 7; Arthurs, “Constitutionalizing the Right of Workers”, supra note 79 at 382. 158 SFL, supra note 2 at para 55 159 Ibid at para 46. 160 Ibid at para 51. 161 Doorey, supra note 28 at 539. 162 See Hill, supra note 4 at para 83ff. However, to date Charter values do not seem to have been used to challenge common law hostility to strikes or collective agreements. 163 I will note here again that, per Professor Fudge, the right to strike is a labour right that does not have an easy or appropriate individual corollary. An individual employee can certainly leave their job. However, it is only when it is done collectively that tort liability becomes likely, and the "strike" is at its heart distinct from an individual resignation. See Fudge, "Labour Rights as Human Rights", supra note 76. 2017] THE RIGHT TOOL FOR THE JOB? 283 right to strike, as the Supreme Court says it does, what does that mean under human rights legislation? For those workers operating within the Wagner Act model, it is almost certain that the restriction on strikes under labour relations legislation will pass constitutional muster. At least, appellate courts have previously upheld the restriction on mid-term strikes164 and Professor Etherington has suggested that recognition of the right to strike in SFL will not require a re-imagining of the existing model of labour relations.165 But that is premised on the Wagner Act model. One argument might be that strike restrictions within existing legislation is justifiable because workers accept a restriction on their strike activity as part of a “bundle” of Wagnerist rights and obligations. Workers operating outside of labour relations legislation do not gain the benefits of the legislation and, therefore, restricting their right to strike under statute may be less justifiable (under either the Charter or human rights legislation). It may also be that the Supreme Court’s ambiguity on the status of the right to strike limits the ambit of the right. If, as Abella J’s reasons suggest, the right to 2017 CanLIIDocs 175 strike is necessarily tied to collective bargaining, then that may serve as an internal limit on the right to strike under human rights legislation. Strike activity or time off work that is not connected to the “meaningful association in pursuit of workplace goals” would not be protected under freedom of association. That, too, gives rise to definitional problems, of course. Surely using the strike to convince an employer to collectively bargain is in pursuit of workplace goals! It seems to me that the ultimate question will be whether human rights legislation or the Charter will require that the “peace obligation” be declared unconstitutional. Paradoxically, the right to strike under human rights legislation may require that, insofar as a strike is pursuing collective workplace goals, an employer is not entitled to retaliate (whether through discipline, or treating a contract as repudiated) against workers who strike. But it might not displace statutory restrictions that would render the strike outright illegal. In that case, and if the Supreme Court's s. 2(d) jurisprudence does lead us to this conclusion (and I am suggesting it does, though I am not entirely convinced of that conclusion myself), an employer whose non-unionized workers undertake strike action may be required to seek relief from a labour relations board. And if that is the case - if the strike remains restricted under the Wagner Act model, and hence workers' ability to access the 164 Grain Workers’ Union, Local 333 et al v British Columbia Maritime Employers Association et al, 2009 FCA 201, [2010] 3 FCR 225 [Grain Workers], leave to appeal denied 2009 CanLII 71475 (SCC) (refusals to cross picket lines); British Columbia Teachers' Federation v British Columbia Public School Employers' Association, 2009 BCCA 39, 306 DLR (4th) 144 [BCTF (2009)], leave to appeal denied 2009 CanLII 44624 (SCC) (mid-term protest strikes). See also Dickson CJ's reasons in the "Dairy Workers" case, RWDSU v Saskatchewan, [1987] 1 SCR 460 at para 29ff, 38 DLR (4th) 277; which suggest that strike prohibitions will be relatively easy to justify under s 1 of the Charter. 165 See also Etherington, supra note 40. 284 UNBLJ RD UN-B [VOL/TOME 68 rights to organize and to collectively bargain is rendered more or less moot - then perhaps the Supreme Court has, despite its protestations, constitutionalized the Wagner Act after all. 6. Conclusion The Supreme Court’s constitutional, administrative law, and human rights jurisprudence raises numerous questions regarding the application of constitutional labour rights to workers outside of the public sector. In most jurisdictions, the question regarding the private sector – except insofar as statutory prohibitions or provisions are involved – can be safely ignored. But in jurisdictions where freedom of association appears within human rights legislation, and therefore is applicable to and enforceable in the private sector, I suggest that such questions need closer examination. The Supreme Court has provided a model of freedom of association that invokes profound and important principles, and also creates a significant, if rudimentary, set of constitutional labour rights and obligations. The implications may be equally profound and important; or they may be much more limited. To 2017 CanLIIDocs 175 determine which, it is important to determine where the Supreme Court's jurisprudence logically leads. SOME INITIAL THOUGHTS ON WILSON V. ATOMIC ENERGY OF CANADA LTD AND EDMONTON (CITY) V. EDMONTON EAST (CAPILANO) SHOPPING CENTRES LTD Diana Ginn* Part I: Introduction Administrative law focusses on the way in which, and the extent to which, courts should oversee the exercise of administrative authority. The law on substantive review of administrative decision-making has changed drastically over the last several decades, particularly around choice of standard of review. In the words of the Honorable John M Evans, courts have returned to this issue “with almost monotonous regularity over the last 30 years”.1 Two Supreme Court of Canada 2017 CanLIIDocs 175 decisions from 2016, Wilson v Atomic Energy of Canada Ltd2 and Edmonton (City) v Edmonton East (Capilano) Shopping Centres Ltd,3 have regenerated discussion about standard of review in relation to questions of law. No less an authority than the Honourable Justice David Stratas has suggested that the Court may be “about to embark on one of its once-a-decade, wholesale revisions to the law of judicial review”.4 To assess how Wilson and Capilano relate to the Supreme Court’s last wholesale revision of the law on substantive review in Dunsmuir v New Brunswick,5 this article: a) considers Justice Abella’s suggestion in Wilson that a separate standard of correctness review is no longer needed; b) assesses the trend, developing pre-Capilano and implicitly accepted by the majority in that decision, of limiting correctness review to the four categories of legal questions identified in Dunsmuir; and c) discusses the difficulties of applying the Dunsmuir understanding of reasonableness where there are only two possible interpretations of the legislative provision in dispute (Wilson and Capilano), or where the administrative decision- maker has not provided reasons on an issue under review (Capilano). *Professor Schulich School of Law, Dalhousie University. My thanks to Nicholas Hooper, JD 2017 Schulich School of Law (prospective) for his editorial assistance. 1 The Honorable John M Evans, “Triumph of Reasonableness: But How Much Does It Really Matter?” (2014) 27 Can J Admin L & Prac 101 at 101. 2 2016 SCC 29, [2016] 1 SCR 770 [Wilson]. 3 2016 SCC 47, [2016] 2 SCR 293 [Capilano]. 4 Honourable Justice David Stratas, “The Canadian Law of Judicial Review: A Plea for Doctrinal Coherence and Consistency” (2016) 42:1 Queen's LJ 27 at 41. 5 2008 SCC 9, [2008] 1 SCR 190 [Dunsmuir]. 286 UNBLJ RD UN-B [VOL/TOME 68 Before moving to an analysis of these three themes, the current law on standard of review is placed in context, followed by a brief overview of Wilson and Capilano. Part II: Background Context Under an earlier iteration of substantive review of administrative decision-making, questions of law decided by an administrative body were automatically subject to correctness review whenever the matter came before the courts by way of appeal, or on judicial review where the original decision was not protected by a privative clause.6 The correctness standard was also applied if the enabling legislation included a privative clause, and the matter in contention was judged to be an issue of jurisdiction. The deferential standard of patently unreasonable was reserved for questions of law where an administrative decision maker protected by a privative clause was deciding on an issue within its jurisdiction.7 Questions of fact were treated with significant deference; courts would intervene only if the administrative decision-maker’s findings were capricious or made without reference to the evidence 2017 CanLIIDocs 175 before it.8 Where the exercise of discretion was challenged, courts refrained from reviewing the outcome and instead confined themselves to asking specific questions about how the decision maker went about its tasks: Courts would ask whether irrelevant considerations had been taken into account, or relevant ones ignored, whether the decision maker had acted for an improper purpose, or whether the decision maker had fettered its statutorily-delegated powers.9 Thus, in this earlier era, substantive review of an administrative decision could be understood as involving a number of different silos, and placement in a particular silo depended on the nature of the question under review, and the route by which the matter came before the courts (appeal, judicial review without a privative clause, judicial review with a privative clause). Each of these silos attracted a specific test for judicial intervention. For issues of fact and the review of discretion, judicial intervention was applied sparingly. Not so with questions of law; such issues were reviewed on a correctness standard unless two criteria were in place: presence of a privative clause and identification of the issue as falling within the administrator’s jurisdiction.10 6 Gus Van Harten et al, Administrative Law: Cases, Text, and Materials, 7th ed (Toronto: Emond Publishing, 2015) at 629–630 [Van Harten et al]. See also David Jones, “Administrative Law In 2016: Update On Caselaw, Recent Trends and Related Developments” (Paper delivered at Newfoundland and Labrador Continuing Legal Education Seminar, 26 September 2016) at 11, online: 7 Van Harten et al, supra note 6 at 630. See also David Jones & Anne de Villars, Principles of Administrative Law, 6th ed (Toronto: Carswell, 2014) at 517. Relevant cases include: National Corn Growers Assn v Canada (Canadian Import Tribunal), [1990] 2 SCR 1324, 74 DLR (4th) 449 & Lester (WW) (1978) Ltd v United Association of Journeymen and Apprentices of the Plumbing and Pipefitting Industry, Local 740, [1990] 3 SCR 644, 88 Nfld & PEIR 15. 8 Van Harten et al, supra note 6 at 629. 9 Baker v Canada, [1999] 2 SCR 817, [1999] 174 DLR (4th) 193 [Baker]. 10 CUPE v NB Liquor Corporation, [1979] 2 SCR 227, 25 NBR (2d) 237 [CUPE 1979]. 2017] SOME INITIAL THOUGHTS ON WILSON AND CAPILANO 287 The Supreme Court of Canada’s last major reworking of substantive review occurred in 2008, with Dunsmuir v New Brunswick.11 In the decade preceding Dunsmuir, the Supreme Court had already made significant changes to the scheme described above. The possibility of deference on appeal was recognized and a middle standard between correctness and patent unreasonableness created;12 a unified approach to determining standard of review was developed, which applied regardless of whether the issue under review was characterized as a question of law, a question of fact or an exercise of discretion and regardless of whether the case involved judicial review or an appeal;13 a four-part pragmatic and functional analysis was established for the purpose of discerning whether the legislature intended the decisions of particular administrative entities to be treated with deference or not;14 and the concept of jurisdiction was largely relegated to the sidelines.15 In Dunsmuir, on judicial review of a decision by a labour adjudicator appointed under the Public Service Labour Relations Act,16 the Court applied a reasonableness standard, but overturned the adjudicator’s decision as unreasonable. The Court also took the opportunity to “develop a principled framework [of 17 substantive review] that is more coherent and workable,” reassessing the approach 2017 CanLIIDocs 175 which it had crafted less than a decade earlier in cases such as Pushpanathan and Baker. At the heart of this new framework was a desire to simplify the approach to determining standard of review. Besides melding the two deferential standards into one, Dunsmuir also: emphasized the role of precedent in establishing standards of review; identified four categories of questions of law to which correctness applies automatically (these being constitutional issues, jurisdictional issues, issues relating to the jurisdictional boundary between two specialized tribunals, and general 11 Dunsmuir, supra note 5. 12 A third, intermediate standard of review, reasonableness simpliciter, was identified in Canada (Director of Investigation and Research) v Southam Inc, [1997] 1 SCR 748, 144 DLR (4th) 1. Even before that, the possibility of deference on an appeal was starting to be accepted in decisions such as Pezim v British Columbia (Superintendent of Brokers), [1994] 2 SCR 557, 92 BCLR (2d) 145 [Pezim]. 13 Pushpanathan v Canada (Minister of Citizenship and Immigration), [1998] 1 SCR 982, 160 DLR (4th) 193 [Pushpanathan]; Baker, supra note 9. 14 Ibid. In order to discern legislative intent, reviewing courts were instructed to consider the purpose of the legislation and the specific provision; the nature of the question under review, the relative expertise of the administrative decision maker and the courts regarding that issue, and the presence or absence of a privative clause. 15 Pushpanathan, supra note 13 at 1005. According to Justice Bastarache, “it should be understood that a question which ‘goes to jurisdiction’ is simply descriptive of a provision for which the proper standard of review is correctness, based upon the outcome of the pragmatic and functional analysis. In other words, ‘jurisdictional error’ is simply an error on an issue with respect to which, according to the outcome of the pragmatic and functional analysis, the tribunal must make a correct interpretation and to which no deference will be shown.” 16 RSNB 1973, c P-25. 17 Dunsmuir, supra note 5 at para 32. 288 UNBLJ RD UN-B [VOL/TOME 68 questions of law of central importance to the legal system as a whole and beyond the expertise of the administrative decision maker); stated that deference may apply to legal questions, such as interpretation of the decision maker’s home or related statutes; and identified deference as almost always the appropriate standard for questions of fact, and discretionary decision making. Where the appropriate standard of review is not immediately obvious, Dunsmuir directed reviewing courts to consider some or all of the four elements of the standard of review analysis (a renamed pragmatic and functional analysis) to determine the legislature’s intention in this regard. The law on substantive review did not stand still after Dunsmuir. For instance, the relevance of Dunsmuir to judicial review of federal administrative decision makers was confirmed;18 certain types of constitutional analysis were carved out of the automatic-correctness classification;19 the relationship between the two arms of Dunsmuir’s description of reasonableness was explained;20 and, as discussed further below, the four categories of correctness came frequently to be treated as exhaustive. Not only was the Dunsmuir framework refined in a number of ways, but the oversight of administrative action is necessarily nuanced (or vague, 2017 CanLIIDocs 175 depending on one’s point of view), so subsequent courts on occasion wrestled with applying various aspects of Dunsmuir. However, most commentators would probably have agreed with Justice Evans’ 2013 statement that the law on standard of review was “reasonably well settled”.21 That sense of stability was at least partially disrupted by Capilano and Wilson. Part III: Overview of Wilson and Capilano A. Wilson v. Atomic Energy of Canada Ltd Wilson raised the question of whether amendments to the Canada Labour Code22 protect non-unionized employees from termination absent just cause. In the decades since these amendments were introduced, labour adjudicators across Canada had been at odds on this issue.23 The adjudicator in Wilson had ruled that under the Labour Code, non-unionized employees could not be dismissed on the provision of reasonable notice or payment in lieu – just cause was required. 18 Canada (Citizenship and Immigration) v Khosa, 2009 SCC 12, [2009] 1 SCR 339 [Khosa]. 19 Doré v Barreau du Québec (Tribunal des professions), 2012 SCC 12, [2012] 1 SCR 395 [Doré]. 20 Newfoundland and Labrador Nurses' Union v Newfoundland and Labrador (Treasury Board), 2011 SCC 62, [2011] 3 SCR 708 [Newfoundland Nurses’ Union]. 21 Evans, supra note 1 at 101. 22 RSC 1985, c L-2. 23 Jones, “Administrative Law In 2016”, supra note 6 at 27. 2017] SOME INITIAL THOUGHTS ON WILSON AND CAPILANO 289 On judicial review, the majority of the Supreme Court applied the reasonableness standard and upheld the adjudicator’s interpretation of the Code as reasonable. Had the discussion of standard of review stopped there, Wilson might not have attracted much notice beyond labour law practitioners. However, Justice Abella took the opportunity to provide “some general comments about standard of review”24 with the aim of “simplify[ing] the standard of review labyrinth we currently find ourselves in.”25 Acknowledging that her comments in this regard would be obiter, Justice Abella offered her proposals “as an option only, for purposes of starting the conversation about the way forward.”26 Her first, and primary, proposal was that the standard of correctness be mothballed, and all substantive review of administrative action be approached from a deferential stance. As a back-up proposal, in case “there prove[d] to be little appetite for collapsing the two remaining standards of review,”27 Justice Abella proposed limiting the scope of correctness review, such that a “residual ‘correctness’ standard” would be available “only in those four circumstances Dunsmuir articulated.”28 Four other members of the majority (Chief Justice McLachlin, and Justices Karakatsanis, Wagner and Gascon) agreed with Abella J that the labour adjudicator’s 2017 CanLIIDocs 175 decision deserved deference and that he had in fact arrived at a reasonable conclusion. On her broader comments, they thanked Justice Abella for her “efforts to stimulate a discussion on how to clarify or simplify our standard of review jurisprudence to better promote certainty and predictability,”29 but declined to “endorse any particular proposal to redraw our current standard of review framework at this time.”30 Justice Cromwell, also in the majority as to choice of standard of review and outcome, was even more forthright in his view that Dunsmuir sets out the appropriate framework for addressing the standard of judicial review. No doubt, that framework can and will be refined so that the applicable standard of review may be identified more easily and more consistently. But the basic Dunsmuir framework is sound and does not require fundamental re-thinking.31 24 Wilson, supra note 2 at para 19. 25 Ibid. 26 Ibid. 27 Ibid at para 38. 28 Ibid. 29 Ibid at para 70. 30 Ibid. 31 Ibid at para 72. 290 UNBLJ RD UN-B [VOL/TOME 68 In dissent, Justices Côté and Brown, writing for themselves and Justice Moldaver, referred appreciatively to the “constructive spirit”32 in which Justice Abella offered her obiter comments, but admitted to “harbour[ing] concerns about their merits,”33 and refused to engage in further speculation regarding “what is already the subject of a peripatetic body of jurisprudence.”34 The dissent would have held “the narrow and distilled legal issue”35 under review to the standard of correctness, on the grounds that rule of law principles of consistency and the “promise of orderly governance” 36 required the Court to provide a definitive interpretation of the provision in question, after decades of conflicting interpretations from labour arbitrators. Further, the dissent held that the adjudicator in this case had interpreted the statute incorrectly. In their view, employers bound by the Canada Labour Code may terminate non- unionized employees without cause, so long as the employee is provided reasonable notice or pay in lieu thereof.37 B. Edmonton (City) v. Edmonton East (Capilano) Shopping Centres Ltd The dispute in Capilano centred on the authority given to local assessment review 2017 CanLIIDocs 175 boards hearing taxpayer appeals under the Alberta Municipal Government Act.38 Specifically, when a taxpayer appeals a municipal tax assessment, can the board raise the assessment, or is it limited to either confirming or lowering the assessment? In this case, the board concluded that it did indeed have the authority to increase the tax assessment, and proceeded to do so. Edmonton East (Capilano) Shopping Centres Ltd challenged the decision under a section of the Municipal Government Act which provided for appeals on “a question of law or jurisdiction of sufficient importance to merit an appeal”.39 Justice Karakatsanis, writing for Justices Abella, Cromwell, Wagner and Gascon held that the board’s decision should be judged against the standard of reasonableness. Further, in the majority’s view, the board’s interpretation of its enabling legislation was reasonable. The relevant section, which referred to the board choosing to “change” the assessment or “decid[ing] that no change is required”,40 could reasonably be understood as allowing an increase in the challenged 32 Ibid at para 78. 33 Ibid. 34 Ibid. 35 Ibid at para 91. 36 Ibid at para 84. 37 Ibid at para 149. 38 Capilano, supra note 3 at para 76. 39 Ibid at para 11. 40 RSA 2000, c M-26, s 467(1). 2017] SOME INITIAL THOUGHTS ON WILSON AND CAPILANO 291 assessment. The majority chose not to revisit of standard of review analysis, as suggested by Justice Abella in Wilson. According to Justice Karakatsanis, [t]he majority appreciated Justice Abella’s efforts to stimulate a discussion on how to clarify or simplify our standard of review jurisprudence to better promote certainty and predictability. In my view, the principles in Dunsmuir should provide the foundation for any future direction. However, any recalibration of our jurisprudence should await full submissions. This appeal was argued on the basis of our current jurisprudence and I proceed accordingly.41 Justice Karakatsanis then set out her understanding of the Dunsmuir principles. Where, as on this appeal, the standard of review is not established by precedent, and the question under review relates to the administrative decision maker’s home statute, “the standard of review is presumed to be reasonableness.”42 The presumption of reasonableness was described as “grounded in the legislature’s choice to give a specialized tribunal responsibility for administering the statutory provisions, and the expertise of the tribunal in so doing.”43 Justice Karakatsanis noted that this expertise could result from “specialization of functions”,44 “a habitual 2017 CanLIIDocs 175 familiarity”45 with one’s home statute, or from provisions in that statute requiring “that members of a given tribunal possess certain qualifications.”46 However, in gauging the expertise of the assessment review board to interpret its enabling legislation, the majority in Capilano did not search for the enumerated indicators of the board’s expertise. Instead, the majority assumed expertise on the part of the board, based on the fact that the board had been created and given statutory authority in the first place, and the fact that it was interpreting its home statute. A statement from Dunsmuir to the effect that adjudicators under the Public Service Labour Relations Act could be presumed to have expertise in their home and related statutes was broadened to apply to administrative decision-makers more generally.47 Justice Karakatsanis stated that the presumption of reasonableness review could be rebutted if the issue in dispute fell into one of the four Dunsmuir categories 41 Capilano, supra note 3 at para 20. 42 Ibid at para 22, citing Mouvement laïque québécois v Saguenay (City), 2015 SCC 16 at para 46, [2015] 2 SCR 3 [Saguenay]. See also Alberta (Information and Privacy Commissioner) v Alberta Teachers' Association, 2011 SCC 61 at para 39, [2011] 3 SCR 654 [Alberta Teachers’ Association]. There is also recent authority for the proposition that the onus lies with the party wishing to rebut the presumption of deference, see McLean v British Columbia (Securities Commission), 2013 SCC 67 at paras 40–41, [2013] 3 SCR 67. 43 Capliano, supra note 3 at para 33. 44 Ibid. 45 Ibid. 46 Ibid. 47 Dunsmuir, supra note 5 at para 68. 292 UNBLJ RD UN-B [VOL/TOME 68 which automatically attract the correctness standard. Of these, the only potentially applicable category identified by the Court was jurisdiction, and the majority dismissed this possibility quickly, saying that “[n]o true question of jurisdiction arises.”48 Edmonton East had argued that the wording of the appeal section indicated a legislative intention for correctness review on “a question of law or jurisdiction of sufficient importance to merit an appeal”,49 and deference otherwise. The majority at the Supreme Court largely treated this as an argument that statutory rights of appeal should be seen as creating a new category of automatic correctness - a contention that was rejected as “go[ing] against strong jurisprudence from this Court.”50 The majority also concluded that no further contextual analysis was needed, both because of a long line of past jurisprudence applying the reasonableness standard to statutory appeals, and because the result would be the same in any case.51 More generally, the majority warned that delving into context has inherent dangers since “[t]he contextual approach can generate uncertainty and endless litigation concerning the standard of review.”52 2017 CanLIIDocs 175 Four members of the Court dissented in Capilano. Justices Côté and Brown, writing for themselves, Chief Justice McLachlin and Justice Moldaver, held that the board’s decision should be measured against a correctness standard, and that the board was incorrect in concluding it had the authority on a taxpayer appeal to raise the municipal assessment. To justify their choice of the correctness standard, the dissent pointed out that a right of appeal was provided for only certain matters, and in their view these matters “transcend[ed] the particular context of a disputed assessment”.53 They continued: The legislature of Alberta created a municipal assessment complaints regime that allows certain questions squarely within the expertise of an assessment review board to be reviewed on a deferential standard through the ordinary mechanism of judicial review. The legislature, however, also 48 Capilano, supra note 3 at para 26. The finding that the question before the assessment board did not fit within one the four Dunsmuir categories that attract automatic correctness seems sensible. No category except jurisdiction was potentially relevant. And if one accepts current-day understandings of jurisdiction as involving the authority (or not) to enter into the very question before the administrative decision maker (an understanding set out in Dunsmuir, supra note 5 at para 61), there was nothing of this nature before the assessment review board in Capilano. Of course, the Supreme Court’s treatment of jurisdictional issues leaves the question of what kinds of specific question might in fact be accepted as jurisdictional, but that is not the focus of this article. Thus, my criticism of the majority decision in Capilano is not its failure to classify the issue before it as falling within one of the four Dunsmuir categories of mandatory correctness review. Instead, my criticism, expanded upon within, relates to the majority’s treatment of the Dunsmuir categories as exhaustive. 49 Ibid at para 11. See also supra note 40, ss 470(1), (5). 50 Capilano, supra note 3 at para 28. 51 Ibid at para 34. 52 Ibid at para 35. 53 Ibid at para 78. 2017] SOME INITIAL THOUGHTS ON WILSON AND CAPILANO 293 designated certain questions of law and jurisdiction — for which standardized answers are necessary across the province — to be the subject of an appeal to the Court of Queen’s Bench. Where the court quashes a decision, its answers to these questions are binding on the Board. This leads to the unavoidable conclusion that the legislature intended correctness review to be applied to these questions.54 Further, the dissent disagreed with the majority’s assessment of the board’s expertise and this too, in their view, supported correctness review. Given that administrative expertise is to be judged in relation to a particular issue and in comparison to the expertise of the court, and given the “vast array of municipal government issues”55 dealt with in the Municipal Government Act, the dissent argued it would be unrealistic to assume that the assessment review board possessed greater expertise than the courts on all such issues, including statutory interpretation.56 Justices Côté and Brown also justified strict scrutiny because of the need for consistency: Because each assessment review board is a distinct entity, there is no overarching institutional body capable of promoting consistency in the interpretation and application of the Act between them. …. Consistency in 2017 CanLIIDocs 175 the understanding and application of these legal questions is necessary, and only courts can provide such consistency.57 Thus, in the dissent’s view, all indicators of legislative intent pointed to correctness review for the question of whether the assessment review board was authorized to raise a tax assessment in the context of a taxpayer appealing to have it lowered. More generally, the dissent in Capilano took issue with the majority’s attempt to limit correctness review to the four categories identified in Dunsmuir, stating that “[a]n approach to the standard of review analysis that relies exclusively on categories and eschews any role for context risks introducing the vice of formalism into the law of judicial review”.58 Justices Côté and Brown were at pains to emphasize they were not advocating a new category, such that all statutory appeals would attract the correctness standard; however, they also emphasized that the existence and wording of a particular appeal right could not be treated as irrelevant. Instead, “[a] statutory right of appeal, like a privative clause, ‘is an 54 Ibid at para 63. 55 Ibid at para 86. 56 This exchange between the majority and dissent on expertise has led Shaun Fluker to comment, “A fissure is developing at the Supreme Court over standard of review, and I suggest it is forming around this notion of relative expertise.” See Sean Fluker, “The Supreme Court of Canada (By a Slim Majority) Confirms the Presumption of Deference in Alberta” (8 November 2016), ABlawg: The University of Calgary Faculty of Law Blog, online: 57 Capilano, supra note 3 at para 80. 58 Ibid at para 70. 294 UNBLJ RD UN-B [VOL/TOME 68 important indicator of legislative intent’ and, depending on its wording, it ‘may be at ease with judicial intervention’”.59 Part IV: Analysis Having provided background on the evolution of the standard of review analysis, and a synopsis of Wilson and Capilano, I now turn to the three themes outlined in the introduction: the merits of maintaining the correctness standard of review; the scope of correctness review and particularly, whether the four categories of automatic correctness review enumerated in Dunsmuir should be treated as exhaustive; and whether the challenges of applying the Dunsmuir understanding of reasonableness in certain situations will inevitably result in judicial reasoning that looks very much like correctness review. A. Maintaining the Availability of Correctness Review 2017 CanLIIDocs 175 The standard of review discussion in Wilson captured attention primarily because of Justice Abella’s suggestion that it is once again time for a significant revision to administrative law, and in particular that correctness review should disappear entirely. Of course, Justice Abella is not alone in this suggestion;60 in Wilson she called the elimination of the correctness standard the “most obvious and frequently proposed reform of the current system”.61 Further, the desire for one deferential standard of review is not new. Jones and de Villars note that after the landmark 1979 decision in CUPE v NB Liquor Corporation,62 the “euphoric (but ultimately incorrect) reaction by many administrative law observers … was that the ‘patent unreasonableness’ test should be applied in all circumstances… to protect all decisions of all statutory delegates from all forms of judicial review.”63 While Justice Abella’s comments certainly engaged attention, none of the other members of the Supreme Court were interested in following up on her proposals, and the dissenting judges in both Wilson and Capilano applied the correctness standard. Further, in Information and Privacy Commissioner of Alberta v Board of Governors of University of Calgary,64 released very shortly after Capilano, 59 Ibid at para 73, citing Khosa, supra note 18 at para 55. 60 See e.g. supra note 1. 61 Wilson, supra note 2 para 28. 62 Supra note 10. 63 Jones & de Villars, supra note 7 at 522 [emphasis in original]. Ultimately, the Supreme Court, in decisions such as National Corn Growers and Lester “clearly recognized that this was not an accurate statement of either the law or the court’s own constitutional role” (at 523). Since, at the time, only two standards existed – patent unreasonableness and correctness – the Court’s rejection of universal application of the former meant a reaffirmation of the continuing role of correctness review. 64 2016 SCC 53, [2016] 2 SCR 555. 2017] SOME INITIAL THOUGHTS ON WILSON AND CAPILANO 295 five of the seven judges identified correctness as the appropriate standard of review for a decision made by a provincial Information and Privacy Commissioner, and a sixth was willing to assume the same, without deciding. Therefore, the standard of correctness does not seem to be in immediate danger of disappearing. However, proposals for a single deferential standard of review are also unlikely to vanish, so it is worth assessing the principled and pragmatic arguments for either retaining or jettisoning the possibility of correctness review on some questions of law.65 Considerations of Principle The notion that there are some constitutional limits on the scope of administrative power runs as a recurrent theme through administrative law jurisprudence and commentary. In the famous words of Justice Rand in Roncarelli v Duplessis, “[i]n public regulation of this sort there is no such thing as absolute and untrammelled ‘discretion’”.66 As Guy Régimbald explains, judicial review of administrative decision-making is a core component of constitutional law: 2017 CanLIIDocs 175 Judicial review is the procedure allowing superior courts to look at a decision of a public body, and determine if the decision is within the scope of its powers as delegated by the legislature. Judicial review is rooted in the basic tenets of constitutional law as a consequence of the relationship between the principles of Parliamentary sovereignty, the rule of law, and the inherent power of the courts to review the legality of actions in order to maintain an adequate balance between these two principles.67 Traditionally, the concept of jurisdiction was seen as a way to maintain this balance. Thus, on judicial review, legislative supremacy is protected through the understanding that so long as legislatures and Parliament do not act in unconstitutional ways, they are empowered “to create various administrative bodies and endow them with broad powers.”68 Further, they are free to signal strongly, via finality or privative clauses, their intention that courts are to be deferential to those entities’ decisions. However, as articulated in Crevier v Quebec (Attorney General), legislatures cannot completely block judicial review, at least on matters of jurisdiction, as this would be inimical to that strand of the rule of law which allows 65 Those who argue that correctness still plays a useful role in the judicial oversight of administrative action are not suggesting that findings of fact, or true discretionary decision making should be subjected to strict scrutiny, and there would likely be little appetite for such a suggestion among judges themselves. As Justice Evans notes, “[b]y and large, judges have little difficulty in upholding tribunals’ findings of fact or exercises of discretion, even when they believe that they might have reached different conclusions, had they been the original decision-makers” (supra note 1 at 107). 66 [1959] SCR 121, 16 DLR (2d) 689. 67 Guy Régimbald, Canadian Administrative Law, 2nd ed (Markham: LexisNexis, 2015) at 27. 68 Ibid. See also Dunsmuir, supra note 5 at para 27. 296 UNBLJ RD UN-B [VOL/TOME 68 individuals access to the courts when their rights are affected.69 Outside the protection of a privative clause, incorrect decisions on matters of law were traditionally seen as depriving an administrative decision-maker of jurisdiction. Thus both deference and strict scrutiny were seen as necessary elements in acknowledging legislative intent, but keeping administrative action within constitutional bounds. While the importance attributed to the concept of jurisdiction is greatly diminished in today’s law on standard of review, the “basic tenets” and the balance outlined by Régimbald above are still routinely referred to by commentators,70 and underlie the Dunsmuir approach to substantive review. In Dunsmuir, Justices Bastarache and LeBel started their re-evaluation of the law regarding choice of standard of review by referencing the twin pillars of rule of law and respect for legislative intent: As a matter of constitutional law, judicial review is intimately connected with the preservation of the rule of law. It is essentially that constitutional foundation which explains the purpose of judicial review and guides its function and operation. Judicial review seeks to address an underlying 2017 CanLIIDocs 175 tension between the rule of law and the foundational democratic principle, which finds an expression in the initiatives of Parliament and legislatures to create various administrative bodies and endow them with broad powers. Courts, while exercising their constitutional functions of judicial review, must be sensitive not only to the need to uphold the rule of law, but also to the necessity of avoiding undue interference with the discharge of administrative functions in respect of the matters delegated to administrative bodies by Parliament and legislatures.71 Arguably, removing the possibility of correctness review would strike at the Court’s consistently-articulated conception of the foundational basis for judicial review. It is certainly true that, as Justice Abella pointed out in Wilson,72 neither Crevier nor Dunsmuir explicitly tied arguments about the constitutionality of administrative action to the continued existence of a correctness standard. However, 69 [1981] 2 SCR 220, 127 DLR (3d) 1. With a provincial administrative decision maker, there would be the additional concern that the province was attempting to create a section 96 court, something that falls within federal jurisdiction. 70 See, for instance, Sara Blake: The purpose of determining which standard of review to apply in each case is to respect the constitutional roles of the court, the legislature and the executive. In keeping with its role to uphold the rule of law, the court exercises its power of review so as to ensure that the tribunal does not overstep its legal authority, while respecting the intentions of the democratically elected legislature by giving deference to the wisdom of the tribunal decision on the merits. Administrative Law in Canada, 5th ed (Markham: LexisNexis Canada, 2011) at 209. Similarly, Jeremy deBeer et al, Standards of Review of Federal Administrative Tribunals, 4th ed (Toronto: LexisNexis Canada, 2012) at 53: “Reviewing courts are torn between their supervisory function as independent guardians of the rule of law and their respect for the will of the legislature to delegate decision-making responsibility to administrative tribunals.” 71 Dunsmuir, supra note 5 at para 27. 72 Wilson, supra note 2 at para 31. 2017] SOME INITIAL THOUGHTS ON WILSON AND CAPILANO 297 at the time of Crevier, correctness review on jurisdictional matters would have been the assumed backdrop, thus making an explicit reference unnecessary. As Jones and de Villars note, the “courts’ treatment of privative clauses” has always related to the “presumed right of the court to rule on the legality of governmental action”,73 and this concern for legality has always been based on the principle that at least sometimes administrative decisions must be correct, not just reasonable. In Dunsmuir, the emphasis was less on the impact of privative clauses and more on the function of judicial oversight of administrative action generally; again however, the idea that some questions of law require correctness review was clearly an integral part of Justices Bastarache and LeBel’s understanding of the courts’ function – else, why identify certain kinds of issues that automatically attract correctness review? If retaining the possibility of correctness review is a core aspect of maintaining the balance between legislative intent and rule of law on judicial review of questions of law, there seems even less of a principled basis for arguing that the possibility of correctness review should be jettisoned in the context of statutory appeals. On an appeal, the rule of law and foundational constitutional principles are unlikely to be in tension. From the time of Crevier, where the Supreme Court first 2017 CanLIIDocs 175 explicitly held that legislatures cannot completely oust judicial oversight of administrative decision making, the point of introducing the rule of law has been to explain why legislative intent, as expressed through strong privative clauses, cannot be given full force. However, where the rule of law’s emphasis on access to the courts is built into the administrative scheme itself by way of an appeal provision, this must surely free courts to concentrate solely on the “polar star” of legislative intention.74 This is certainly not to advocate for a return to pre-Pezim75 days, with automatic application of the correctness standard to all appeals; however, where the legislature has invited judicial oversight by way of a statutory appeal, and the wording of a particular appeal provision or other indicia of legislative invite strict scrutiny, there is no principled reason for a court not to implement that. (Unless of course one accepts the conclusion of Justice Robertson of the New Brunswick Court of Appeal, that “the deference obligation is no longer the product of the will of the legislature or Parliament.”76) My argument in this section, then, is that there is no principled reason for excising the possibility of correctness review, for either judicial review or appeals. In fact, longstanding understandings about the constitutional relationship among 73 Jones & de Villars, supra note 7 at 569. 74 CUPE v Ontario (Ministry of Labour), 2003 SCC 29 at para 149, 66 OR (3d) 735. See also Khosa, supra note 18 at para 93.. 75 Pezim, supra note 12. 76 The Honourable Joseph Robertson, “Judicial Deference to the Decisions of Administrative Tribunals: A Guide to 60 Years of Supreme Court Jurisprudence” in Joseph Robertson, Peter Gall & Paul Daly, eds, Judicial Deference to Administrative Tribunals in Canada: Its History and Future (Markham: LexisNexis, 2014) 1 at 3. 298 UNBLJ RD UN-B [VOL/TOME 68 legislatures, courts, and those who exercise administrative power, point to the necessity of retaining the standard of correctness, even while accepting that it will be used less often than reasonableness. On judicial review under a privative clause, correctness remains a way of protecting the rule of law, by ensuring that legislatures cannot insulate statutorily-created entities from the challenge that they have overstepped their delegated authority. On an appeal, if correctness review appears to be the legislature’s intent, preserving the correctness standard allows the courts to fulfil their constitutional role by giving full weight to that intent. There could be no grounds for refusing to apply a correctness standard where this is expressly mandated by the legislature,77 and there seems no principled reason to ignore a legislative desire for correctness review simply because this is communicated through other indicators of intent. Pragmatic Considerations If an argument is to be made for excising correctness review entirely, it must be made on pragmatic, rather than principled grounds. Two such arguments might be 2017 CanLIIDocs 175 offered, both relating to efficient use of the judicial system. Would moving to a single deferential standard of review save court resources, either through eliminating the need for arguments about, and judicial determination of, the appropriate standard of review, or through reducing the instances of judicial review or appeal? Dunsmuir hoped to reduce the amount of time spent arguing about standard of review, and the merging of the two deferential standards was clearly intended to be a step in this direction. In Wilson, similar concerns about time and effort led to Justice Abella’s proposal for a single standard of review: A substantial portion of the parties’ factums and the decisions of the lower courts were occupied with what the applicable standard of review should be. This, in my respectful view, is insupportable, and directs us institutionally to think about whether this obstacle course is necessary or whether there is a principled way to simplify the path to reviewing the merits.78 To the extent that Dunsmuir has encouraged lower courts to approach standard of review analysis more briskly, it seems likely that the merging of reasonableness simpliciter and patently unreasonable has played a role; however, I suspect that other elements of the Dunsmuir revisions such as the focus on precedent, the reversal of the previous requirement to do a complete standard of review analysis every time, and delineation of categories where deference will usually apply are 77 As Justice Karakatsanis stated in Capilano, “[s]ubject to constitutional constraints, the legislature can specify the applicable standard of review. In British Columbia, for example, the legislature has displaced almost the entire common law on the standard of review … Unfortunately, clear legislative guidance on the standard of review is not common” (supra note 3 at para 35). 78 Wilson, supra note 2 at para 20. 2017] SOME INITIAL THOUGHTS ON WILSON AND CAPILANO 299 equally significant.79 The amalgamation of deferential standards will have had some effect on brevity in that courts no longer have the thankless task of trying to explain the difference between unreasonable and not patently unreasonable. Other than that, however, discussion that previously occurred at the stage of choosing between unreasonableness or patent unreasonableness has largely moved to the application stage. Keeping in mind that “reasonableness takes its colour from the context”,80 at the application stage the court must evaluate arguments regarding the range of outcomes that could be considered reasonable. Contextual arguments that in pre- Dunsmuir days would have been seen as relevant to deciding which deferential standard of review to apply will now be relevant to deciding whether the span of acceptable outcomes is broad or narrow.81 Similarly, if correctness review disappeared, that would certainly avoid any initial debate about the appropriate standard of review. However, arguments that are now used at that initial stage (for instance, arguments about the significance of the issue to the legal system as a whole) would likely crop up at the stage of determining the range of possible defensible interpretations that could be given to the provision under dispute. Pragmatic arguments for doing away with correctness review could relate 2017 CanLIIDocs 175 not just to a desire for greater brevity in legal argument and judicial analysis, but also to concerns about the number of challenges mounted to administrative decision- making. Does the continued possibility that courts will review against a correctness standard encourage appeals and applications for judicial review? There seems little empirical evidence that getting rid of correctness review would reduce litigation. Despite the frequent application of deference in fields such as labour relations, judicial review is still sought regularly – as is clear from both Dunsmuir and Wilson. In fact, the goal of preserving judicial resources might actually be undermined by eliminating the possibility of correctness review. David Jones contends that: 79 I say this based at least in part on a comprehensive study of substantive review of administrative action by the lower courts since Dunsmuir, undertaken by William Lahey and myself. Our findings and analysis will be reported in a series of three articles in the Canadian Journal of Administrative Law and Practice. Our first piece (coauthored along with Lauren Soubolsky and Madison Veinotte and to be published shortly), focusses on substantive review in the federal courts. We found that federal courts have taken seriously the Supreme Court’s strictures on spending excessive time on choice of standard of review – but our analysis of the 216 cases forming the basis of our first article led us to conclude that this was the result of a number of elements in the Dunsmuir framework, not simply the reduction in the number of available standards of review. 80 Khosa, supra note 18 at para 59, and Catalyst Paper Corp v North Cowichan District, 12 SCC 2 at para 18, [2012] 1 SCR 5. 81 Again, I make this suggestion based on what we have seen thus far in our analysis, referenced at supra note 80, of lower courts’ application of Dunsmuir. See also Jones & de Villars, supra note 7 at 544–545. A list of questions still unanswered by Dunsmuir includes the following: “Does the merging of the two former deferential standards actually achieve anything? Or does it merely shift the discussion from determining which of the two deferential standards applies to determining whether the substance of the administrative decision is ‘unreasonable’ in any one of a myriad of possible ways of being unreasonable?” 300 UNBLJ RD UN-B [VOL/TOME 68 if reasonableness ever becomes the only standard of review, that will not achieve either certainty or less litigation— there will be every incentive for litigants to take every case up the line in the hope of persuading the next judge that their interpretation is reasonable. Nor will it achieve consistency where there are two or more different reasonable interpretations.82 Having argued that neither considerations of principle nor pragmatic concerns support the elimination of the correctness standard of review, I now turn to the question of how broadly available this standard should be. B. The Scope of Correctness Review In Dunsmuir, Justices Bastarache and LeBel started their discussion on selecting the appropriate standard of review with the following statement: [Q]uestions of fact, discretion and policy as well as questions where the legal issues cannot be easily separated from the factual issues generally 2017 CanLIIDocs 175 attract a standard of reasonableness while many legal issues attract a standard of correctness. Some legal issues, however, attract the more deferential standard of reasonableness.83 The deferential standard would “usually automatically apply” to factual findings, on review of discretion, and “where the legal and factual issues are intertwined with and cannot be readily separated.”84 Deference would “usually”85 be appropriate: where a tribunal is interpreting its own statute or statutes closely connected to its function, with which it will have particular familiarity [or] …. where an administrative tribunal has developed particular expertise in the application of a general common law or civil law rule in relation to a specific statutory context.86 Further, a reasonableness test should be applied on questions of law where this intention is signalled by the legislature, for instance through factors such as the presence of a privative clause (although the presence or absence of such a clause is not determinative) or the legislature’s creation of a “discrete and special administrative regime in which the decision-maker has special expertise”.87 The 82 David Jones, “Administrative Law in 2016 Part II – An Additional Case”, Case Comment on Edmonton (City) v Edmonton East (Capilano) Shopping Centres Limited, 2016 SCC 47, [2016] 2 SCR 293, online: 83 Dunsmuir, supra note 5 at para 51. 84 Ibid at para 53. 85 Ibid at para 54. 86 Ibid. 87 Ibid at para 55. 2017] SOME INITIAL THOUGHTS ON WILSON AND CAPILANO 301 possibility of reviewing questions of law for reasonableness was justified on the grounds that “[t]here is nothing unprincipled in the fact that some questions of law will be decided on the basis of reasonableness. It simply means giving the adjudicator’s decision appropriate deference in deciding whether a decision should be upheld, bearing in mind the factors indicated.”88 (This of course needs to be read in light of Justices Bastarache and LeBel’s earlier statement that “many legal issues attract a standard of correctness.”89) After justifying the possibility of deference to some questions of law, Justices Bastarache and LeBel identified four kinds of legal questions for which correctness review is mandatory. Thus, ‘[a] question of law that is of “central importance to the legal system . . . and outside the . . . specialized area of expertise’ of the administrative decision maker will always attract a correctness standard”,90 because “such questions require uniform and consistent answers.”91 Justices Bastarache and LeBel next referred to previous case law as indicating that division of powers issues, “as well as other constitutional issues are necessarily subject to correctness review because of the unique role of s. 96 courts as interpreters of the Constitution”.92 As well, “[a]dministrative bodies must … be correct in their determinations of true questions 2017 CanLIIDocs 175 of jurisdiction or vires”,93 narrowly defined. Finally, correctness review also applies automatically to “[q]uestions regarding the jurisdictional lines between two or more competing specialized tribunals”.94 In the years separating Dunsmuir from Wilson and Capilano, this framework had been fine-tuned in various ways, all tending in the direction of greater deference on questions of law. While Dunsmuir stated that “many legal issues attract a standard of correctness”,95 it would be difficult to make such a sweeping statement today. Since Dunsmuir, the availability of correctness review has been narrowed by: a growing presumption of deference, untethered to contextual factors such as the existence of a privative clause or a specialized, expert administrative regime; a narrow reading of the four categories identified in Dunsmuir 88 Ibid at para 56. 89 Ibid at para 51. 90 Ibid at para 55. 91 Ibid at para 60. 92 Ibid at para 58. 93 Ibid at para 59. “Jurisdiction is intended in the narrow sense of whether or not the tribunal had the authority to make the inquiry. In other words, true jurisdiction questions arise where the tribunal must explicitly determine whether its statutory grant of power gives it the authority to decide a particular matter. The tribunal must interpret the grant of authority correctly or its action will be found to be ultra vires or to constitute a wrongful decline of jurisdiction”. 94 Ibid at para 61. 95 Ibid at para 51. 302 UNBLJ RD UN-B [VOL/TOME 68 as always requiring correctness review; and a tendency to treat those four categories as exhausting the possibility of review for correctness. On the first of these three trends, there is now a broad starting presumption of deference, without the necessity of first showing the existence of a “discrete and special administrative regime in which the decision maker has special expertise”.96 The extent to which a court should determine whether a specialized regime has in fact been created, and should look for legislative indicators of expertise, was a point of disagreement between the majority and dissent in Capilano. As noted earlier, the majority assumed expertise on the part of the board, based on the fact that the board had been created and given statutory authority in the first place, and the fact that it was interpreting its home statute. Dunsmuir’s assessment of adjudicators under the Public Service Labour Relations Act as possessing expertise regarding their home and related statutes was broadened to apply to administrative decision makers more generally. 97 The dissent, on the other hand, found it unlikely that decision makers appointed to deal with a “vast array of municipal government issues”98 would be more expert than the courts in statutory interpretation. 2017 CanLIIDocs 175 A leaning toward deference can also be seen in the way the four Dunsmuir categories calling for automatic correctness review have been kept within narrow boundaries. Regarding jurisdictional questions, the Supreme Court has been warning itself (and others) since the time of CUPE v NB Liquor Corporation not to be “alert to brand” issues as jurisdictional,99 a warning reiterated by Justices Bastarache and LeBel in Dunsmuir. This was taken one step further some years after Dunsmuir, when Justice Rothstien, in Alberta (Information and Privacy Commissioner) v Alberta Teachers' Association, questioned the very concept of jurisdictional questions. 100 A second category of automatic correctness named in Dunsmuir is, by its nature, unlikely to occur frequently. Presumably it is fairly rare for cases to turn on the jurisdictional line between two expert tribunals. Further, to the extent that this line relates to division of powers issues (for instance, discerning whether a matter should be dealt with by the federal or a provincial Labour Relations Board), then this class of legal questions would collapse into a third Dunsmuir category – that of constitutional questions. As a category, constitutional issues might seem least susceptible to either contraction or expansion – either a question is constitutional in nature or it isn’t. However, in Doré, the Court applied a reasonableness analysis to the question of whether an administrative decision infringed Charter rights, thus limiting the constitutional issues which automatically attract correctness review to 96 Ibid at para 55. See also Capilano, supra note 3 at 22, and Van Harten et al, supra note 6 at 675. 97 Dunsmuir, supra note 5 at para 68. 98 Wilson, supra note 2 at para 86. 99 [1979] 2 SCR 227 at 233, 25 NBR (2d) 237. 100 Alberta Teachers’ Association, supra note 42. While not the focus on this paper, I would note that a lively debate can be found in the literature as to whether the category of jurisdictional questions plays any real role in today in the judicial oversight of administrative action. In our analysis of the federal courts’ application of the Dunsmuir framework (see supra note 80), we found that in 13 of the 216 federal court cases we reviewed, a question of law was identified as jurisdictional. 2017] SOME INITIAL THOUGHTS ON WILSON AND CAPILANO 303 constitutional challenges to the enabling legislation itself. The final category (general questions of law of importance to the legal system as a whole and beyond the expertise of the decision-maker) has been used by the Supreme Court only twice since Dunsmuir. In Saguenay101 the majority of the Court held that the parameters of the state’s duty of neutrality vis-à-vis religion could be classified as such, as did the majority in University of Calgary, regarding the issue of whether Alberta’s Freedom of Information and Protection of Privacy Act102 “allows solicitor-client privilege to be set aside”.103 Issues which the Court has refused to classify in this way include: a labour arbitrator’s interpretation of a management rights clause in a collective agreement, in the context of an attempt to impose mandatory alcohol testing;104 a labour arbitrator’s interpretation of the concept of estoppel;105 the question of whether a human rights tribunal had the power to grant legal costs to a party; 106 and most recently, a labour arbitrator’s rulings on evidentiary issues which were intertwined with concerns about deliberative secrecy.107 In this last decision, Justice Gascon, writing for the majority, stated that “questions of this nature are rare and tend to be limited to situations that are detrimental to ‘consistency in the fundamental legal order of our country’”.108 2017 CanLIIDocs 175 Both because of the nature of the Dunsmuir categories, and the way in which they have been interpreted subsequently by the Supreme Court, most questions of law that come before the courts are not going to be placed in a class which requires automatic application of the correctness standard. The more pressing question, then, for discerning the potential scope of correctness review, is whether or not the four Dunsmuir categories represent the only instances where a court can deviate from deference. Nothing in Dunsmuir itself limited correctness review of questions of law to the four named categories. Quite the opposite in fact. As noted above, it was taken as a given that “many legal issues attract a standard of correctness”.109 Further, 101 Saguenay, supra note 42 at para 46. 102 RSA 2000, c F-25. 103 Supra note 64 at para 20. 104 Communications, Energy and Paperworkers Union of Canada, Local 30 v Irving Pulp & Paper, Ltd, 2013 SCC 34, [2013] 2 SCR 458. 105 Nor-Man Regional Health Authority Inc v Manitoba Association of Health Care Professionals, 2011 SCC 59, [2011] 3 SCR 616. 106 Canada (Canadian Human Rights Commission) v Canada (AG), 2011 SCC 53, [2011] 3 SCR 471. 107 Commission scolaire de Laval v Syndicat de l’enseignement de la région de Laval, 2016 SCC 8, [2016] 1 SCR 29. 108 Ibid at para 34, citing supra note 106 at para 22. Justice Côté, writing a minority decision, warned against interpreting this category of legal issues “too narrow[ly]” (at para 78). 304 UNBLJ RD UN-B [VOL/TOME 68 Justices Bastarache and LeBel directed courts to look to relevant precedents, and it cannot be the case that precedent is relevant only where the earlier case selected the reasonableness standard. Finally, if standard of review could be determined solely by determining whether or not the issue in dispute falls into one of the four categories of automatic correctness review, there would be no need to retain the four-part standard of review analysis. The Court in Dunsmuir did, however, retain it, and instructed courts to engage in this analysis where the case law has not “already determined in a satisfactory manner the degree of deference to be accorded with regard to a particular category of question”.110 Despite these indications in Dunsmuir that correctness review could apply more broadly, subsequently there has been “a marked tendency”111 by the Supreme Court to treat the four enumerated categories as exhausting the scope of correctness review. In Wilson, Justice Abella suggested formalizing this tendency, as a back-up option should her colleagues be unpersuaded by arguments to move to a single deferential standard of review. While there was no explicit uptake on this proposal in either Wilson or Capilano, arguably the majority in Capilano came very close to accepting implicitly that the four categories of automatic correctness are exhaustive, 2017 CanLIIDocs 175 given its reluctance to engage in contextual analysis regarding what the legislature might have intended in terms of deference. In Dunsmuir, Justices Bastarache and LeBel summarized the process of identifying the appropriate standard of review as involving two steps: First, courts ascertain whether the jurisprudence has already determined in a satisfactory manner the degree of deference to be accorded with regard to a particular category of question. Second, where the first inquiry proves unfruitful, courts must proceed to an analysis of the factors making it possible to identify the proper standard of review.112 The analysis at the second stage “must be contextual”,113 and is to be based on the four-part pragmatic and functional analysis, renamed as the standard of review analysis, although it may not always be necessary to consider all four factors. However, in Capilano, Justice Karakatsanis chastised the Appeal Court for engaging in a lengthy discussion of contextual matters. In her view, given the prior Supreme Court jurisprudence applying the reasonableness standard on statutory appeals: 109 Supra note 95 at para 51. As David Jones notes: “The judges in Dunsmuir clearly contemplated that there were other circumstances in which correctness would be the appropriate standard of review, beyond the four categories they specifically identified as examples. That is why they referred to the need for an analysis of the context—which was perfectly in line functionally with the previous jurisprudence about a pragmatic and functional approach for determining the nature of the issue and legislative intent about the standard of review” (Jones, “Administrative Law Part II”, supra note 82 at 12). 110 Dunsmuir, supra note 5 at para 62. 111 Supra note 82 at 12. 112 Dunsmuir, supra note 5 at para 62. 113 Ibid at para 64. 2017] SOME INITIAL THOUGHTS ON WILSON AND CAPILANO 305 …there was no need for the Court of Appeal to engage in a long and detailed contextual analysis. Inevitably, the result would have been the same as in those cases. The presumption of reasonableness is not rebutted. … I would add this comment. The contextual approach can generate uncertainty and endless litigation concerning the standard of review. 114 The dissent in Capilano argued strenuously that correctness review can apply outside the four Dunsmuir categories, where there are sufficient indicators that the legislature intended this, particularly when combined with other contextual factors such as the need for consistency among various local boards interpreting the same legislation. Adopting the dissent’s approach would not create more categories of automatic correctness; the dissent in Capilano were clear that they had no interest in returning to pre-Pezim days,115 where a right of appeal led inevitably to correctness review. However, eschewing all focus on context, and thus precluding any inquiry into legislative intent, or the impact of the standard of review chosen on the administrative scheme as a whole, seems at odds with the very purpose of judicial oversight of administrative action. Legislative intention may sometimes indicate a 2017 CanLIIDocs 175 preference for strict scrutiny on other questions of law outside the four Dunsmuir categories, including where the particular wording of an appeal section invites such scrutiny, where the administrative body has little expertise in statutory interpretation, or where significant inconsistency of outcome would undermine the administrative scheme involved.116 A rigid system of classification which dictates whether or not an administrative decision is owed deference based solely on the box within which the issue is seen as fitting would not only run the risk of ignoring legislative intent, but would also mark a surprising return to the “categorical and nominate”117 approach, which was has been criticized for over two decades. 114 Capilano, supra note 3 at paras 34–35. 115 Ibid at para 70–77. 116 Inconsistency of decisions, by itself, is unlikely to justify correctness review: the inconsistency may be on a minor matter, or the administrative scheme may have its own methods for ensuring consistency over time (for example, as in IWA v Consolidated-Bathurst Packaging Ltd, [1990] 1 SCR 282, 73 OR (2d) 676). But where there is no internal mechanism for ultimately achieving agreement on the meaning of core provisions, or where alternate interpretations of these provisions will provide significantly different rights for individuals in the same situation, then the need for consistency does seem to add weight to any arguments for correctness review based on legislative intent. Or, to bring arguments about consistency within the rubric of legislative intent, perhaps it could be argued that on issues of such significance, the failure to establish an internal method for ensuring consistency must indicate that the legislature intended this function to be undertaken by the courts. 120 Dr Q v College of Physicians and Surgeons of British Columbia, 2003 SCC 19 at para 24, [2003] 1 SCR 226. See also Justice Côté’s comment on the “vice of formalism” in Capilano, supra note 3 at para 70. 306 UNBLJ RD UN-B [VOL/TOME 68 C. Application of reasonableness review in particular circumstances Thus far, I have argued that the possibility of correctness review should be retained and that its application should not be strictly limited to the four categories identified in Dunsmuir. This is not an attempt to undermine a general norm of deference for review of administrative action, but to say that there may be times when that norm should be set aside so as to give weight to legislative intent, or to serve other aims of the legal system, such as providing consistency on legal questions of significant import. If deference is to remain as the general norm, we will, however, have to become used to the prospect of courts sometimes identifying reasonableness as the appropriate standard, then performing their own analysis of the question in dispute, rather than simply asking whether the impugned decision is reasonable. Here, I am not speaking of instances where an overly-interfering court pays only lip service to deference, but times when what looks like correctness in the “guise”118 of reasonableness may be caused not by overly interventionist tendencies on the part of the reviewing judge, but by complexities inherent in the widespread use of the 2017 CanLIIDocs 175 reasonableness standard. According to Dunsmuir, [i]n judicial review, reasonableness is concerned mostly with the existence of justification, transparency and intelligibility within the decision-making process. But it is also concerned with whether the decision falls within a range of possible, acceptable outcomes which are defensible in respect of the facts and law.119 Yet, as highlighted by Wilson and Capilano, there are at least two scenarios where it would be difficult to apply the Dunsmuir understanding of reasonableness. The first is reasonableness review where a decision turns on a choice between two possible interpretations of the legislation in question, and the court determines that only one such interpretation can withstand even deferential review. The notion of a range of defensible outcomes seems to indicate that there will in fact be a range, which creates a dilemma, at least conceptually, when the court concludes there is only one right answer to be chosen from two possible alternatives. The second scenario is where the administrative decision maker has provided no reasons for its decision, thus making it difficult to review administrative action for justification, transparency and intelligibility. These challenges are examined in the next two sections. 118 Jones & de Villars, supra note 7 at 821. 119 Dunsmuir, supra note 5 at para 47. 2017] SOME INITIAL THOUGHTS ON WILSON AND CAPILANO 307 Reasonableness and choosing definitively between two possible interpretations Although it is now axiomatic that there is only one reasonableness standard as opposed to a sliding scale, reasonableness still “takes its colour from the context”, 120 and so there may be a wider or narrower range of outcomes reasonably available to the decision-maker. In fact, sometimes there may be only one outcome that a reviewing court will countenance as “defensible in respect of the facts and law”.121 This contrasts with cases such as CUPE 1979, where the Supreme Court accepted that there were numerous ways in which the disputed provisions of the Public Service Labour Relations Act122 (described by Justice Dickson as “very badly drafted” and “bristl[ing] with ambiguities”123) might be interpreted.124 The Court in CUPE accepted that the interpretation chosen by the Public Service Labour Relations Board was not patently unreasonable, but went on to say, “[t]he ambiguity of s.102(3)(a) is acknowledged and undoubted. There is no one interpretation which can be said to be ‘right’.”125 Even where “only a single defensible answer is available”,126 other aspects of the standard of review analysis may well indicate a legislative intention that 2017 CanLIIDocs 175 deference be accorded the administrative decision-maker. In that situation, however, the reviewing court may find itself writing a decision that sounds very much like correctness review. Unless the court wishes, in essence, to answer the question before it with a shrug - an approach hardly likely to enhance confidence in either the administrative state or the judicial system– the court may well end up indicating which of the two possible, and incompatible, interpretations is reasonable, and which is not. In Wilson, the only two interpretations offered for the disputed provisions of the Canada Labour Code127 were either that non-unionized employees could be dismissed without just cause, or that they could not be. Likewise, in Capilano, the 120 Khosa, supra note 18 at para 59. 121 Dunsmuir, supra note 5 at para 47. Which of course sounds very like the traditional understanding of correctness review where “[t]here can be only a ‘single right answer’ to the questions under review” (Régimbald, supra note 67 at 423, referring to Law Society of New Brunswick v Ryan, 2003 SCC 20, [2003] 1 SCR 247 at para 51). 122 RSNB 1973, c P-25. 123 CUPE, supra note 10 at 230. 124 Ibid at 237. Justice Dickson noted: “Mr. Justice Limerick of the New Brunswick Appeal Division, in the course of his reasons in the present litigation, said: ‘Four possible interpretations immediately come to mind’”. 125 Ibid. 126 Wilson, supra note 2 at para 23. 127 RSC 1985, c L-2. 308 UNBLJ RD UN-B [VOL/TOME 68 relevant section of the Municipal Government Act either allowed an assessment review board to increase the tax assessment on a taxpayer appeal, or it did not. The majority in Wilson, applying a standard of reasonableness, did not explicitly reject an interpretation of the Canada Labour Code allowing employees to be terminated with reasonable notice; however, the majority concluded that: [t]he text [of the Code], the statements of the Minister when the legislation was introduced, and the views of the overwhelming majority of arbitral and labour law scholars, confirm that the entire purpose of the statutory scheme was to ensure that non-unionized federal employees would be entitled to the protection from being dismissed without cause”.128 In the face of such unequivocal language, it would be a foolhardy adjudicator who, in the future, decided to apply the alternate interpretation. Similarly, the majority in Capilano not only found it reasonable for the board to interpret its authority as allowing for an increase in the assessment rate on a taxpayer appeal, but it also stated that the alternate interpretation would frustrate the purpose of Act. 129 Again, there seems little room for an assessment review board to delineate its powers differently in the future. 2017 CanLIIDocs 175 While both cases provide much-needed consistency on the questions before the Court, they also illustrate the difficulty of performing traditional reasonableness review in certain circumstances. David Jones’ intriguing question in the context of another Supreme Court decision seems equally relevant to Wilson and Capilano: “At what point in time is the reasonableness of a decision determined—before the court determines the applicable standard of review, or only after the applicable standard of review has been determined?”130 Reviewing the reasonableness of non-existent reasons The Dunsmuir approach to reasonableness may also pose a challenge where there are no reasons to be reviewed for “justification, transparency and intelligibility”.131 In Dunsmuir, Justices Bastarache and LeBel were alert to the possibility that an administrative decision-maker might not have provided reasons for the issue in dispute. While offering assurances that the merger of patent unreasonableness and reasonableness simplicter into one standard would not promote greater judicial intervention, they quoted with approval Professor David Dyzenhaus’ description of deference as “a respectful attention to the reasons offered or which could be offered in support of a decision”.132 The possibility of supplementing the reasons of an 128 Wilson, supra note 2 at para 39. 129 Capilano, supra note 3 at para 61. 130 Jones, “Administrative Law In 2016”, supra note 6 at 14. 131 Dunsmuir, supra note 5 at para 47. 132 Ibid at para 48, citing David Dyzenhaus, “The Politics of Deference: Judicial Review and Democracy” in M Taggart, ed, The Province of Administrative Law (Oxford: Hart Publishing, 1997) 279 at 286. 2017] SOME INITIAL THOUGHTS ON WILSON AND CAPILANO 309 administrative body received unanimous support from the Supreme Court in Newfoundland Nurses’ Union, which added the gloss that the two aspects of reasonableness described in Dunsmuir were not to be treated as two separate tests but rather as an organic whole.133 However, unless this means that either reasoning that offers justification and is transparent and intelligible or an outcome that comes within a range of acceptable outcomes is, by itself, sufficient to pass the reasonableness standard, the Newfoundland Nurses’ Union gloss offers little assistance where the decision under review does not provide any reasons on the issue in dispute. In Alberta Teachers’ Association, the Court was faced with reviewing an issue raised for the first time on judicial review. Justice Rothstein offered the following advice for a reviewing court in this situation: Obviously, where the tribunal’s decision is implicit, the reviewing court cannot refer to the tribunal’s process of articulating reasons, nor to justification, transparency and intelligibility within the tribunal’s decision- making process. … 2017 CanLIIDocs 175 However, … when the decision concerns an issue that was not raised before the decision maker … [i]f there exists a reasonable basis upon which the decision maker could have decided as it did, the court must not interfere. …. [D]eference under the reasonableness standard is best given effect when administrative decision makers provide intelligible and transparent justification for their decisions, and when courts ground their review of the decision in the reasons provided. … [But] parties cannot gut the deference owed to a tribunal by failing to raise the issue before the tribunal and thereby mislead the tribunal on the necessity of providing reasons.134 Justice Statas of the Federal Court of Appeal has been scathing about the Dunsmuir direction that courts should first seek to supplement, rather than subvert, an administrative decision-maker’s reasoning, calling it “a rule that has been decreed, not deduced from an underlying doctrinal concept” and noting that “[w]ithout a coherent underlying concept to guide this rule, no one knows its limits or when or how it should be applied”.135 Perhaps rather than continuing the fiction of “supplementing”, where reasons are non-existent, it should simply be acknowledged Stratas, supra note 4 at 38 argues that the Supreme Court in Dunsmuir “adopted this rule [of supplementing reasons] on the basis of a quote plucked out of context from a single academic article, that, if read in its entirely, deals with another subject entirely, and, in fact, advocates something quite different”. 133 Newfoundland Nurses’ Union, supra note 20 at para 14. 134 Alberta Teachers’ Association, supra note 42 at paras 52–54. 135 Stratas, supra note 4 at 38. 310 UNBLJ RD UN-B [VOL/TOME 68 that, even if precedent or the standard of review analysis indicates deferential review, the court will be forced to undertake its own analysis. In Capilano, at the hearing before the assessment review board, Edmonton East’s counsel acknowledged that the board did have the authority to raise an assessment on a taxpayer appeal. Therefore, as Justice Karakatsanis noted, “it is hardly surprising the Board did not explain why it was of the view that it could increase the assessment: the Company expressly conceded the point”.136 Yet as Justice Karakatsanis also noted, “[w]hen a tribunal does not give reasons, it makes the task of determining the justification and intelligibility of the decision more challenging.”137 The majority considered the reasons which the board might have offered for its interpretation by examining the ordinary meaning of the section under review, how the section was interpreted by a predecessor body, the purpose of the Act, and general arguments about equity. There was little else the Court could do in the absence of reasons from the Board. There seems little point in requiring a decision-maker to provide reasons not only for the questions that are clearly before it, but also for those questions which do 2017 CanLIIDocs 175 not appear to be live issues. Yet, without an explanation from the assessment review board as to why it interpreted its authority as it did, it was difficult for the Court to review for the very thing that Dunsmuir tells us is the primary concern of a court engaged in determining whether a decision is reasonable or not: “the existence of justification, transparency and intelligibility within the decision-making process.” In such a situation, a court will be forced to follow its own process of reasoning. My point here is that this should not be interpreted as a covert attempt to expand correctness review beyond its appropriate purview; instead it is the inevitable outcome of a system of judicial oversight where deference is the norm. Conclusion Should those of us who are interested in standard of review analysis be holding our collective breath, waiting for the Supreme Court to embark on a sea change of the magnitude wrought by Dunsmuir? A complete recalibration of Dunsmuir seems unlikely, given the lack of uptake on Justice Abella’s suggestions in this regard. On the other hand, in light of the significant differences of opinion on the Court regarding key elements of the role of courts in overseeing administrative decision- making, we are likely to witness more lively exchanges on the availability and scope of correctness review; future decisions may also provide further commentary on how courts should go about reasonableness review where there are only two possible interpretations of a legislative provision or where no reasons were given by the body under review. 136 Capilano, supra note 2 at para 40. 137 Ibid at para 36. 2017] SOME INITIAL THOUGHTS ON WILSON AND CAPILANO 311 In this article, I have argued that while Dunsmuir makes deference the norm, there are no compelling reasons, from either a principled or pragmatic standpoint, for eliminating the possibility of correctness review for some questions of law. The dissents in Wilson and Capilano, and the majority decision in University of Calgary indicate that correctness review has a future. However, University of Calgary based its choice of the correctness standard on classifying the issue in dispute as a question of general law, of central importance to the legal system as a whole and outside the expertise of the administrative decision maker. This raises the question of whether the trend, visible in the majority decision in Capilano, of limiting correctness review to the four categories of legal questions identified in Dunsmuir, will continue to gain momentum or whether on occasion, the Supreme Court will adopt the reasoning of the dissent, such that correctness can on occasion be used outside those four categories. Of course, even if the Dunsmuir categories of automatic correctness are not treated as exhausting the possibility of correctness review, Dunsmuir indicates that deference should be the predominate approach to reviewing administrative action. Therefore, we will have to get used to the idea that sometimes the application of the reasonableness standard will look singularly like correctness review. Where this occurs because the court feels compelled to choose 2017 CanLIIDocs 175 between two possible interpretations of the legislative provision in dispute, or because the decision-maker has not provided reasons on a key issue, courts’ reliance on their own reasoning arises less from inappropriate interventionism and more from the difficulties of applying the Dunsmuir understanding of reasonableness in those contexts. REVISITING THE APPLICATION OF SECTION 7 OF THE CHARTER IN IMMIGRATION AND REFUGEE PROTECTION Gerald Heckman* I. Introduction Section 7 of the Canadian Charter of Rights and Freedoms states that “everyone has the right to life, liberty and security of the person and the right not to be deprived thereof except in accordance with the principles of fundamental justice.”1 To establish an infringement of s. 7, it is necessary to establish first, that state action has resulted in depriving an individual of their life, liberty or security of the person and second, that this deprivation was achieved in a manner inconsistent with one or more principles of fundamental justice. This article focuses on the first step of the analysis: whether proceedings under Canada’s immigration and refugee protection laws 2017 CanLIIDocs 175 engage the life, liberty or security of the person of non-citizens. The very first case decided by the Supreme Court of Canada involving a s. 7 claim outside of the criminal context was Singh v Canada (Minister of Employment and Immigration).2 Three of six judges recognized that the denial of a Convention refugee’s right under the Immigration Act, 1976 not to be removed from Canada to a country where his life or freedom would be threatened amounted to a deprivation of his security of the person within the meaning of s. 7. Justice Bertha Wilson’s judgment in Singh is remarkable in several ways.3 It established that the word “everyone” in s. 7 applies to “every human being physically present in Canada and by virtue of such presence amenable to Canadian law.” It recognized that “security of the person” encompasses not only freedom from physical punishment or suffering but freedom from the threat of such punishment. In determining whether s. 7 of the Charter applied to government acts, it refused to embrace the distinction between acts said to impact “rights” and those affecting “mere privileges”, focusing instead on their consequences for the affected person’s s. 7 interests. * Associate Professor, Faculty of Law, University of Manitoba. I thank my colleagues Professors Audrey Macklin and Colin Grey for their comments on an earlier draft of this manuscript. I also thank the anonymous reviewers and the editorial staff of the University of New Brunswick Law Journal for their efforts in preparing this article for publication. All errors and omissions are mine alone. 1 Canadian Charter of Rights and Freedoms, Part I of the Constitution Act, 1982, being Schedule B to the Canada Act 1982 (UK), 1982, c 11 [Charter]. 2 Singh v Canada (Minister of Employment and Immigration), [1985] 1 SCR 177, 17 DLR 4th 422 [Singh]. 3 See generally Catherine Dauvergne, “How the Charter Has Failed Non-Citizens in Canada: Reviewing Thirty Years of Supreme Court of Canada Jurisprudence” (2013) 58 McGill LJ 663 at 668–671 [Dauvergne, “How the Charter Has Failed”]. 2017] REVISITING THE APPLICATION OF SECTION 7 313 Over thirty years later, significant questions remain about the application of s. 7 in the sphere of immigration and refugee law. Following over a decade of inconsistent Federal Court of Appeal decisions on whether the right to liberty was engaged by immigration and refugee proceedings, the Supreme Court laconically declared that “the deportation of a non-citizen in itself cannot implicate the liberty and security interests protected by s. 7 of the Canadian Charter of Rights and Freedoms.”4 Two years later, it adjusted its position, holding that “[w]hile the deportation of a non-citizen in the immigration context may not in itself engage s. 7 of the Charter, some features associated with deportation, such as detention in the course of the [security] certificate process or the prospect of deportation to torture, may do so.”5 More recently, the Supreme Court appeared to endorse, in obiter, the view that liberty and security of the person are not engaged in the earlier decision making stages of the immigration and refugee protection regime so long as these interests can be considered in proceedings that immediately precede removal.6 Thus, s. 7 was not engaged at the stage of determining whether a refugee claimant was inadmissible to Canada because the claimant had access to a subsequent pre-removal risk assessment where the risk of removal to face death, torture or cruel and unusual 7 treatment or punishment would be considered and s. 7 would be engaged. In a 2017 CanLIIDocs 175 recent decision, the Federal Court of Appeal held that to decide whether a bar on pre- removal risk assessments filed within a year of the rejection of a refugee protection claim violated a non-citizen’s s. 7 right to security of the person, it would be necessary to revisit the reasoning underlying the thirty-year-old Singh judgment.8 In this article, I provide a brief and general overview of the Supreme Court’s jurisprudence on the application of s. 7. Against this background, I take stock of and critically assess, in historical context, the current state of the law on the engagement of liberty and security of the person in immigration and refugee proceedings. I conclude that in the refugee and immigration context, Canadian courts have adopted a narrow approach to the engagement of s. 7 that is inconsistent with their approach in the cognate areas of criminal law and extradition law and for which they have failed to articulate a transparent and principled justification. I examine four aspects of the s. 7 engagement jurisprudence that illustrate this inconsistency and cry out for a principled reappraisal by the Supreme Court. 4 Medovarksi v Canada (Minister of Citizenship and Immigration); Esteban v Canada (Minister of Citizenship and Immigration), 2005 SCC 51 at para 46, [2005] 2 SCR 539 [Medovarksi]. 5 Charkaoui v Canada (Citizenship and Immigration), 2007 SCC 9 at para 17, [2007] 1 SCR 351 [Charkaoui]. 6 B010 v Canada (Citizenship and Immigration), 2015 SCC 58, 390 DLR (4th) 385 [B010]. 7 Ibid at para 75. 8 Peter v Canada (Minister of Public Safety and Emergency Preparedness); Savunthararasa v Canada (Minister of Public Safety and Emergency Preparedness), 2016 FCA 51 at paras 28–29, 395 DLR (4th) 758 [Savunthararasa]. 314 UNBLJ RD UN-B [VOL/TOME 68 First, the Federal Court of Appeal’s inconsistent early jurisprudence on whether s. 7 is engaged in immigration and refugee proceedings was marked by persistent confusion between two distinct components of s. 7 analysis: whether these proceedings engage non-citizens’ life, liberty and security of the person and whether, viewed in their statutory context, they offend principles of fundamental justice. The Federal Court and Federal Court of Appeal have continued to rely on some of these early decisions, which are ripe for re-examination. Second, the Supreme Court’s bald assertion in Medovarski that the deportation of non-citizens does not, in itself, implicate their liberty and security of the person fails to address key arguments, some grounded in the Court’s own s. 7 jurisprudence, that support s. 7 engagement. As recognized in some Federal Court of Appeal judgments, immigration and refugee protection proceedings involve the threat of detention incidental to forced removal. The possibility of detention engages liberty in the extradition and penal contexts which, with the advent of “crimmigration” – the convergence of criminal law and immigration law – are not far removed from the context of removal proceedings. Deportation can also engage non-citizens’ liberty by preventing them from making fundamental personal choices, such as nurturing or caring for their Canadian-born children, that go beyond the bare assertion of mobility rights. Interference with such 2017 CanLIIDocs 175 profoundly intimate choices could also produce an effect on non-citizens’ psychological integrity serious and profound enough to engage their security of the person. Third, more than thirty years after Singh, uncertainty persists on whether non-citizens’ security of the person is engaged in any circumstance where deportation places them at risk of persecution, torture or cruel and unusual punishment or whether s. 7 engagement hinges on non-citizens’ ability to establish a violation of their statutory rights. While the Supreme Court has hinted in some of its judgments that non-citizens’ right not to be deprived of their security of the person except in accordance with the principles of fundamental justice is a freestanding constitutional right, it has not yet expressly and unequivocally addressed this fundamental aspect of s. 7 engagement. Finally, by opining that s. 7 does not apply to determinations of exclusion or inadmissibility because these proceedings are not sufficiently proximate to removal, the Supreme Court has without justification imposed in the immigration and refugee protection context a standard of causation more onerous than that which it applies for s. 7 engagement generally. In my concluding remarks, I briefly address how a principled approach to s. 7 engagement in immigration and refugee protection decision making could make a real difference for non-citizens who seek to challenge their removal from Canada. Abandoning the narrow approach to s. 7 engagement in this context would shift the courts’ focus to the crucial question of whether the state has interfered with non- citizens’ liberty and security of the person in a manner rationally connected and proportionate to the objectives of Canada’s immigration laws, as required by our fundamental constitutional values. 2017] REVISITING THE APPLICATION OF SECTION 7 315 II. The scope of application of section 7 of the Charter: a brief overview To demonstrate a violation of s. 7, one must establish, first, that a law or state action interferes with or deprives natural persons9 present in Canada and thus subject to Canadian law10 of their life, liberty or security of the person11 and, second, that this deprivation is not in accordance with the principles of fundamental justice.12 This part summarizes the state of the law on the question of whether s. 7 is engaged, the first of these issues, through a review of leading cases decided in a variety of contexts touching on the administration of justice. 1. State action implicating the administration of justice The dominant strand of jurisprudence on s. 7 sees its purpose as guarding against the kinds of deprivation of life, liberty and security of the person “that occur as a result of an individual’s interaction with the justice system and its administration,”13 a term which refers to “the state’s conduct in the course of enforcing and securing 14 compliance with the law.” Thus, s. 7 protects against measures that can be 2017 CanLIIDocs 175 attributed to state action implicating the “administration of justice,”15 broadly interpreted by the Court as extending beyond processes operating in the criminal law sphere16 to the investigation of complaints of discrimination under human rights legislation,17 parental rights in relation to state-imposed medical treatment18 and in 9 Irwin Toy Ltd v Québec (AG), [1989] 1 SCR 927 at 1002–3, 58 DLR (4th) 577. 10 Singh, supra note 2 at para 35. Exceptionally, the Charter may apply to the actions of state agents participating in activities of a foreign state or its agents that are contrary to Canada’s international obligations: Canada (Justice) v Khadr, 2008 SCC 28 at para 18, [2008] 2 SCR 125. 11 Carter v Canada (AG), 2015 SCC 5 at para 55, [2015] 1 SCR 331 [Carter]. 12 Ibid at para 35. 13 Gosselin v Quebec (AG), 2002 SCC 84 at para 77, [2002] 4 SCR 429 [Gosselin]. 14 Ibid, citing New Brunswick (Minister of Health and Community Services) v G(J), [1999] 3 SCR 46 at 79, 216 NBR (2d) 25 [G(J)]. While some Supreme Court judges have expressly argued in favour of extending the application of s. 7 to contexts other than those linked to the administration of justice, this position has not in my view clearly been adopted by a majority of the Court: Chaoulli v Quebec (AG), 2005 SCC 35 at paras 195–199, [2005] 1 SCR 791 [Chaoulli] per Binnie, LeBel and Fish JJ; see Gerald Heckman, “Charte Canadienne: droit à la vie, à la liberté et à la sécurité de la personne et justice fondamentale” in Stéphane Beaulac & Jean-François Gaudreault-Desbiens, eds, Jurisclasseur Québec – Collection Droit Public – Droit Constitutionnel (Toronto: LexisNexis, 2015) at para 9. 15 Ibid. 16 Ibid at para 78. 17 Ibid, citing Blencoe v British Columbia (Human Rights Commission), 2000 SCC 44, [2000] 2 SCR 307 [Blencoe]. 18 B(R) v Children’s Aid Society of Metropolitan Toronto, [1995] 1 SCR 315, 21 OR (3d) 479 [B(R)]. 316 UNBLJ RD UN-B [VOL/TOME 68 the child custody process19 and the right to refuse state-imposed addiction treatment.20 Canada’s conduct in enforcing and securing non-citizens’ compliance with its immigration laws falls well within this concept of administration of justice. 2. Life, Liberty and Security of the Person (a) Life The right to life under s. 7 is engaged where a law or state action imposes death or an increased risk of death on a person, either directly or indirectly.21 State measures that interfere with patients’ timely access to potentially life-saving medical care have been found to engage the right to life.22 Deporting a refugee “where there are grounds to believe that this would subject the refugee to a substantial risk of torture” would also violate the guarantee of life, liberty and security of the person.23 2017 CanLIIDocs 175 (b) Liberty The right to liberty is engaged where the state subjects an individual to physical restraint or to the threat of physical restraint. An offense has the potential of depriving persons of their liberty and engages s. 7 “as of the moment it is open to the judge to impose imprisonment”: there is no need that imprisonment “be made mandatory.”24 Immigration detention, such as that provided under the Immigration and Refugee Protection Act25 for individuals designated by a security certificate, also engages liberty.26 The s. 7 liberty interest is no longer restricted to “mere freedom from physical restraint” but is engaged “where state compulsions or prohibitions affect 19 G(J), supra note 14. 20 Winnipeg Child and Family Services (Northwest Area) v DFG, [1997] 3 SCR 925, 121 Man R (2d) 241. 21 Carter, supra note 11 at para 62. 22 See Chaoulli, supra note 14 at para 62 (prohibition on the purchase of private health insurance) and Canada (AG) v PHS Community Services Society, 2011 SCC 44 at para 91, [2011] 3 SCR 134 [PHS] (measures preventing health professionals from offering medical supervision and counselling to their addicted clients at a safe injection site). 23 Suresh v Canada (Minister of Citizenship and Immigration), 2002 SCC 1 at para 129, [2002] 1 SCR 3 [Suresh]. 24 Reference re Motor Vehicle Act (British Columbia) s 94(2), [1985] 2 SCR 486 at 79, 24 DLR (4th) 536 [Motor Vehicle Reference]. 25 Immigration and Refugee Protection Act, SC 2001, c. 27 [IRPA]. 26 As the Supreme Court held in Charkaoui, supra note 5 at para 13: “The provisions at issue… clearly deprive detainees such as the appellants of their liberty. The person named in a certificate can face detention pending the outcome of the proceedings.” 2017] REVISITING THE APPLICATION OF SECTION 7 317 important and fundamental life choices.”27 For example, the Identification of Criminals Act, which provided for the fingerprinting of persons charged with but not convicted of an offense, engaged their right to liberty because it “require[d] a person to appear at a specific time and place and oblige[d] that person to go through an identification process on pain of imprisonment for failure to comply.”28 Similarly, the statutory power of an administrative tribunal to compel any person “to appear at a specific time and place to testify subject to legal consequences for failure to comply” constitutes a deprivation of liberty and engages s. 7 of the Charter. 29 The right to liberty “protects within its ambit the right to an irreducible sphere of personal autonomy wherein individuals may make inherently private choices free from state interference.”30 It does not guarantee unconstrained freedom nor protect any and all decisions that individuals may make in conducting their affairs. Rather, it encompasses only those matters “that can properly be characterized as fundamentally or inherently personal such that, by their very nature, they implicate basic choices going to the core of what it means to enjoy individual dignity and independence.”31 2017 CanLIIDocs 175 (c) Security of the person Security of the person “encompasses ‘a notion of personal autonomy involving… control of one’s bodily integrity free from state interference’ […] and it is engaged by state interference with an individual’s physical or psychological integrity, including any state action that causes physical or serious psychological suffering.”32 27 Blencoe, supra note 17 at para 49. 28 R v Beare; R v Higgins, [1988] 2 SCR 387 at 402, 71 Sask R 1 [Beare] cited in Blencoe, supra note 17 at para 49. 29 Thomson Newspapers Ltd. v Canada (Director of Investigation and Research, Restrictive Trade Practices Commission), [1990] 1 SCR 425 at 573 (L’Heureux-Dubé J), cited in Blencoe, supra note 17 at para 49 [Thomson]. The relevant statute gave the members of the Restrictive Trade Practices Commission the authority to order any person to be examined upon oath before a member and to exercise the powers of a superior court for the enforcement of subpoenas to witnesses “or punishment of disobedience thereof”: ibid, at para 24. 30 Godbout v Longeuil (City), [1997] 3 SCR 844 at para 66, 152 DLR (4th) 577 [Godbout] per La Forest J, writing for L’Heureux-Dubé and McLachlin JJ, cited with approval by a majority of the Court in Blencoe, supra note 17 at para 51. 31 Ibid. By denying individuals with grievous and irremediable medical conditions the right to request a physician’s assistance in dying, the Criminal Code’s prohibition on assisted suicide interfered “with their ability to make decisions concerning their bodily integrity and medical care and thus trenche[d] on liberty”: Carter, supra note 11 at para 66. Several Supreme Court judges (but not a majority) would have recognized that the right to liberty encompassed the right of parents to make decisions regarding the medical care provided to their children and a person’s right to choose where to establish his or her home: respectively, B(R), supra note 18 and Godbout, supra note 30 at paras 66–67. 32 Carter, supra note 11 at para 64 [case citations omitted]. 318 UNBLJ RD UN-B [VOL/TOME 68 Security of the person is engaged by state action that interferes with individuals’ physical integrity. Delays inherent in the Criminal Code regime governing the provision of therapeutic abortions increased the risk of medical complications and mortality and infringed the physical aspect of women’s right to security of the person.33 Similarly, legislation that prohibited patients from purchasing private medical insurance and forced them to accept delays in the public medical system denied them timely access to care “for a condition… clinically significant to their current and future health,” adversely impacted their physical and psychological health and engaged their security of the person.34 Security of the person encompasses “freedom from the threat of physical punishment or suffering as well as freedom from such punishment itself.”35 Accordingly, denying a Convention refugee the right under Canada’s Immigration Act, 1976 not to be removed to a country “where his life or freedom would be threatened” would amount to a deprivation of security of the person.36 State action that has a serious and profound effect on a person’s psychological integrity restricts that person’s security of the person. The effects of the state’s interference, assessed objectively “with a view to their impact on the 2017 CanLIIDocs 175 psychological integrity of a person of reasonable sensibility,”37 need not rise to the level of “nervous shock or psychiatric illness,” but must be greater than “ordinary stress or anxiety.”38 Security of the person will be violated only by serious psychological incursions resulting from state interference with an individual interest of fundamental importance or, in other words, the profoundly intimate and personal choices of an individual.39 Breaches of security of the person were found where the state interfered with a woman’s choice to end her pregnancy,40 a person’s choice to end her life41 and a parent’s interest in raising and caring for a child.42 Such 33 R v Morgentaler, [1988] 1 SCR 30 at 59, 63 OR (3d) 281 [Morgentaler]. 34 Chaoulli, supra note 14 at para 123 per McLachlin CJ and Major and Bastarache JJ and at paras 191, 203–6 per Binnie, LeBel and Fish JJ. By prohibiting medical marijuana users from choosing methods of administration of the drug other than smoking dry marijuana, Parliament breached their right to security of the person by subjecting them to the risk of cancer and bronchial infections and forcing them to choose between legal and inadequate treatment and an illegal but more effective choice: R. v Smith, 2015 SCC 34 at para 18, [2015] 2 SCR 602. Similarly, Criminal Code prohibitions on bawdy houses, living on the avails of prostitution and communicating in public for the purposes of prostitution engaged prostitutes’ security of the person by preventing them from taking steps to protect themselves from the risks inherent in prostitution, thereby heightening the risk of disease, violence and death: Canada (AG) v Bedford, 2013 SCC 72 at paras 60, 88, [2013] 3 SCR 1101 [Bedford]. 35 Singh, supra note 2 at 207 per Wilson J (Dickson CJ and Lamer J concurring). 36 Ibid. 37 G(J), supra note 14 at para 60. 38 Ibid. 39 Blencoe, supra note 17 at paras 82–83. 40 Morgentaler, supra note 33. 41 Carter, supra note 11 at paras 64–66. 42 G(J), supra note 14 at para 61. 2017] REVISITING THE APPLICATION OF SECTION 7 319 fundamental personal choices “would not easily include the type of stress, anxiety and stigma that result from administrative or civil proceedings.”43 3. Causation Section 7 is engaged only if a law or state action is the causal source of an interference with a rights claimant’s life, liberty or security of the person. The rights claimant must establish “a sufficient causal connection” between the state-caused effect and the prejudice suffered by the claimant: A sufficient causal connection standard does not require that the impugned government action or law be the only or the dominant cause of the prejudice suffered by the claimant, and is satisfied by a reasonable inference, drawn on a balance of probabilities… A sufficient causal connection is sensitive to the context of the particular case and insists on a real, as opposed to a speculative, link.44 In Bedford the Supreme Court rejected a higher standard of causation that would 2017 CanLIIDocs 175 have required the rights claimant to show that the state action was a foreseeable and necessary cause of the prejudice to the claimant’s security interest. In its view, a “sufficient causal connection” represented a fair and workable threshold for engaging s. 7 of the Charter: This is the port of entry for s. 7 claims. The claimant bears the burden of establishing this connection. Even if established, it does not end the inquiry, since the claimant must go on to show that the deprivation of her security of the person is not in accordance with the principles of fundamental justice. Although mere speculation will not suffice to establish causation, to set the bar too high risks barring meritorious claims. What is required is a sufficient connection, having regard to the context of the case.45 Over the Charter’s 35-year history, the Supreme Court has gradually eased the threshold for the engagement of s. 7. In particular, it has broadened the scope of liberty and security of the person and adopted a relatively low standard of causation. In the next part of my paper, I argue that in the immigration and refugee protection 43 Blencoe, supra note 17 at para 83. Despite the personal hardship endured by the respondent to a human rights complaint in the face of significant delay in a human rights commission’s investigation of the complaint, including the stigma associated with the complaint, the depletion of his financial resources and the associated physical and psychological suffering, there was no breach of security of the person because the state had not interfered with the respondent and his family’s ability to make essential life choices: ibid at para 86. 44 Bedford, supra note 34 at para 76. 45 Ibid at para 78. 320 UNBLJ RD UN-B [VOL/TOME 68 context, the Court has, without acknowledgement or justification, resiled from this more relaxed approach to the engagement of s. 7. III. The application of s. 7 in immigration and refugee law – a critical appraisal Against the background of the Court’s current approach to the engagement of s. 7, this part focuses on how courts have dealt with the question of whether liberty and security of the person are engaged in immigration and refugee protection proceedings and with the issue of causation. Each section begins with a review of the foundational cases that marked the evolution of the jurisprudence in this context, including early Federal Court and Federal Court of Appeal decisions. Some of these remain relevant today not only as historical context but because, unlike some recent decisions of the Supreme Court, they squarely address the impact of removal on non- citizens’ s. 7 interests. Indeed, Federal Court judges still return to some of these precedents when resolving claims of s. 7 engagement. 2017 CanLIIDocs 175 1. Liberty Early Federal Court and Federal Court of Appeal decisions were split on whether the removal of non-citizens engaged their liberty interest. One line of jurisprudence recognized that forcibly deporting someone against his will necessarily interfered with his liberty, while other decisions found no engagement of liberty. This was due in large measure to a misreading of Chiarelli v Canada (Minister of Employment and Immigration),46 a decision in which, as noted in the following section, the Supreme Court had expressly declined to address the question of s. 7 engagement. (a) Chiarelli v Canada (Minister of Employment and Immigration) In Chiarelli, the Supreme Court first considered, without deciding, whether the right to liberty is engaged in proceedings leading to non-citizens’ removal from Canada. To this day, the judgment casts a long shadow over s. 7 jurisprudence in the context of immigration and refugee law.47 When Chiarelli, a permanent resident who arrived in Canada at age 15, was convicted of a serious criminal offense, a deportation order was issued against him. He asked the Immigration Appeal Board to set aside the order on humanitarian and compassionate grounds. Before the Board could hear the appeal, the Security Intelligence Review Committee (SIRC), at the behest of the Canadian government, inquired into whether Chiarelli was likely to engage in organized crime if allowed to remain in Canada. Relying on the SIRC’s findings, the 46 Chiarelli v Canada (Minister of Employment & Immigration), [1992] 1 SCR 711, 90 DLR (4th) 289 [Chiarelli]. 47 See, generally, Dauvergne, “How the Charter Has Failed”, supra note 3 at 680–2. See also Catherine Dauvergne, Humanitarianism, Identity and Nation: Migration Laws of Australia and Canada (Vancouver: UBC Press, 2005) at 202ff [Dauvergne, Humanitarianism, Identity and Nation]. 2017] REVISITING THE APPLICATION OF SECTION 7 321 Minister of Employment and Immigration issued a certificate with the result that Chiarelli’s appeal on humanitarian and compassionate grounds was dismissed. Chiarelli contested the constitutionality of the scheme on several grounds, including that reliance upon the certificate deprived him of his liberty under s. 7 through a process that did not accord with fundamental justice.48 All three Federal Court of Appeal judges agreed that s. 7 was engaged: The filing of the certificate had the effect of depriving the Immigration Appeal Board of its power to allow the appellant’s appeal on compassionate grounds. This, in itself, did not directly interfere with the appellant’s right to life, liberty and security of the person. However, if things are looked at realistically, it cannot be denied that, as a result of the filing of the certificate, the appellant will be deported to Italy while he otherwise might have been allowed to remain in the country. As, in my view, deportation necessarily implies an interference with the liberty of the person, I would say that a violation of section 7 of the Charter has been established.49 2017 CanLIIDocs 175 The Supreme Court declined to decide “whether deportation per se engages s. 7, that is, whether it amounts to a deprivation of life, liberty or security of the person” because it found no breach of the principles of fundamental justice.50 To determine the scope of these principles, Justice Sopinka adopted a contextual approach to Charter interpretation and looked to the principles and policies underlying immigration law including “the most fundamental principle of immigration law… that non-citizens do not have an unqualified right to enter or remain in the country.”51 The requirement that permanent residents not be convicted of a serious criminal offence was a “legitimate, non-arbitrary choice by Parliament of a situation in which it is not in the public interest to allow a non-citizen to remain in the country.”52 Where non-citizens deliberately violated essential conditions under which they could remain in Canada, giving practical effect, through deportation, to the termination of their right to remain did not breach fundamental justice,53 nor did 48 The SIRC hearing was held, in part, in camera and in the absence of Chiarelli and his counsel. 49 Chiarelli v Canada (Minister of Employment and Immigration), [1990] 2 FCR 299 at 318–19, 67 DLR (4th) 697 [Chiarelli FCA] per Pratte JA, dissenting but not on this point. Stone and Urie JJA agreed with Pratte JA on the engagement of s. 7 and with his finding that the SIRC hearing did not accord with the principles of fundamental justice but disagreed with his conclusion that the breach of s. 7 was justified under s. 1 of the Charter. 50 Chiarelli, supra note 46 at 731–2. 51 Ibid at 733. 52 Ibid at 734. 53 Ibid. The Court stated that it was “not necessary, in order to comply with fundamental justice, to look beyond this fact [of a deliberate violation of the prohibition on committing serious crimes] to other aggravating or mitigating circumstances.” 322 UNBLJ RD UN-B [VOL/TOME 68 the absence of a compassionate appeal from the deportation order.54 Finally, the SIRC procedure did not violate principles of fundamental justice.55 Thus, Chiarelli established that it was not in itself fundamentally unjust for Parliament to devise criteria to govern the entry and residency of non-citizens in Canada, as contemplated by the Charter’s stipulation of differing mobility rights for citizens and non- citizens,56 and to provide for their enforcement. Non-citizens’ lack of an unqualified right to enter or remain in Canada supplied the context which informed the scope of the principles of fundamental justice. However, the Court did not tie this “fundamental principle of immigration law” to the scope of liberty or security of the person. It very deliberately made no decision on the engagement of s. 7. (b) Conflicting decisions at the Federal Court of Appeal Around the time that Chiarelli was before the Supreme Court, the Federal Court of Appeal issued conflicting decisions on whether proceedings involving the potential removal of non-citizens engaged their right to liberty. In Grewal v Canada (Minister 57 of Employment and Immigration), a permanent resident being deported to India for 2017 CanLIIDocs 175 criminal activity unsuccessfully appealed the deportation to the Immigration Appeal Board. His application to the Minister for humanitarian and compassionate relief was also dismissed. Finally, an immigration adjudicator refused to re-open the immigration inquiry at which he had been ordered deported so that he might register a refugee claim. Grewal argued that, in his circumstances, s. 7 required the re- opening. Following its decision in Chiarelli, the Federal Court of Appeal accepted that s. 7 applied: It has already been determined that the deportation of refugees infringes their right to security of the person. (Singh…). This, of course, does not mean that people cannot be deported for good reason, that is, as long as there is no violation of the principles of fundamental justice … Hence, it is permissible to deport a permanent resident for the commission of a serious offence without violating the Charter, as long as fundamental justice has been accorded to that person before doing so. … The legislation and the earlier jurisprudence of this court must yield to the dictates of section 7.58 In Hoang v Canada (Minister of Employment and Immigration),59 a permanent resident of Vietnamese origin previously recognized as a Convention 54 Ibid at 739. Significantly, a ministerial humanitarian and compassionate review would still have been available to Chiarelli under s. 114(2) of the Immigration Act, RSC 1985, c I-2, s 114(2). 55 Ibid at 746. 56 Ibid at 733–4. 57 Grewal v Canada (Minister of Employment and Immigration), [1992] 1 FCR 581, 85 DLR (4th) 166 (CA) [Grewal]. 58 Ibid at 587–8. 2017] REVISITING THE APPLICATION OF SECTION 7 323 refugee unsuccessfully appealed a removal order issued against him as a result of convictions for serious criminal offenses. Before the Federal Court of Appeal, he argued that in light of his possible deportation to Vietnam, the procedures mandated by the Immigration Act violated ss. 7 and 12 of the Charter. Justice MacGuigan quoted at length from the Federal Court of Appeal’s conclusion in Chiarelli that a deportation order made against a permanent resident as a result of a conviction was not contrary to s. 7 because: There is no injustice in requiring the deportation of a person who has lost the right to remain in the country; there is no injustice, either, in prescribing that a foreigner who has been admitted here as a permanent resident will lose the right to remain in the country if he is found guilty of an offence which, in itself, Parliament considers to be serious.60 In this passage from Chiarelli, the Federal Court of Appeal rejected Chiarelli’s s. 7 claim because he had not shown that the removal of persons convicted of a serious offense “raised any injustice” or, in other words, violated fundamental justice. It did not find that s. 7 was not engaged. Nevertheless, without adverting to the Federal Court of Appeal’s unanimous view in Chiarelli that deportation “necessarily implies 2017 CanLIIDocs 175 an interference with the liberty of the person,” Justice MacGuigan erroneously concluded that “… on the authority of Hurd and Chiarelli, deportation for serious offenses affect neither s. 7 nor s. 12 rights, since it is not to be conceptualized as either a deprivation of liberty or a punishment.”61 The Federal Court of Appeal compounded this error in Canepa v Canada (Minister of Employment and Immigration),62 once again dismissing the argument that the removal of a permanent resident who had established a substantial connection with Canada engaged s. 7. Acknowledging that the Supreme Court had, in Chiarelli, “left open the question whether deportation for serious offences can be conceptualized as a deprivation of liberty under s. 7,” Justice MacGuigan reasoned that the Court of Appeal had answered that question in the negative in Hoang and was “bound by its previous decisions,”63 a conclusion he reiterated in Barrera v Canada (Minister of Employment and Immigration),64 which involved a proceeding 59 Hoang v Canada (Minister of Employment and Immigration) (1990), 13 Imm LR (2d) 35, 42 ACWS (3d) 1140 [Hoang]. 60 Chiarelli FCA, supra note 49 at 310 [emphasis added]. 61 Hoang, supra note 59 at 41. In Hurd v Canada (Minister of Employment and Immigration), [1989] 2 FCR 594, 12 ACWS (3d) 328 (CA), the Federal Court of Appeal determined that deportation proceedings were not proceedings that could lead to truly penal consequences and to which s. 11(h) of the Charter could apply. 62 Canepa v Canada (Minister of Employment and Immigration), [1992] 3 FCR 270, 93 DLR (4th) 589 leave to appeal dismissed, [1993] 1 SCR v, [1992] SCCA No. 410 [Canepa]. 63 Ibid at 277. 64 Barrera v Canada (Minister of Employment and Immigration), [1993] 2 FCR 3, 99 DLR (4th) 264. 324 UNBLJ RD UN-B [VOL/TOME 68 to deport a Convention refugee as a result of serious criminal convictions. To the extent they rely exclusively on the Federal Court of Appeal’s decision in Chiarelli for the proposition that deportation does not engage non-citizens’ liberty interest, Hoang, Canepa and Barrera were based on a misreading of that decision and were wrongly decided. The question of the application of s. 7 to proceedings under the Immigration Act was once again considered by the Federal Court of Appeal in Nguyen v Canada (Minister of Employment and Immigration).65 Nguyen, a landed immigrant convicted of serious criminal offenses, challenged the constitutionality of two decisions under the Immigration Act: first, that he was a person convicted of a serious criminal offence and thus subject to deportation; and second, that he was not eligible to have his refugee claim referred to the Refugee Division of the Immigration and Refugee Board for determination (the Minister having issued a certificate stating that he constituted a danger to the public in Canada). Relying on the Supreme Court’s judgment in Chiarelli, Justice Marceau held that the requirement of no serious criminal convictions was neither illegitimate nor arbitrary and that the procedure to determine whether a non-citizen had breached this requirement did not violate 2017 CanLIIDocs 175 fundamental justice and thus complied with s. 7. With regard to whether s. 7 was engaged by removal, he concluded that “forcibly deporting an individual against his will has the necessary effect of interfering with his liberty, in any meaning that the word can bear, in the same manner as extradition was found to interfere in Kindler, supra.”66 Justice Marceau held that the decision finding Nguyen ineligible to have his refugee claim determined by the Convention Refugee Determination Division did not, in itself, engage s. 7 since “contrary to the first decision which entailed forced deportation and therefore deprivation of liberty, a declaration of ineligibility does not imply or lead, in itself, to any positive act which may affect life, liberty or security of the person.”67 However, this did not end the matter. Justice Marceau proceeded to examine the constitutionality of the two provisions in the context of the entire scheme: The Supreme Court [in Chiarelli], following in that respect the approach of this Court, examined the constitutional challenge as being aimed at the scheme viewed as a whole. The removal of the special right to appeal was perceived as the removal of a means to oppose the deportation order and, as a result, might engage section 7 of the Charter. Similarly in our case, while a determination of ineligibility under subparagraph 46.01(1)(e)(ii) of the Act is only indirectly linked to the deportation order, nevertheless it has the effect of taking away the only possible barrier to the issuance of an unconditional deportation order, and as such participates in the deprivation of liberty and, possibly, the security of the individual which results from 65 Nguyen v Canada (Minister of Employment and Immigration), [1993] 1 FCR 696, 100 DLR (4th) 151 (CA) [Nguyen]. 66 Ibid at para 7, footnote 5. 67 Ibid at 704. 2017] REVISITING THE APPLICATION OF SECTION 7 325 deportation. More generally, the deprivation of liberty involved in any forced deportation is given a new dimension by the fact that the individual to be deported claims to be a refugee. It is appropriate, therefore, to assume that section 7 of the Charter is brought into play with respect to the scheme as a whole, that is to say with respect not only to the issuance of the deportation order, but also to the ineligibility decision based on the public danger certificate. The question becomes whether the issuance of the public danger certificate, the central feature of the scheme as a whole, could be said to have violated a principle of fundamental justice.68 The underlined passages in Justice Marceau’s judgment support the proposition that immigration proceedings linked to the deportation of non-citizens engage their s. 7 liberty interest so long as they make deportation more likely.69 This approach to causation, sensitive to the specific statutory context, is consistent with the standard recently set by the Supreme Court in Bedford. As will be discussed further, it is, for reasons unexplained, no longer followed in the deportation context by the Federal Court of Appeal nor, arguably, by the Supreme Court itself.70 In Williams v Canada (Minister of Citizenship and Immigration),71 a 2017 CanLIIDocs 175 permanent resident of Jamaican origin convicted of serious criminal offenses was ordered deported. The Minister of Citizenship and Immigration issued an opinion that Williams constituted a danger to the Canadian public, stripping him of his right to appeal to the Immigration Appeal Board. The Federal Court of Appeal considered whether this engaged Williams’s liberty or security of the person. Justice Strayer acknowledged that the “jurisprudence of this Court on this subject has not been entirely consistent”72 and contrasted the line of decisions finding s. 7 engagement (Chiarelli and Nguyen) with that finding no engagement (Hoang, Canepa and Barrera). He determined that liberty was not engaged: … I have difficulty understanding how the refusal of a discretionary exemption from a lawful deportation order, as applied to a non-refugee who has no legal right to be in the country, must be seen as involving a deprivation of liberty. Unless “liberty” is taken to include the freedom to be anywhere one wishes, regardless of the law, how can it be “deprived” by the lawful execution of a removal order? … 68 Ibid at para 10 [underlining added]. 69 Nguyen was followed by the Federal Court Trial Division in Kaberuka v Canada (Minister of Employment and Immigration), [1995] 3 FCR 252 at 262, 32 Imm LR (2d) 38 (TD) [Kaberuka]. 70 I discuss this point further in section 3, infra. 71 Williams v Canada (Minister of Citizenship and Immigration), [1997] 2 FCR 646, 147 DLR (4th) 93 (CA) [Williams]. 72 Ibid at para 23. 326 UNBLJ RD UN-B [VOL/TOME 68 On the basis of the jurisprudence to date, then, I am unable to conclude that “liberty” includes the right of personal choice for permanent residents to stay in this country where, as the Supreme Court said in Chiarelli: [t]hey have all deliberately violated an essential condition under which they were permitted to remain in Canada.73 The Federal Court of Appeal’s reliance on the Supreme Court’s decision in Chiarelli to conclude that deportation did not engage non-citizens’ liberty – a question expressly not considered by the Supreme Court – indicates that it once more74 confused this question with the question answered by Chiarelli: whether a (presumed) deprivation of non-citizens’ liberty accorded with substantive principles of fundamental justice in circumstances where they had violated an essential condition under which they could remain in Canada.75 Justice Strayer also dismissed the respondent’s argument that, consistent with a broadening understanding of the liberty interest, deportation engaged s. 7 because it interfered with non-citizens’ personal autonomy over important decisions intimately affecting their private lives.76 While he correctly held that this broader 2017 CanLIIDocs 175 view of liberty had not yet been accepted by a majority of the Supreme Court, this would happen only four years later, in Blencoe. In Romans v Canada (Minister of Citizenship and Immigration), a case decided shortly after Blencoe, the Federal Court Trial Division held that deportation engaged a deportee’s liberty interest in its broad sense: The consequence of the issuance of the deportation order against an individual is profound. The deportation order prohibits Mr. Romans from making the fundamental personal choice to remain in Canada where he receives the love and support of his family, financial support, and the support of his social worker and the health-care system. I am satisfied that in the circumstances before me the issuance of a deportation order… engages section 7 of the Charter.77 What can be gleaned from a decade of Federal Court decisions on whether the deportation of non-citizens engages their liberty interest under s. 7? Decisions of 73 Ibid at paras 24 and 26. 74 As it had in Hoang, Canepa and Barrera, supra notes 59, 60 and 62. 75 The Court of Appeal did not engage with the reasoning in Chiarelli and Nguyen that deportation of non- citizens necessarily interfered with their liberty, adopted by the Federal Court, Trial Division subsequent to Williams in Al Yamani v Canada (Minister of Citizenship and Immigration), [2000] 3 FCR 433 at paras 59 and 61, 5 Imm LR (3d) 235 (TD). 76 The respondent relied on the judgment of LaForest, L’Heureux-Dubé, Gonthier and McLachlin JJ in B(R), supra note 18. 77 Romans v Canada (Minister of Citizenship and Immigration), 2001 FCT 466 at para 22, 14 Imm LR (3d) 215 [Romans]. Romans, who had come to Canada at the age of two, was later diagnosed with chronic paranoid schizophrenia. Canada sought to remove him on grounds of serious criminality. Relying on Chiarelli, the Court found no breach of the principles of fundamental justice, a conclusion upheld by the Federal Court of Appeal which accepted, without deciding, that s. 7 was engaged: Romans v Canada (Minister of Citizenship and Immigration), 2001 FCA 272 at para 1, 17 Imm LR (3d) 34. 2017] REVISITING THE APPLICATION OF SECTION 7 327 the Federal Court of Appeal answering this question in the negative were based on a misreading of the Chiarelli decision, which did not decide the question of engagement. Not one of them provides a principled or compelling answer to the observations of Justice Pratte in Chiarelli or of Justice Marceau in Nguyen that forcibly deporting an individual against his will necessarily interferes with his liberty. Medovarksi offered the Supreme Court an opportunity to address in a principled manner the question of s. 7 engagement in the deportation context and perhaps, as Justice Dawson had done in Romans, apply to it the expanded conception of liberty it had recently adopted in Blencoe. (c) The Supreme Court speaks: Medovarski and Charkaoui The question of whether liberty and security of the person were engaged by proceedings leading to non-citizens’ removal from Canada was squarely before the Supreme Court in Medovarksi.78 Medovarksi and Esteban were permanent residents who had been ordered deported for serious criminality. They had appealed their removal to the Immigration Appeal Division and their removal orders were 2017 CanLIIDocs 175 automatically stayed under provisions of the Immigration Act. When the IRPA was enacted, their appeals were discontinued under transitional provisions. Medovarksi argued that on a proper interpretation of these provisions, her right of appeal should have been preserved. In the alternative, she claimed that its discontinuance infringed s. 7: She claims that deportation removes her liberty to make fundamental decisions that affect her personal life, including her choice to remain with her partner. Medovarski argues her security of the person is infringed by the state-imposed psychological stress of being deported. Medovarski further alleges that the process by which her appeal was extinguished was unfair, contrary to the principles of fundamental justice.79 The Supreme Court of Canada declared that: The most fundamental principle of immigration law is that non-citizens do not have an unqualified right to enter or remain in Canada: Chiarelli … at p. 733. Thus the deportation of a non-citizen in itself cannot implicate the liberty and security interests protected by s. 7 of the Canadian Charter of Rights and Freedoms.80 78 Charkaoui, supra note 5. 79 Medovarksi, supra note 4 at para 45. The Federal Court of Appeal did not decide whether s. 7 was engaged by Medovarksi’s removal from Canada, finding that, based on Chiarelli, the principles of fundamental justice were not offended by the discontinuance of her appeal: Medovarksi v Canada (Minister of Citizenship and Immigration), 2004 FCA 85 at paras 58–62, [2004] 4 FCR 48. 80 Medovarksi, supra note 4 at para 46. 328 UNBLJ RD UN-B [VOL/TOME 68 It also held that even if liberty and security of the person were engaged, Medovarksi had not established that any unfairness wrought by the transition to IRPA breached the principles of fundamental justice.81 According to Chiarelli, fundamental justice did not mandate an appeal on humanitarian and compassionate grounds, which could, in any event, be considered by the Minister if Medovarksi applied to remain in Canada under s. 25(1) of IRPA.82 In Charkaoui,83 the Supreme Court significantly qualified its holding in Medovarksi on the engagement of s. 7. Adil Charkaoui, a permanent resident, and Hassan Almrei and Mohammed Harkat, both foreign nationals recognized as Convention refugees, were named in certificates of inadmissibility (“security certificates”) issued by the Minister of Public Safety and Emergency Preparedness and the Minister of Citizenship and Immigration under the IRPA. Following the issuance of the certificates, which deemed them to be threats to Canada’s national security, all three individuals were detained pending the completion of a multistage process for their removal.84 First, a Federal Court judge determined whether the certificate was reasonable in proceedings conducted, at the Ministers’ request, in 85 camera and ex parte. A certificate determined to be reasonable became a removal 2017 CanLIIDocs 175 order. Second, the named person could apply to the Minister of Citizenship and Immigration for a pre-removal risk assessment (PRRA), which would consider whether removal would subject him to a danger of torture or to a risk to his life or of cruel and unusual treatment or punishment, and whether his claim for protection should be refused because of the nature and severity of acts he had committed or because of the danger he constituted to the security of Canada.86 A successful PRRA application would result in a stay of the removal order. Charkaoui, Almrei and Harkat challenged the constitutionality of the procedure for determining the reasonableness of the certificate, claiming, inter alia, that it infringed their rights to life, liberty and security of the person under s. 7 of the Charter. The Court observed that the claimants were required to prove two matters: [F]irst, that there has been or could be a deprivation of the right to life, liberty and security of the person, and second, that the deprivation was not or would not be in accordance with the principles of fundamental justice.87 81 Ibid at para 47. 82 A humanitarian and compassionate application to the Minister had also been available to Chiarelli under the Immigration Act, RSC 1985, c I-2, s 114(2). 83 Charkoui, supra note 5. 84 The IRPA provided that upon issuance of a certificate, permanent residents may be held in detention but that foreign nationals must be detained: Charkaoui, supra note 5 at para 6. 85 The Ministers and the designated judge could rely on undisclosed material that neither the person named in the certificate nor their counsel could see. The judge disclosed to the named person a summary of the case against him but could not disclose information that might compromise national security. 86 IRPA, supra note 25 at ss 113(d) and 97. 87 Charkaoui, supra note 5 at para 12 [underlining added]. 2017] REVISITING THE APPLICATION OF SECTION 7 329 The Court’s use of “could be” or “would be” signals that a successful s. 7 claim does not require that the claimant actually be detained or subjected to treatment causing psychological or physical suffering; the risk of such treatment is enough. Indeed, the Court determined that the security certificate provisions “clearly deprive detainees such as the appellants of their liberty,” noting that “the person named in a security certificate can face detention pending the outcome of the proceeding,”88 and that detention was automatic for foreign nationals. In other words, while the deprivation of liberty in Charkaoui was clear because the named persons were actually detained, the possibility of detention also engaged the liberty interest. Rejecting the Attorney General’s claim that Medovarksi excluded the application of s. 7 to removal proceedings, the Court stated that “[w]hile the deportation of a non-citizen in the immigration context may not in itself engage s. 7 of the Charter, some features associated with deportation, such as detention in the course of the certificate process or the prospect of deportation to torture, may do so.”89 Professor Hamish Stewart has observed that the Supreme Court of Canada’s decisions in Charkaoui and Medovarksi are in tension on the question of whether s. 7 is engaged in proceedings leading to removal from Canada: 2017 CanLIIDocs 175 Criminal proceedings, and most other penal proceedings as well, have to comply with section 7 from the outset because of the potential for imprisonment that they create … Because of the holding in Medovarksi, this logic apparently does not apply to deportation proceedings; thus, in Poshteh v Canada (Minister of Citizenship and Immigration), the Federal Court of Appeal held that the initial steps in proceedings that may lead to deportation, such as a finding that a person is inadmissible to Canada, do not engage section 7 because those initial steps do not necessarily mean that individual will ever be detained. But because of the holding in Charkaoui, there must be some point in the proceedings where the likelihood of detention incidental to deportation is sufficiently high that the liberty interest is engaged and section 7 applies.90 Another possibility, of course, is that the tension between Medovarksi and Charkaoui cannot be resolved and that Medovarksi should be reconsidered, a question I examine in the next section. 88 Ibid at para 13 [underlining added]. 89 Ibid at para 17. 90 Hamish Stewart, Fundamental Justice: Section 7 of the Canadian Charter of Rights and Freedoms (Toronto: Irwin Law, 2007) at 81. 330 UNBLJ RD UN-B [VOL/TOME 68 (d) Liberty and immigration and refugee proceedings: an appraisal of the case law Medovarksi squarely raised a question that had long been contested in the Federal Court of Appeal: whether deportation in itself can implicate s. 7 interests. The Supreme Court’s decision that it could not, based entirely on its assertion of non- citizens’ qualified right to enter or remain in Canada – a principle mobilized in Chiarelli to narrow the content of fundamental justice in the immigration context – was perplexing. Professors Donald Galloway and Jamie Liew saw in this passage from Medovarksi “a remarkable extrapolation… based on the failure to distinguish between, on the one hand, not interfering with a right to liberty and security, and, on the other hand, interfering with the right but doing so in a manner that accords with the principles of fundamental justice,” a distinction expressly drawn by the Court in Chiarelli but “glossed over” in Medovarski.91 Rather than confusing the question of s. 7 engagement and that of compliance with fundamental justice, the Supreme Court may, in referring to the qualified rights of non-citizens, have intended to imply that because s. 6 of the 2017 CanLIIDocs 175 Charter confers exclusively on citizens the constitutional right to enter, remain in and leave Canada, the deportation of a non-citizen would not violate s. 6. However, the fact that the deportation of non-citizens does not violate their mobility rights does not mean that it cannot engage other Charter rights.92 The Supreme Court correctly rejected a similar claim in the extradition context, holding that the fact that the breach of extraditees’ s. 6 rights was generally justifiable did not insulate the extradition process from scrutiny for violation of other Charter rights, including s. 7.93 It could not seriously be argued, for example, that a law aimed at prioritizing for 91 Donald Galloway & Jamie Chai Yun Liew, Immigration Law, 2d ed (Toronto: Irwin Law, 2015) at 656. 92 Ibid at 80. 93 See R v Schmidt, [1987] 1 SCR 500, 39 DLR (4th) 18. Schmidt was facing extradition to the United States to face a state charge of child stealing after having been acquitted of a federal charge of kidnapping. She argued that her extradition would violate her right, under s. 11(h) of the Charter, not to be tried again for an offence of which she had been finally acquitted. Noting that the Ontario Court of Appeal, in Federal Republic of Germany v Rauca (1983), 38 OR (2d) 225, 145 DLR (3d) 638 (CA) had determined that extradition was a reasonable infringement on the right of Canadian citizens, under s. 6 of the Charter, to remain in Canada, the Ontario High Court of Justice had decided that any argument that specific aspects of extradition were contrary to other Charter rights, including those guaranteed by ss. 11(h) and 7, was ruled out: R v Schmidt (1983), 41 OR (2d) 399 at 407, 147 DLR (2d) 616. Justice La Forest, for a majority of the Supreme Court, disagreed with this conclusion: … I am far from thinking that the Charter has no application to extradition. The surrender of a person to a foreign country may obviously affect a number of Charter rights. In Rauca, supra, for example, the Ontario Court of Appeal recognized that extradition intruded on a citizen’s right under s. 6 to remain in Canada, although it also found that the beneficial aspects of the procedure in preventing malefactors from evading justice, a procedure widely adopted all over the world, were sufficient to sustain it as a reasonable limit under s. 1 of the Charter. Section 6 was not raised in this case, though Schmidt is a Canadian citizen, no doubt because her counsel believed, as I do, that it was properly disposed of in the Rauca case. However, it does not follow from the fact that the procedure is generally justifiable that the manner in which the procedures are conducted in Canada and the conditions under which a fugitive is surrendered can never invite Charter scrutiny. The pre-eminence of the Constitution must be recognized; the treaty, the 2017] REVISITING THE APPLICATION OF SECTION 7 331 removal non-citizens of African origin would not engage and violate their equality rights under s. 15.94 Why, then, should non-citizens be precluded from asserting that their forced removal from Canada subjects them to possible detention or would deprive them of the opportunity to parent and care for their children, a fundamental choice recognized by several Supreme Court judges as an aspect of liberty under s. 7, or interfere with another similarly fundamental choice? Lacking in Medovarski, as Professor Stewart notes, is “a more careful analysis of the nature and effect of deportation on a person present in Canada.”95 Equally absent is any analysis of whether non-citizens’ physical liberty is engaged by the possibility of detention incidental to removal, an argument considered by the Federal Court of Appeal in its earlier decisions. As Professor Stewart points out, deportation engages the liberty interest “because a deportation order includes the possibility of detaining the deportee in order to carry it out, just as penal proceedings engage the liberty interest because a finding of guilt includes possibility of imprisonment as punishment.”96 While not giving non-citizens “an unqualified right to enter or remain in Canada,” this solution would “require the legal rules governing deportation from Canada to comply with the principles of fundamental justice,” an appropriate requirement “given the importance to a permanent resident or Convention refugee of remaining in 2017 CanLIIDocs 175 Canada.”97 Indeed, under the IRPA, an officer may, without warrant, arrest and detain a foreign national, other than a protected person, who the officer has reasonable grounds to believe is inadmissible and is a danger to the public or is unlikely to appear for removal from Canada or other proceeding that could lead to the making of a removal order by the Minister under ss. 44(2).98 The IRPA establishes a regime that places non-citizens under the administrative control of the state99 in large measure through the threat of their forced removal from Canada. As part of the process set up to achieve this end, the enforcement provisions of IRPA establish a statutory extradition hearing in this country and the exercise of the executive discretion to surrender a fugitive must all conform to the requirements of the Charter, including the principles of fundamental justice. [Underlining added]. See also F Pearl Eliadis, “The Swing from Singh: The Narrowing Application of the Charter in Immigration Law” (1995) 26 Imm LR (2d) 130 at 142. 94 See, for example, YZ v Canada (Minister of Citizenship and Immigration), 2015 FC 892, 387 DLR (4th) 676, where the Federal Court determined that a provision of IRPA that denied refugee claimants from designated countries of origin access to an appeal before the Refugee Appeal Division violated s. 15(1) of the Charter. 95 Stewart, supra note 90 at 80. 96 Ibid. 97 Ibid. 98 IRPA, supra note 25, s 55(2). Permanent residents or foreign nationals may also be arrested and detained pursuant to a warrant on the same grounds: s 55(1). 99 Canadian Doctors for Refugee Care v Canada (AG), 2014 FC 651, 28 Imm LR (4th) 1 [Canadian Doctors for Refugee Care]. 332 UNBLJ RD UN-B [VOL/TOME 68 compulsion on non-citizens to appear at a specific time and place subject to legal consequences, including arrest and detention, with or without a warrant depending on the circumstances. Just as this kind of statutory compulsion triggered the liberty interest in the Beare and Thomson judgments,100 the potential of detention incidental to removal and to proceedings that could lead to removal should suffice to engage non-citizens’ right to liberty under s. 7, an outcome hinted at, as noted above, in the Supreme Court’s decision in Charkaoui. Contrasting the scope of application of s. 7 in immigration and refugee protection proceedings to the extradition and penal contexts is instructive. In Canada, extradition proceedings begin when a foreign state requests that Canada surrender a person to be prosecuted or to serve a sentence for extraditable conduct.101 The proceedings that follow involve several steps. First, the Minister of Justice issues an “authority to proceed” authorizing the Attorney General to seek a court order for the committal of the extraditee if satisfied that the conditions for extradition are met in respect of one or more offenses mentioned in the request.102 The Attorney General may then apply for the issuance of a summons to the extraditee or a warrant for that 103 person’s arrest. At the judicial phase of the extradition process, an extradition 2017 CanLIIDocs 175 hearing before a superior court judge, the extradition judge must decide whether there is “evidence… of conduct that, had it occurred in Canada, would justify committal for trial in Canada on the offense set out in the authority to proceed” and, if so, order the committal of the person into custody to await surrender.104 In the ministerial phase following committal, the Minister of Justice must decide whether to surrender the person to the requesting state105 and make surrender conditional on assurances from the requesting state.106 The Supreme Court has stated that “section 7 permeates the entire extradition process” and is engaged at both the stages of committal and surrender.107 At the committal stage, s. 7 requires the extradition judge to ensure “that the committal order, if it is to issue, is the product of a fair judicial process.”108 The liberty interest is engaged because “the person sought may be detained while an extradition request is dealt with, and will certainly be detained if that request is 100 Beare, supra note 28 and Thomson, supra note 29. 101 Extradition Act, SC 1999, c 18, ss 2 and 3. Requests are usually made pursuant to an extradition treaty. 102 Ibid at s 15. 103 Ibid at s 16. 104 Ibid at s 29(1). 105 The reasons for which the Minister may refuse to surrender the extraditee are listed in ss. 44 to 47 of the Extradition Act, ibid. 106 Ibid at s 58(f). 107 United States of America v Cobb, 2001 SCC 19 at para 34, [2001] 1 SCR 587 [emphasis added]. 108 Ibid. 2017] REVISITING THE APPLICATION OF SECTION 7 333 granted.”109 Indeed, even the very first step of the process, the issuance by the Minister of an authority to proceed, is subject to s. 7. Issuance of an authority to proceed in circumstances disclosing bad faith or improper motives or where the authority to proceed provides the person sought with inadequate notice of the case he or she faces will violate the principles of fundamental justice, breach s. 7 and justify the quashing of the authority to proceed under s. 24(1) of the Charter.110 Similarly, in penal proceedings, it is the possibility of detention and imprisonment as an outcome of the proceedings which justifies the application of s. 7 from their outset. As Professor Stewart notes, “it would be odd if the principles of fundamental justice came into play only at the point where the accused had been convicted and the judge had decided to imprison him, or if the content of the applicable principles was different depending on whether the Crown announced its intention to seek a term of imprisonment before the trial began.”111 The Supreme Court’s approach ensures that “the principles of fundamental justice will always apply in penal proceedings, whether or not imprisonment, another form of detention, or probation will ultimately be imposed.”112 2017 CanLIIDocs 175 That the risk of immigration detention may be lower than the risk of imprisonment in penal proceedings should not defeat the claim that the liberty interest is engaged by proceedings under the IRPA. A majority of the Supreme Court found that the availability of imprisonment for the offence of simple possession of marijuana was sufficient to trigger s. 7 scrutiny113 despite the fact that imprisonment was only imposed by the Courts in exceptional circumstances.114 The exceptional nature of imprisonment and the relatively short sentences associated with conviction spoke not to the engagement of s. 7, which flowed from the availability of imprisonment, but to whether this availability breached any principles of fundamental justice – in particular, the principle against gross disproportionality.115 An approach to the engagement of the liberty interest in immigration and refugee proceedings which, consistent with that adopted by the Court in the context of extradition and penal proceedings, recognizes that liberty is engaged from the outset of the proceedings given the possibility of detention incidental to deportation 109 Stewart, supra note 90 at 71 [emphasis added]. See also United States of America v Ferras; United States of America v Latty, 2006 SCC 33 at para 49, [2006] 2 SCR 77. 110 Froom v Canada (Minister of Justice), 2005 FCA 352 at paras 18–19, [2005] 2 FCR 19; United States of America v Saas, (2004) 237 DLR (4th) 623, 61 WCB (2d) 325 (ONCA). 111 Stewart, supra note 90 at 69. 112 Ibid. 113 R v Malmo-Levine; R v Caine, 2003 SCC 74 at paras 84, 89, [2003] 3 SCR 571. 114 Ibid at para 154. 115 Ibid at paras 158–161. 334 UNBLJ RD UN-B [VOL/TOME 68 seems particularly apt considering the growing convergence and overlap between criminal law and immigration law known as “crimmigration.”116 A hallmark of this convergence has been that “immigration enforcement measures – particularly detention and deportation – are used much more commonly in response to suspected criminal activity than ever before.”117 Writing on the longstanding characterization of deportation by American courts as a civil rather than criminal or penal proceeding and thus subject to a dramatically lower level of constitutional scrutiny, Kanstroom has argued that the “increasing real world convergence” between the United States’ criminal justice and deportation systems “compels a rethinking of the foundational principles underlying the constitutional status of deportation.”118 Since the deportation of long-term permanent residents for post-entry criminal conduct serves an incapacitating function to the deported, a deterrent function to others and could be understood as a form of retribution – justifications accepted as part of criminal law – one might assume, Kanstroom observes, that persons subject to these types of proceedings “would at least have the most basic constitutional rights accorded to criminal defendants,” an assumption supported by the fact that deportation proceedings are “initiated by a government agency, are directly based on criminal conduct, involve incarceration and forced movement of persons, and may result in 2017 CanLIIDocs 175 lifetime banishment.”119 Legislative developments in Canada too have seen a marked erosion of the statutory protections afforded to permanent residents against deportation on grounds of serious criminality.120 These legislative efforts culminated in the enactment of the Faster Removal of Foreign Criminals Act,121 which subjected permanent residents sentenced in Canada to more than six months imprisonment (including conditional sentence orders) to automatic removal with no IAD review of the circumstances of their case and removed the ability of the Minister to consider humanitarian and compassionate factors against removal for permanent residents inadmissible on grounds of organized criminality.122 Parliament has thus made deportation the automatic consequence of receiving a sentence over six months on conviction of one 116 Sharryn Aiken, David Lyon & Malcolm Thorburn, “Criminalization, Surveillance and ‘Security Threats’: A Multidisciplinary Dialogue” (2014) 40:1 Queen’s LJ i–xi; Juliet Stumpf, “The Crimmigration Crisis: Immigrants, Crime, and Sovereign Power” (2006) 56 Am U L Rev 367. See also Katja Franko Aas & Mary Bosworth, “Preface” in Katja Franko Aas & Mary Bosworth, eds, The Borders of Punishment: Migration, Citizenship and Social Exclusion (Oxford: Oxford University Press, 2013) at vii. 117 Aiken, Lyon & Thorburn, supra note 116 at ii. 118 Daniel Kanstroom, “Deportation, Social Control, and Punishment: Some Thoughts About Why Hard Laws Make Bad Cases” (2000) 113 Harv L Rev 1889 at 1892. 119 Ibid at 1894. 120 Canadian Bar Association, Citizenship and Immigration Law Section, Submission on Bill C-31 – Immigration and Refugee Protection Act (May 2002), online: CBA 121 An Act to amend the Immigration and Refugee Protection Act, SC 2013, c 16. 122 Canadian Bar Association, National Immigration Law Section, Bill C-43, Faster Removal of Foreign Criminals Act (November 2012), online: CBA 2017] REVISITING THE APPLICATION OF SECTION 7 335 of a broad range of criminal offences. Catherine Dauvergne observes that the imposition of eligibility provisions based on criminality as a precondition of access to domestic asylum processes is an example of the “criminalization of asylum seeking.”123 In light of Kanstroom’s criticisms of the American constitutional jurisprudence on deportation, the emergence of crimmigration in Canada highlights the weaknesses of an approach to the engagement of constitutional protections that would hinge on whether deportation should be labelled as a “penal,” “criminal,” “civil” or “immigration” proceeding. The Supreme Court appeared to have recognized this in Charkaoui when it dismissed the claim that Medovarksi stood for the proposition that deportation proceedings were immune from s. 7 scrutiny: In determining whether s. 7 applies, we must look at the interests at stake rather than the legal label attached to the impugned legislation. As Professor Hamish Stewart writes: Many of the principles of fundamental justice were developed in criminal cases, but their application is not restricted to criminal cases: they apply whenever one of the three protected interests is engaged. Put another way, the principles of fundamental justice apply in 2017 CanLIIDocs 175 criminal proceedings, not because they are criminal proceedings, but because the liberty interest is always engaged in criminal proceedings. [Emphasis in original.]124 In addition to engaging non-citizens’ liberty interest, narrowly defined as including freedom from the threat of detention incidental to deportation, deportation decisions also arguably engage liberty broadly defined in Blencoe as protecting important and fundamental life choices. The Supreme Court has on several occasions come close to recognizing that “the right to nurture a child, to care for its development and to make decisions for it in fundamental matters… are part of the liberty interest of a parent.”125 Justice Wilson, who first accepted this proposition, described the parental liberty interest as an aspect of the right to respect for an individual’s private and family life protected at international law: [The appellant] has the right, I believe, to raise his children in accordance with his conscientious beliefs. The relations of affection between an individual and his family and his assumption of duties and responsibilities towards them are central to the individual’s sense of self and of his place 123 Catherine Dauvergne, “The Troublesome Intersections of Refugee Law and Criminal Law” in Katja Franko Aas & Mary Bosworth, eds, The Borders of Punishment: Migration, Citizenship and Social Exclusion (Oxford: Oxford University Press, 2013) 76 at 76. 124 Charkaoui, supra note 5 at para 18, citing Hamish Stewart, “Is Indefinite Detention of Terrorist Suspects Really Constitutional?” (2005) 54 UNBLJ 235 at 242 125 B(R), supra note 18 at para 83 per La Forest, Gonthier and McLachlin JJ; G(J), supra note 14 per L’Heureux-Dubé, Gonthier and McLachlin JJ; and Chamberlain v Surrey School District No 36, 2002 SCC 86 at para 87, [2002] 4 SCR 710 per Gonthier and Bastarache JJ. 336 UNBLJ RD UN-B [VOL/TOME 68 in the world. The right to educate his children is one facet of this larger concept. This has been widely recognized. Article 8(1) of the European Convention for the Protection of Human Rights and Fundamental Freedoms… states in part “Everyone has the right to respect for his private and family life...”126 Under this approach, liberty is engaged where deportation would interfere with a non-citizen’s ability to nurture and care for his or her children. Section 7 would be breached in such circumstances if deportation violated fundamental justice by causing a deprivation of liberty grossly disproportionate to the state’s objective in pursuing removal.127 The framework set down by the Supreme Court in Blencoe to determine whether the liberty interest is engaged requires an analysis of whether “in the circumstances of this case,” the state has prevented the rights claimant from making any fundamental personal choices – basic choices going to the core of what it means to enjoy individual dignity and independence. No such analysis appears in the Court’s decision in Medovarksi.128 Instead, the Court invoked non-citizens’ lack of an “unqualified right to enter or remain in Canada” to defeat the s. 7 claim. But 2017 CanLIIDocs 175 under the Blencoe framework, this argument would only suffice in circumstances where a non-citizen’s liberty claim could be reduced to the bare assertion of a mobility right – the right to enter Canada freely and remain there as if the international border did not exist – as the expression of the core of his or her individual dignity and independence. It is not an answer to non-citizens’ assertion of fundamental interests that go beyond mobility. 126 R v Jones, [1986] 2 SCR 284 at 319, 47 Alta LR (2d) 97. 127 See Bedford, supra note 34 at para 120: a law or state action violates fundamental justice where its “effects on life, liberty or security of the person are so grossly disproportionate to its purposes that they cannot rationally be supported.” The requirement that deportation be proportionate to a legitimate state objective is an integral part of the jurisprudence under art 8 of the European Convention for the Protection of Human Rights and Fundamental Freedoms, 213 UNTS 222 (1950) which guarantees the right to respect of one’s private and family life. While recognizing the power of European states to control the entry of aliens into their territories and their residence there, the European Court of Human Rights has held that in some circumstances, the expulsion of an alien will violate art. 8: Üner v The Netherlands (18 Oct. 2006), no. 46410/99 (European Court of Human Rights) at para 57, online (http://hudoc.echr.coe.int/eng?=001-77542). A deportation decision that interferes with family life will be found to violate art. 8 if it is not “in accordance with the law and necessary in a democratic society, that is to say, justified by a pressing social need and, in particular, proportionate to the legitimate aim pursued”: ibid, at para 54. The United Nations Committee on the Elimination of Racial Discrimination, the treaty body responsible for monitoring the implementation of the International Convention on the Elimination of All Forms of Racial Discrimination, 7 March 1966, 660 UNTS 195, Can TS 1970 No 28 (entered into force 4 January 1969, ratified by Canada 14 October 1970) recommends that state parties avoid expulsions of non-citizens, especially of long-term residents, that would result in disproportionate interference with the right to family life: CERD, General Recommendation 30: Discrimination Against Non Citizens, 64th sess, 2004, UN Doc CERD/C/64/Misc.11/rev.3 (2004). 128 One could infer from the Court’s statement that deportation “in itself” does not engage s. 7 that the circumstances of Medovarksi’s case disclosed no fundamental personal choices that could ground a liberty claim. If this inference is correct, the Court should have made an express finding to that effect. 2017] REVISITING THE APPLICATION OF SECTION 7 337 The jurisprudence on whether immigration and refugee protection proceedings engage the liberty of non-citizens, including the Supreme Court’s sweeping conclusion in Medovarksi that deportation does not, in itself, implicate liberty is unsatisfactory because it fails to transparently address two key arguments that strongly support the engagement of non-citizens’ liberty in this context. First, liberty is engaged by the possibility of detention incidental to removal – a claim based on an analogy to penal and extradition proceedings that is particularly apt in a context where immigration enforcement is commonly used as a response to criminal activity. Second, deportation engages non-citizens’ liberty by preventing them from making fundamental personal choices beyond the bare assertion of a mobility right. In Charkaoui, the Supreme Court may have addressed this latter point by qualifying its holding in Medovarksi to allow for the engagement of s. 7 by some “features associated with deportation,” thereby “leaving the door open” for advocates to persuade immigration decision makers and courts that s. 7 is engaged by the hardships that accompany deportation, including separation from family.129 As the following section demonstrates, clarification is also sorely needed on whether immigration and refugee protection proceedings engage non-citizens’ security of the person and, particularly, on whether the right to security of the person is a 2017 CanLIIDocs 175 freestanding constitutional right that does not hinge on non-citizens’ statutory entitlements. 2. Security of the person The seminal decision on whether security of the person is engaged in the immigration and refugee protection context remains Singh v Canada (Minister of Employment and Immigration).130 Under the procedure in place under the Immigration Act, 1976, Singh’s claim that he was a Convention refugee could be denied by a decision maker who had not heard his claim in person on the basis of country conditions information to which he was not given access. Singh claimed that this statutory scheme infringed s. 7 of the Charter and advanced two arguments in support of its engagement. First, he claimed that “because a Convention refugee is, by definition, a person who has a “well-founded fear of persecution”, the refusal to give him refuge exposes him to jeopardy of death, significant diminution of his physical liberty or physical punishment in his country of origin.”131 Second, he claimed that by empowering immigration officials to detain him for purposes of examination and removal, the Immigration Act deprived him of his liberty. Justice Wilson began her analysis of whether the appellant had been deprived of life, liberty and security of the person by first “determining what rights 129 Galloway & Liew, supra note 91 at 656. 130 Singh, supra note 2. 131 Ibid at 203. 338 UNBLJ RD UN-B [VOL/TOME 68 the appellants have under the Immigration Act, 1976”132 – namely, the right to a determination from the Minister as to whether a permit should issue entitling him to enter and remain in Canada, the right not to be returned to a country where his life or freedom would be threatened and the right to appeal a removal order or a deportation order made against him – and second, by asking “whether the deprivation of these rights constitutes a deprivation of the right to life, liberty and security of the person…”133 While she acknowledged that there might be some merit in the Minister’s submission that “closing off the avenues of escape provided by the Act [did] not per se deprive a Convention refugee of the right to life or to liberty,” because it was not certain that others would deprive him of life or liberty,134 this was not the case for his right to security of the person, which “must encompass freedom from the threat of physical punishment or suffering as well as freedom from such punishment itself”: I note particularly that a Convention refugee has the right under s. 55 of the Act not to “… be removed from Canada to a country where his life or freedom would be threatened…” In my view, the denial of such a right must amount to a deprivation of security of the person within the meaning 2017 CanLIIDocs 175 of s. 7.135 Justice Wilson also recognized that, as refugee claimants, the appellants were not “at this stage entitled to certain rights as Convention refugees” but instead asserted they were entitled to fundamental justice in the determination of whether they were Convention refugees or not.136 Noting that a determination that the appellants were Convention refugees under the Act would have entitled them to the incidents of that status provided for in the Act (including the right not to be refouled), Justice Wilson concluded: Given the potential consequences for the appellants of a denial of that status if they are in fact persons with a “well-founded fear of persecution”, it seems to me unthinkable that the Charter would not apply to entitle them to fundamental justice in the adjudication of their status.137 It is noteworthy that in the underlined portion of the extract from her judgment, above, Justice Wilson deliberately tied her analysis of whether s. 7 interests were engaged to whether the appellants had been deprived of rights under the Immigration Act, 1976. A possible explanation for doing so is found in her reasons for rejecting the Minister’s invitation to adopt the approach taken by 132 Ibid at 204. 133 Ibid. 134 Ibid at 206. As will be discussed in more detail in section 3 of this part, below, the Minister’s submission in this regard is inconsistent with the modern standard of causation adopted by the Supreme Court in Bedford, supra note 34. 135 Ibid at 207 [underlining added]. 136 Ibid at 208. 137 Ibid at 210. 2017] REVISITING THE APPLICATION OF SECTION 7 339 American courts to the constitutional protection of non-citizens in immigration proceedings. In her view, an approach denying constitutional due process protections to aliens seeking entry on the ground that the power to expel or exclude them was a “fundamental sovereign attribute exercised by the Government’s political departments largely immune from judicial control”138 (a manifestation of the political questions doctrine) should not govern the application of s. 7 because: [I]n the Canadian context Parliament has in the Immigration Act, 1976 made many of the “political” determinations which American courts have been justifiably reluctant to attempt to get involved in themselves. On these appeals this Court is being asked by the appellants to accept that the substantive rights of the Convention refugees have been determined by the Immigration Act, 1976 itself and the Court need concern itself only with the question whether the procedural scheme set up by the Act for the determination of that status is consistent with the requirements of fundamental justice articulated in s. 7 of the Charter.139 2017 CanLIIDocs 175 Justice Wilson may have tied the deprivation of Charter rights to a deprivation of statutory rights in order to pre-emptively defend against claims that by applying constitutional due process norms to the political branches’ treatment of non-citizens, the Supreme Court was treading in an area of decision-making reserved to these branches. This approach relieved the Court of the need to recognize that rights to life, liberty and security of the person sprang from the Charter alone; it could find support for its decision in the fact that Parliament itself had elected to recognize substantive rights arising from the recognition of Convention refugee status. The importance placed by Justice Wilson on Singh’s rights under the Immigration Act, 1976 raised “very important questions about the extent to which s. 7 procedural claims are founded on the existence of statutory substantive rights as opposed to independent or free-standing constitutional rights.”140 Indeed, this issue was recently raised by the Federal Court of Appeal in Savunthararasa.141 The appellants’ claims to refugee protection were denied by the Refugee Protection Division (RPD) because they had failed to demonstrate that if returned to Sri Lanka, they would face a serious possibility of persecution. The appellants were scheduled to be removed from Canada. Because less than twelve months had passed since their claim for refugee protection was last rejected, they were barred from applying for a pre-removal risk assessment under s. 112(2)(b.1) of the IRPA. Claiming that new 138 Ibid at 211, citing Shaugnessy v US ex rel Mezei, 345 US 206 (1953) at 210. 139 Ibid at 212. 140 David J Mullan, Administrative Law – Cases, Text and Materials, 5th ed (Toronto: Emond Montgomery, 2003) at 229. 141 Savunthararasa, supra note 8. 340 UNBLJ RD UN-B [VOL/TOME 68 evidence of risk was available that had not been put in evidence before the RPD, they requested that their removal be deferred pending an assessment of the risks in light of the new evidence. When their requests were denied by enforcement officers of the Canada Border Services Agency, they sought judicial review of these decisions on the grounds that s. 112(2)(b.1) and the removals process violated their s. 7 rights: In the appellants’ submission, section 7 of the Charter is engaged when a person claims he would be at “risk of harm” if removed from Canada. Further, the “risk of harm” which engages section 7 is broad enough to encompass the kinds of risks assessed under both section 96 of the Act (a well-founded fear of persecution for reasons of race, religion, nationality, membership in a particular social group or political opinion) and section 97 of the Act (a risk of torture or a risk to life or a risk of cruel and unusual treatment or punishment). The appellants argue that enforcement officers do not, and are not permitted to, assess this full spectrum of risk.142 At the Federal Court, Justice Annis dismissed the applications. He determined that the appellants had not presented evidence of risks they faced that 2017 CanLIIDocs 175 could not be assessed by an enforcement officer. Moreover, in a lengthy Charter analysis, he determined that s. 112(2)(b.1) was constitutional. The Federal Court of Appeal agreed with the appellants that Justice Annis had erred in embarking on the Charter analysis without a proper evidentiary record and held that his comments and analysis on that issue were obiter dicta. However, it provided guidance on the nature of the analysis that would be required to deal with the Charter issue, assuming that an applicant for deferral could show that he or she faced a risk of harm that would not be assessed by an enforcement officer: In Singh …, in order to decide whether the appellants had been deprived of the right to life, liberty or security of the person, the Court began by determining which rights the appellants possessed under the applicable immigration legislation. … Once the rights possessed by the appellants as refugee claimants were identified, the inquiry turned to whether the deprivation of those rights constituted a deprivation of the right to life, liberty and security of the person within the meaning of section 7 of the Charter. The Court concluded that security of the person encompassed “freedom from the threat of physical punishment or suffering as well as freedom from such punishment itself”… The Court expressly left open the question of whether a more expansive approach to security of the person should be taken… Because the Court left this question open, in the context of a claim asserting a broader concept of security of the person, the Federal Court must be mindful of the need to properly analyze at the first stage of the section 7 analysis whether the removals scheme imposes limits on the security of the person, thus engaging section 7 of the Charter.143 142 Ibid at para 8. 143 Ibid at paras 27–29. 2017] REVISITING THE APPLICATION OF SECTION 7 341 Savunthararasa confirms that the Federal Court of Appeal remains mindful that Singh tied the deprivation of security of the person under s. 7 to the deprivation of rights conferred under the Immigration Act, 1976. Indeed, it had previously distinguished Singh on this basis.144 For example, in Berrahma v Canada (Minister of Employment and Citizenship),145 it considered the constitutionality of a provision whereby a refugee claimant was ineligible to have a refugee claim referred to the Refugee Division because he had filed the claim less than ninety days after first having been denied refugee status. The Court held that security of the person was not engaged and distinguished Singh as follows: As I understand it, the reason the Supreme Court concluded as it did in Singh is that, to give effect to international obligations assumed earlier, Parliament had recognized and granted foreign nationals the right to claim refugee status, but failed at the same time to create along with the exercise of this right - a right connected with the protection of life and security - a procedure consistent with the requirements of fundamental justice. That, I think, is the difference between Singh and the case of an ineligible claimant: Singh was denied a status which the law gave him the right to claim without having any opportunity of showing that he met the 2017 CanLIIDocs 175 conditions for obtaining it, whereas the ineligible claimant is not denied a status he is entitled to claim.146 The very idea that finding a deprivation of life, liberty or security of the person should hinge on proof of the existence of a right or status conferred by statute is plainly inconsistent with Justice Wilson’s criticism, in Singh,147 of the dichotomy between “rights” and “privileges” which had narrowed the scope of the application of the Canadian Bill of Rights.148 A majority of the Supreme Court had held in Mitchell v The Queen that s. 2(e) of the Bill of Rights, which provided that “no law of Canada shall be construed or applied so as to deprive a person of the right to a fair hearing in accordance with the principles of fundamental justice for the determination of his rights and obligations” did not apply to the decision of the Parole Board to revoke an individual’s parole because he “had no right to parole.” Rather, parole was granted “as a matter of discretion” and subject to revocation at the Board’s “absolute discretion.”149 This approach was consistent with the Court’s view 144 See Sharry J Aiken et al, Immigration and Refugee Law – Cases, Materials and Commentary, 2d ed (Toronto: Emond, 2015) at 242. 145 Berrahma c Canada (Ministre de l’Emploi et de l’Immigration) (1991), 132 NR 202, 25 ACWS (3d) 925 (FCA) (WL Can) [Berrahma]. 146 Ibid at para 12. See Nguyen, supra note 65 at para 9, insisting on the fact that Singh did not assist claimants found ineligible, because it dealt with “the right to claim refugee status, a right previously granted.” See also Williams, supra note 71 at para 22. 147 Singh, supra note 2 at 209. 148 Galloway & Liew, supra note 91 at 652. 342 UNBLJ RD UN-B [VOL/TOME 68 that parole revocation decisions did not attract procedural protections at common law; statutorily defined as being at the discretion of the Board, they could not be said to affect rights and were therefore not “judicial” in character.150 In a spirited dissent, Laskin C.J. attacked this failure to recognize the right of parolees to minimum procedural safeguards in parole revocation under the common law and the Bill of Rights. In his view, the application of the rules of natural justice should be determined not by the judicial character of the decision maker but by “the substantive issue that a tribunal is called upon to determine, and its consequences for the affected person, whether in respect of his person, his status or his property…”151 Relying on the United States Supreme Court’s decision in Morrisey v Brewer152 that “there was more in parole than mere privilege that could be granted or withdrawn at the pleasure of the state,” he emphasized the serious consequences of revocation for a parolee, including prolonged imprisonment, “loss of job,… loss of conditional liberty, loss of family and other association,” and concluded that parole revocation without minimum procedural safeguards breached ss. 2(c)(i) and 2(e) of the Bill of Rights. To Justice Wilson, an analysis based on the distinction between rights and privileges was not acceptable in relation to the Charter.153 She preferred Chief Justice Laskin’s dissenting opinion which focused instead “on the consequences of 2017 CanLIIDocs 175 parole revocation for the individual.”154 In other words, engagement of s. 7 should hinge on the consequences of the impugned state act on the life, liberty and security of the person interests of the individual, not on whether that act can be categorized as involving the determination of a statutory right rather than the discretionary revocation of a privilege. The judgment in Singh of Beetz, Estey and McIntyre JJ., based on s. 2(e) of the Bill of Rights, relies on the rights-privilege distinction. Having laid out the “list of legal rights given to Convention refugees in Canada by the Immigration Act, 1976 and Regulations,”155 counsel for the appellants claimed that the regime set out in the Act under which a person could claim Convention refugee status provided for a procedure “for the determination of his rights” in the meaning of s. 2(e) of the Bill of 149 Mitchell v The Queen, [1976] 2 SCR 570 at 588, 61 DLR (3d) 77 [Mitchell], cited by Justice Wilson in Singh, ibid at 209. 150 A “judicial decision” was a decision that had a conclusive effect, was adjudicative and had a serious adverse effect on rights: Howarth v Canada (National Parole Board), [1976] 1 SCR 453 at 465, 50 DLR (3d) 349 per Dickson J. 151 Mitchell, supra note 149 at 580 [emphasis added]. 152 Morrisey v Brewer (1972), 408 US 471. The Court found a violation of constitutional due process in the failure to give a parolee faced with revocation a simple factual hearing. 153 Singh, supra note 2 at 209. 154 Ibid at 210. 155 Ibid at 226–7: these included the right not to be removed to a country where life or freedom is threatened (s. 55), the right to re-enter Canada if a safe country cannot be found (s. 14(1)(c)) and the right to be considered under the criteria provided in the Regulations for “employment authorization” while residing in Canada. 2017] REVISITING THE APPLICATION OF SECTION 7 343 Rights.156 Accepting that “what is protected by the right to a fair hearing is the determination of one’s ‘rights and obligations’ whatever they are”157 Justice Beetz concluded that: [T]he process of determining the appellants’ refugee claims involved the determination of rights and obligations for which the appellants have, under s. 2(e) of the Canadian Bill of Rights, the right to a fair hearing in accordance with the principles of fundamental justice. It follows also that this case is distinguishable from cases where a mere privilege was refused or revoked, such as Prata v. Minister of Manpower and Imigration, [1976] 1 S.C.R. 376, and Mitchell v. The Queen, [1976] 2 S.C.R. 570.158 Justice Beetz recognized that s. 2(e) of the Bill of Rights was engaged in Singh only because the appellants were able to point to rights, defined in the Immigration Act, 1976, that were “determined” through the impugned refugee status determination regime. Had non-refoulement or re-entry been cast in discretionary terms as privileges instead of rights, the reasoning in Mitchell might have defeated Singh’s claim to protection under the Bill of Rights. The approach to the application of s. 7 advocated by Justice Wilson avoids such a result under the Charter because it 2017 CanLIIDocs 175 focuses on the consequences of denial of refugee status to the life, liberty and security of the person of those with a well-founded fear of persecution.159 Thus, the deprivation of security of the person which arose from the risk of harm to the non-citizen if removed from Canada was aligned with the risks created by Canada’s failure to put in place an effective process to determine whether claimants were Convention refugees (resulting in a deprivation of their statutory right to have their status determined). However, it was not contingent on the existence of this statutory right. For reasons previously explained, Justice Wilson’s cautious approach may have been understandable in the context of Singh, an early decision from a court likely divided160 on whether s. 7 applied to Parliament’s control and regulation of non-citizens. It is no longer necessary or appropriate today. 156 Ibid at 227. Indeed, the Attorney General of Canada conceded that the determination of refugee claims involved the determination of rights of refugee claimants and that it was “only in that respect that his submissions with respect to s. 2(e)… differ from his submissions with regard to section 7 of the Charter…” 157 Ibid at 228. 158 Ibid [underlining added]. 159 Her position finds resonance in the Supreme Court’s admonition, in Charkaoui, supra note 5 at para 18, that “[in] determining whether s. 7 applies, we must look at the interest at stake rather than the legal label attached to the impugned legislation.” The Supreme Court was responding to the Canadian government's blanket claim, based on Medovarksi, supra note 4, that s. 7 could not apply in immigration proceedings. 160 The fact that following the hearing of the appeal the Court asked the parties to address, through written arguments, the application of the Canadian Bill of Rights, SC 1960, c. 44 [Bill of Rights] and that three judges chose to allow the appeal based on a breach of s. 2(e) of the Bill of Rights strongly indicates that 344 UNBLJ RD UN-B [VOL/TOME 68 Support for the view that a deprivation of life, liberty or security of the person should not hinge on proof of the denial of a statutory right is also found in the Supreme Court’s subsequent decision in Suresh v Canada (Minister of Citizenship and Immigration).161 Suresh examined whether a provision of the Immigration Act authorizing the deportation of a Convention refugee on security grounds even where the refugee’s life or freedom “would be threatened” by the return violated s. 7. The Court noted that it was conceded that “deportation to torture may deprive a refugee of liberty, security and perhaps life.”162 It reiterated the principle enunciated in the extradition context that the guarantee of fundamental justice applied “even to deprivations of life, liberty or security effected by actors other than our government, if there is a sufficient causal connection between our government’s participation and the deprivation ultimately effected.”163 In other words, whether s. 7 was engaged had to take into account not only the Minister’s act of deporting but “the possibility of grievous consequences such as torture and death, if a risk of those consequences is established.”164 Notably absent in Suresh is the notion that a deprivation of security of the person is contingent on the deprivation of a right conferred in relevant legislation. Indeed, Suresh had no unqualified statutory right not to be returned to a country where his life or freedom would be threatened. As a person found 2017 CanLIIDocs 175 inadmissible on grounds of membership in an organization believed to be involved in terrorism, his right to non-refoulement was expressly subject to the minister’s broad discretion under s. 53(1)(b) of the Immigration Act to issue an opinion that he should be removed because he constituted a danger to Canada’s security. The Court recognized that deportation may involve a risk to the “fundamental right to be protected from torture or similar abuses,”165 grounded in the right not to be deprived of security of the person except in accordance with the principles of fundamental justice. To access the procedural protections of fundamental justice, the refugee needed to show that security of the person was engaged by showing not “proof of the risk of torture to that person,” but “a prima facie case that there may be a risk of torture upon deportation.”166 The Charkaoui decision also supports the view that the right not to be deprived of security of the person except in accordance with fundamental justice is a freestanding constitutional right. The Court concluded that “the appellants’ challenges to the fairness of the process leading to possible deportation and the loss the Court was divided on whether s. 7 of the Charter applied in Singh: Singh, supra note 2 at 185. Indeed, Justice Beetz refrained from expressing any view on whether the Charter was “applicable at all to the circumstances of these cases”: ibid, at 223–4. See also Dauvergne, Humanitarianism, Identity and Nation, supra note 47 at 186. 161 Suresh, supra note 23. 162 Ibid at para 44. 163 Ibid at para 52. 164 Ibid at para 52. 165 Ibid at para 127. 166 Ibid. See also Ahani v Canada (Minister of Citizenship and Immigration), 2002 SCC 2 at para 2, [2002] 1 SCR 72 [Ahani]. 2017] REVISITING THE APPLICATION OF SECTION 7 345 of liberty associated with detention raise important issues of liberty and security” and that “s. 7 of the Charter [was] engaged.”167 While it focused primarily on the impact of detention on Charkaoui’s liberty interest, the Court described the “issues of security” as follows: The detainee’s security may be further affected in various ways. The certificate process may lead to removal from Canada, to a place where his or her life or freedom would be threatened: see e.g. Singh … at p. 207, per Wilson J. A certificate may bring with it the accusation that one is a terrorist, which could cause irreparable harm to the individual, particularly if he or she is eventually deported to his or her home country. Finally, a person who is determined to be inadmissible on grounds of security loses the protection of s. 115(1) of the IRPA, which means that under s. 115(2), he or she can be deported to torture if the Minister is of the opinion that the person is a danger to the security of Canada. In Suresh …, this Court stated, at para. 76, that “barring extraordinary circumstances, deportation to torture will generally violate the principles of fundamental justice protected by s. 7 of the Charter.” … The appellants claim that they would be at risk of torture if deported to 2017 CanLIIDocs 175 their countries of origin. But in each of their cases, this remains to be proven as part of an application for protection under the provisions of Part 2 of the IRPA. The issue of deportation to torture is consequently not before us here.168 The Court’s description of the security issues engaged in Charkaoui is noteworthy for three reasons. First, a named person’s security may be affected because “the certificate process may lead to removal from Canada, to a place where his or her life or freedom would be threatened.” The Court recognizes this effect of the certificate process on security despite the fact that, at the stage of assessing the reasonableness of the certificate, removal from Canada is not inevitable.169 It does not exclude the application of s. 7 because constitutional scrutiny may be applied at a subsequent stage of the proceedings. Second, Singh is cited in support of the proposition that security interests are engaged by a named person’s removal to a place where his life or freedom would be threatened; no mention is made of the fact that, in Singh, the deprivation of security of the person was linked to the denial of Singh’s statutory right to non-refoulement. As the Court notes, named persons determined to be inadmissible on security grounds do not benefit from a statutory protection against refoulement. Third, the Court’s observation that the appellants’ claim that they would be at risk of torture if deported to their countries of origin “remains to be proven as part of an application for protection under the provisions of Part 2 of the IRPA” and that the “issue of deportation to torture is consequently not 167 Charkaoui, supra note 5 at para 18. 168 Ibid at paras 14–15. 169 Ibid at para 14. See also para 18, where the Court associates Charkaoui’s “possible deportation” with engagement of his security interest. 346 UNBLJ RD UN-B [VOL/TOME 68 before us”170 should not be taken to mean that the Court found that the appellants’ security of the person was not engaged by the certificate process.171 The Court clearly stated that features associated with deportation such as detention in the course of the certificate process “or the prospect of deportation to torture” may engage s. 7 and ultimately concluded that s. 7 was engaged because the process raised “important issues of liberty and security.”172 Thus, the engagement of a security interest would not appear to be contingent on the outcome of the application for protection. The Court’s reference to the risk of torture remaining to be proven must also be read subject to its admonition, in Suresh, that engagement of security of the person did not require a “proof of the risk of torture” but a prima facie case that there may be a risk.173 Based on the preceding analysis, it is open to non-citizens to claim that their s. 7 right to security of the person is engaged in circumstances where deportation places them at risk of persecution, torture or cruel and unusual treatment or punishment, whether or not exposing them to this risk of harm also violates their statutory rights. In other words, the right not to be deprived of one’s security of the person except in accordance with the principles of fundamental justice is a 2017 CanLIIDocs 175 freestanding constitutional right. What of the impact of Medovarksi? I earlier argued that a possible, though unarticulated, justification for the Court’s finding that the deportation of non-citizens cannot in itself implicate s. 7 interests – that deportation does not violate the mobility rights of non-citizens and thus cannot violate other Charter rights – does not stand up to scrutiny. In any event, the Supreme Court has qualified this finding by allowing in Charkaoui that deportation proceedings were not immune from s. 7 scrutiny and that “some features associated with deportation, such as detention in the course of the certificate process or the prospect of deportation to torture” may engage s. 7. In other words, in certain circumstances, the nature of the impact of deportation on an affected person may trigger that person’s security of the person interests, whether in its physical or psychological dimensions. Thus, where the deportation of individuals suffering from medical conditions “clinically significant to their current and future health” would deprive them of access to essential health care, security of the person should be engaged as it was for those citizens denied access to timely health care in Chaoulli.174 Lorne Waldman suggests that, in this sense, Medovarksi and Charkaoui could be read consistently with the Supreme Court’s approach to the 170 Ibid at para 15. 171 The Federal Court of Appeal appeared to suggest as much in JP v Canada (Minister of Public Safety and Emergency Preparedness); B306 v Canada (Minister of Public Safety and Emergency Preparedness); Hernandez v Canada (Minister of Public Safety and Emergency Preparedness), 2013 FCA 262 at paras 120–22, 368 DLR (4th) 524 [B306]. 172 Ibid at para 18 [emphasis added]. 173 Suresh, supra note 23 at para 127. 174 See, on this question, Lorne Waldman, “The Charter of Rights and its Application in Immigration Proceedings”, (Paper delivered at the 10th Annual Conference of the Canadian Association of Refugee Lawyers, Toronto, 21 March 2016) at 39–41 [unpublished]. 2017] REVISITING THE APPLICATION OF SECTION 7 347 application of s. 7 in Blencoe, “where the Court held that in a non-criminal context, questions of the engagement of s. 7 must be considered on a case-by-case basis considering the serious impact of the state-imposed psychological stress on the individual.”175 It is certainly conceivable that deportation may produce a serious and profound effect on a person’s psychological integrity by interfering in profoundly intimate and personal choices, including a parent’s interest in raising and caring for a child, recognized by the Supreme Court as engaging security of the person in the context of child custody proceedings.176 I have argued that the Supreme Court should approach the question of whether proceedings leading to the deportation of non-citizens engage their liberty and security of the person in a manner consistent with its broad definition of these interests in contexts other than immigration and refugee protection. Similarly, in the next section, I claim that the Court’s apparent refusal to find that s. 7 is engaged by decisions under the Immigration and Refugee Protection Act that do not immediately precede removal conflicts with the more relaxed standard of causation it adopted in Bedford. 2017 CanLIIDocs 175 3. Causation Canadian courts’ response to non-citizens’ claim that their liberty or security of the person are engaged by decisions taken at preliminary stages of the administrative process eventually leading to removal is that s. 7 does not apply because these interests are more directly engaged and considered in the stages of this process that immediately precede removal. This prematurity “principle” was best described by Justice John Evans, then a judge of the Federal Court Trial Division, in Jekula v Canada (Minister of Citizenship and Immigration).177 Jekula, a Liberian citizen recognized in Sierra Leone as a refugee, claimed refugee status in Canada but was found ineligible as a person recognized as a Convention refugee by a country other than Canada to which he could be returned. An exclusion order was issued against him on the grounds he did not have authorization to remain in Canada. Justice Evans 175 Ibid at 25. See also Galloway & Liew, supra note 91 at 656. 176 G(J), supra note 14. The contrary outcome, which prevails under American jurisprudence, is that parents have constitutional rights where “the state seeks to take their children” but no such rights “if they or their children face separation as a result of one or the other’s deportation” – an odd result, as Kanstroom notes, which flows from the lack of a “unified theory of constitutional punishment”: Kanstroom, supra note 118 at 1934. An argument could be made that s. 7 cannot be engaged by the psychological impact of deportation on parents who “choose” to leave their Canadian-born children in Canada because, due to their intervening choice, the state would not be directly responsible for the interference with their ability to nurture their children. This argument ignores that, under the test for causation set out in Bedford, supra note 34 at para 76, parents need show only a sufficient causal connection between the state’s action and the prejudice they have suffered, not that the state action is the only or even the dominant cause. 177 Jeluka v Canada (Minister of Citizenship and Immigration), [1999] 1 FCR 266, 47 Imm LR (2d) 218 [Jekula]. 348 UNBLJ RD UN-B [VOL/TOME 68 held that the first step in a s. 7 analysis was to ask whether the “administrative action under review… deprive[d] the applicant of the right to life, liberty and security of the person.”178 In his opinion, the eligibility decision did not have this effect: First, while it is true that a finding of ineligibility deprives the claimant of access to an important right, namely the right to have a claim determined by the Refugee Division, this right is not included in “the right to life, liberty and security of the person”: Berrahma […] at page 213; Nguyen […] Second, it may well be a breach of the rights protected by section 7 for the government to return a non-citizen to a country where she fears that she is likely to be subjected to physical violence or imprisoned. However, a determination that a refugee claimant is not eligible to have access to the Refugee Division is merely one step in the administrative process that may lead eventually to removal from Canada. The procedure followed at the risk assessment to which the applicant will be entitled under section 53 before she is removed can be subject to constitutional scrutiny to ensure that it complies with the principles of fundamental justice, even though the procedure is not prescribed in the Act or regulations: Kaberuka […] at page 271. Moreover, while holding that it was not inconsistent with 2017 CanLIIDocs 175 section 7 for the Immigration Act to limit access to the Refugee Division, Marceau J.A. also said in Nguyen […] at pages 708-709: It would be my opinion, however, that the Minister would act in direct violation of the Charter if he purported to execute a deportation order by forcing the individual concerned back to a country where, on the evidence, torture and possibly death will be inflicted. It would be, it seems to me . . . at the very least, an outrage to public standards of decency, in violation of the principles of fundamental justice under section 7 of the Charter. In summary, section 7 rights are not engaged at the eligibility determination and exclusion order stages of the process. However, the applicant cannot be lawfully removed from Canada without an assessment of the risks that she may face if returned to Sierra Leone. And the manner in which that assessment is conducted must comply with the principles of fundamental justice.179 While Justice Evans relies on Nguyen for the proposition that s. 7 is not engaged at the eligibility determination and exclusion order stages of the process set out under the Immigration Act, his judgment does not advert to the fact that, as noted earlier, both the Federal Court of Appeal in Nguyen and the Federal Court Trial Division in Kaberuka, also cited in Jekula, had concluded that the scheme as a whole did in fact engage s. 7 of the Charter. However, Justice Evans’ decision is most remarkable because it segments the “administrative process that may lead eventually to removal from Canada” into discrete steps and posits that s. 7 should only apply to those steps which immediately precede the applicant’s deportation. This approach was later taken up by the Federal Court of Appeal in deciding, in Poshteh v Canada 178 Ibid at para 31. 179 Ibid at paras 31–33. 2017] REVISITING THE APPLICATION OF SECTION 7 349 (Minister of Citizenship and Immigration),180 that s. 7 was not engaged by a determination of inadmissibility. The Immigration Division of the IRB had determined that Poshteh was inadmissible to Canada under s. 34(1)(f) of IRPA on the grounds that, as a child, he had been a member of a terrorist organization in Iran. While the appeal was mainly concerned with whether Poshteh was a member of the organization and whether his status as a minor was relevant to this determination, Justice Rothstein commented on Poshteh’s claim that, even though his life, liberty and security of the person were not engaged in the proceeding, IRPA should “be interpreted in a manner consistent with the principles of fundamental justice.” Justice Rothstein determined that the inadmissibility decision did not engage s. 7, relying on Barrera (a problematic precedent, as noted earlier181) but also pointing to the fact that other proceedings were more proximate to his deportation: [A]ll that is being determined is whether Mr. Poshteh is inadmissible to Canada on the grounds of his membership in a terrorist organization. The authorities are to the effect that a finding of inadmissibility does not engage an individual’s section 7 Charter rights. (See, for example, Barrera v. Canada (MCI) (1992), 99 D.L.R. (4th) 264 (F.C.A.).) A number of proceedings may yet take place before he reaches the stage at 2017 CanLIIDocs 175 which his deportation from Canada may occur. For example, Mr. Poshteh may invoke subsection 34(2) to try to satisfy the Minister that his presence in Canada is not detrimental to the national interest. Therefore, fundamental justice in section 7 of the Charter is not of application in the determination to be made under paragraph 34(1)(f) of the Act.182 This reasoning183 was recently on display in obiter statements by the Supreme Court in B010 v Canada (Citizenship and Immigration).184 B010 and other Tamil refugee claimants from Sri Lanka arrived in Canada on a dilapidated cargo ship. The Immigration and Refugee Board (Immigration Division) found them inadmissible under s. 37(1)(b) of the IRPA on grounds of organized criminal people smuggling. As a result, their refugee claims were ineligible to be referred to the RPD for consideration on their merits.185 For a unanimous court, Chief Justice McLachlin concluded that s. 37(1)(b) targeted “procuring illegal entry in order to obtain, directly or indirectly, a financial or other material benefit in the context of transnational 180 Poshteh v Canada (Minister of Citizenship and Immigration), 2005 FCA 85, [2005] FCR 487 [Poshteh]. 181 See the text accompanying note 64. 182 Poshteh, supra note 180 at para 63. 183 See also Torre v Canada (Minister of Citizenship and Immigration), 2016 FCA 48 at para 4; and Brar v Canada (Minister of Public Safety and Emergency Preparedness), 2016 FC 1214 at para 21, 273 ACWS (3d) 603. 184 B010, supra note 6. 185 IRPA, supra note 25 at s 101(1)(f); B010, supra note 6 at para 14. 350 UNBLJ RD UN-B [VOL/TOME 68 organized crime.”186 The appellants, who merely aided in the illegal entry of other asylum-seekers in the course of their collective flight to safety, were not “people smugglers.”187 Accordingly, while it decided that it was unnecessary to address the appellants’ alternative argument that s. 37(1)(b) was overbroad and violated s. 7 of the Charter, the Court noted in obiter that this argument could not assist them “as s. 7 of the Charter is not engaged at the stage of determining admissibility under s. 37(1)”: This Court recently held in Febles v. Canada (Citizenship and Immigration), 2014 SCC 68, [2014] 3 S.C.R. 431, that a determination of exclusion from refugee protection under the IRPA did not engage s. 7, because “even if excluded from refugee protection, the appellant is able to apply for a stay of removal to a place if he would face death, torture or cruel and unusual treatment or punishment if removed to that place” (para. 67). It is at this subsequent pre-removal risk assessment stage of the IRPA’s refugee protection process that s. 7 is typically engaged. The rationale from Febles, which concerned determinations of “exclusion” from refugee status, applies equally to determinations of “inadmissibility” 188 to refugee status under the IRPA. 2017 CanLIIDocs 175 The Court’s reliance on Febles as authority for the proposition that “exclusion from refugee protection under the IRPA did not engage s. 7” is problematic. Febles focused on the interpretation of article 1F(b) of the Refugee Convention189 which excludes from the protection of the Convention any person with respect to whom there are serious reasons for considering that “he has committed a serious non-political crime outside the country of refuge prior to his admission to that country as a refugee.” Article 1F(b) was directly incorporated into Canadian law through s. 98 of IRPA, which provides that a person excluded under sections E or F of Article 1 of the Refugee Convention is not a Convention refugee or person in need of protection. Responding to Febles’ argument that a narrow interpretation of s. 98 should be adopted because it was consistent with the Charter, the Supreme Court held that its broader interpretation of the provision was consistent with the Charter, since excluded persons could apply for a pre-removal risk assessment or could challenge their removal to a country where their Charter rights are jeopardized pursuant to the principles set out in Suresh:190 While the appellant would prefer to be granted refugee protection than have to apply for a stay of removal, the Charter does not give a positive right to refugee protection. The appellant is excluded from refugee protection as a result of his commission of serious non-political crimes. If 186 B010, supra note 6 at para 72. 187 Ibid. 188 Ibid at para 75. 189 Convention Relating to the Status of Refugees, 4 June 1969, Can TS 1969 No 6. 190 Febles v Canada (Minister of Citizenship and Immigration), 2014 SCC 68 at para 67, [2014] 3 SCR 431 [Febles]. 2017] REVISITING THE APPLICATION OF SECTION 7 351 removal of the appellant to Cuba jeopardizes his Charter rights, his recourse is to seek a stay of removal, as discussed earlier.191 While Febles may be read as affirming that a provision restricting the authority of the IRB to grant refugee status to excluded persons does not in itself violate their s. 7 rights, the Court did not expressly find that s. 7 was not even engaged because their liberty or security of the person were not engaged. It certainly supplied no reasoning to justify such a conclusion. A more plausible reading of Febles is that while s. 7 of the Charter may be engaged by the process to which Febles was subjected, he had not shown that his exclusion from proceedings that would result in a grant of refugee protection violated fundamental justice because the IRPA provided alternate avenues by which his security of the person interests could be addressed and protected. Is there another basis, apart from Febles, for the Court’s obiter views in B010 that s. 7 is not engaged in determinations of inadmissibility to or exclusion from refugee status? The Court’s rationale mirrors that set out by the Federal Court of Appeal in one of the two decisions appealed from in B010.192 Quoting at length from Jekula, the Federal Court of Appeal had determined that s. 7 of the Charter would only be engaged at “a stage under the process in IRPA which is subsequent to the 2017 CanLIIDocs 175 inadmissibility finding.”193 What can we make of the claim that s. 7 does not apply in the IRPA’s administrative process so long as other “steps” or proceedings are available before a non-citizen reaches the stage at which deportation from Canada may occur? As revealed in the discussion of decision making in the extradition and penal contexts in section 2, above, this logic has not prevailed in the context of other multi-stage proceedings that may result in detention or imprisonment. What reasoning underlies the claim that an ineligibility determination or a finding of inadmissibility does not attract s. 7 protection because it is merely one step in the administrative process that may lead eventually to removal from Canada, with others to follow? The argument appears to be that s. 7 is not engaged at that step because there are steps later in the process more directly and foreseeably linked to a deprivation of a non-citizen’s s. 7 interests where the person’s circumstances can be scrutinized to ensure that this deprivation complies with the principles of fundamental justice. This reasoning implies a standard of causation more onerous than the “sufficient causal connection” standard adopted by the Supreme Court in Bedford. It requires that state action be a foreseeable and necessary cause of the prejudice to the person’s s. 7 interests – a standard expressly rejected in Bedford. It is instructive to contrast this approach to Justice Marceau’s decision, in Nguyen, to examine the eligibility determination in the context of the scheme viewed as a whole: 191 Ibid at para 68. 192 B306, supra note 171. 193 Ibid at para 125. 352 UNBLJ RD UN-B [VOL/TOME 68 [W]hile a determination of ineligibility… is only indirectly linked to the deportation order, nevertheless it has the effect of taking away the only possible barrier to the issuance of an unconditional deportation order, and as such participates in the deprivation of liberty and, possibly, the security of the individual which results from deportation.194 Justice Marceau’s approach is consistent with the “sufficient causal connection” test which, “sensitive to the context of the particular case,” does not require that the impugned government action – here, the eligibility determination – be the only or the dominant cause of the claimant’s prejudice. It is also consistent with the Supreme Court’s own approach in Charkaoui, where it raised the impact of the security certificate process on Charkaoui’s security of the person, recognizing that s. 7 was engaged because this process raised “important issues of liberty and security” despite the fact that removal from Canada was not inevitable at the stage of the proceedings subject to constitutional scrutiny in that case – the designated judge’s assessment of the security certificate’s reasonableness. The approach to the engagement of s. 7 that underlies Jekula and the Supreme Court’s obiter comments in B010 is not, as required by Bedford, sensitive 2017 CanLIIDocs 175 to the context of the particular case. It artificially reduces the “immigration context” to a set of discrete processes whose impact on non-citizens’ liberty and security of the person can be analyzed independently and in isolation from the overarching regime of immigration control to which they are subjected under IRPA. In Canadian Doctors for Refugee Care v Canada (Attorney General),195 Federal Court Justice Anne Mactavish described the immigration context in more realistic terms. One of the issues in that case was whether the Government of Canada’s decision to withdraw health care coverage for certain refugee claimants constituted “treatment” for the purposes of the prohibition in s. 12 of the Charter against subjecting individuals to “any cruel and unusual treatment or punishment.” Justice Mactavish noted that to constitute “treatment”, positive actions, inaction or prohibitions by the state affecting a rights claimant had to be part of an active state process involving an exercise of state control over that individual:196 In this case, those seeking the protection of Canada are under immigration jurisdiction, and as such are effectively under the administrative control of the state. Some claimants may be detained, and obligations such as reporting requirements may be imposed upon others. In addition, their rights and opportunities (such as their right to work or their ability to receive social assistance benefits) may be limited in a number of different ways by the state. Indeed, their entitlement to a range of benefits is wholly dependent upon decisions made by various branches of the Government of Canada as to their right to seek protection, and the ultimate success of their claims for protection.197 194 Nguyen, supra note 65 at para 10. 195 Canadian Doctors for Refugee Care, supra note 99. 196 Ibid at paras 582–583. 197 Ibid at para 585. 2017] REVISITING THE APPLICATION OF SECTION 7 353 Canada’s immigration law can more realistically be seen as an instrument of social control, with deportation as a “method of continual control” of non-citizens’ behaviour.198 Under this model, proceedings under IRPA can be usefully compared to a system of railway lines, some of which, as Justice Evans observes, ultimately lead to removal from Canada. Along the way, switches or turnouts could allow the train to move from the mainline, heading towards removal, to a secondary line or even to a siding. These correspond to various proceedings, like eligibility, that provide opportunities to avoid removal through a process by which they may gain refugee protection or that require decision-makers to consider additional factors that could weigh against removal and that might not be considered at a later stage. As Justice Marceau recognized, when these switches are closed and a non-citizen is denied access to these proceedings, the likelihood of removal and the risk of deprivation of life, liberty or security of the person increase. This increased risk meets the standard of causation required by the Supreme Court in Bedford to establish engagement of s. 7. What may animate decisions like Jekula is the courts’ concern that if they recognize that s. 7 is engaged by decisions made at each and every stage prior to 2017 CanLIIDocs 175 removal, non-citizens will seek to judicially review every decision on the ground that it infringes their rights to liberty and security of the person, paralyzing IRPA’s enforcement. This is by no means the inevitable or even likely outcome of recognizing s. 7 engagement through a principled application of the s. 7 framework developed by the Supreme Court in other contexts, including the Bedford standard of causation. The Federal Court has the discretion to refuse to entertain an application for judicial review where an adequate alternative remedy in the form of an internal or external appeal or other statutory mechanism is available to the applicant.199 Rather than holding that s. 7 is not engaged in immigration and refugee protection proceedings that do not immediately precede removal, a position inconsistent with the standard of causation adopted by the Supreme Court, a court could decline to entertain an application for judicial review based on s. 7 of the Charter on the ground that the applicant’s s. 7 rights to life, liberty or security of the person would be considered in a subsequent adequate alternative proceeding. However, before dismissing an application on this ground, the court would have to satisfy itself that this proceeding was “adequate”, providing the non-citizen with a fair hearing before a decision maker with the independence and statutory authority to substantially address life, liberty and security of the person claims and to provide an appropriate remedy.200 198 Kanstroom, supra note 118 at 1898. 199 David J Mullan, “The Discretionary Nature of Judicial Review” in RJ Sharpe & K Roach, eds, Taking Remedies Seriously: 2009 (Montréal: Canadian Institute for the Administration of Justice, 2010) 420 at 432. Parliament has preserved the traditionally discretionary nature of judicial review through its use of permissive language in s 18.1(3) of the Federal Courts Act, RSC 1985, c F-7. 354 UNBLJ RD UN-B [VOL/TOME 68 IV. Conclusion The Supreme Court’s current approach to the application of s. 7 in the immigration and refugee protection context is inconsistent with its approach to s. 7 engagement in other legal regimes. No principled and transparent reasons have yet been offered to justify this discrepancy. Liberty is engaged in removal proceedings under IRPA because this statute effectively establishes an administrative regime to control non- citizens in large measure through the threat of their forced removal from Canada and exposes them to the possibility of detention in order to carry out this threat. Moreover, deportation may in certain circumstances engage non-citizens’ liberty in its broad sense by preventing them from making fundamental personal choices that go beyond the bare assertion of a right to mobility. Non-citizens’ security of the person is engaged where deportation would place them at risk of physical or serious and profound psychological harm, including that caused by the resulting interference with their profoundly intimate and personal choices, regardless of whether this also involves the breach of their statutory rights. Finally, as in other contexts where there is a risk of state deprivation of liberty or security of the person, and consistently with the relaxed standard of causation adopted by the Supreme Court in Bedford, courts 2017 CanLIIDocs 175 should recognize that these s. 7 interests are engaged in the early stages of the administrative process and not only at the stage most proximate to deportation. Will a principled approach to the application of s. 7 make any real difference for non-citizens seeking to challenge their removal from Canada? After all, in a legion of cases, including Chiarelli and Medovarski, courts have held that, even assuming that liberty and security of the person are engaged, the removal of non-citizens would not breach any principle of fundamental justice. As Binnie and LeBel JJ. observed in Chaoulli: Claimants whose life, liberty or security of the person is put at risk are entitled to relief only to the extent that their complaint arises from a breach of an identifiable principle of fundamental justice. The real control over the scope and operation of s. 7 is to be found in the requirement that the applicant identify a violation of a principle of fundamental justice.201 At its most basic level, my argument is that courts should consistently apply the same principles to define the scope of the life, liberty and security of the person interests of citizens and non-citizens because these relate to our basic and common humanity – the same essential insight that underlies Justice Wilson’s decision, in Singh, that s. 7 applies to every human being present in Canada and thus amenable to Canadian law.202 A review of the jurisprudence on the application of s. 7 in the 200 For a recent review of courts’ discretionary remedial powers on judicial review, see Gerald Heckman, “Developments in Remedial Discretion on Judicial Review: Prematurity and Adequate Alternative Remedies” (2017) 30:1 CJALP 1. 201 Chaoulli, supra note 14 at para 199. 202 The United Nations Committee on Human Rights has stated that “[a]liens have the full right to liberty and security of the person” and that while the International Covenant on Civil and Political Rights does not “recognize their right to enter or reside in the territory of a state party,” they may enjoy its protection 2017] REVISITING THE APPLICATION OF SECTION 7 355 immigration and refugee protection context reveals no principled or compelling justification for a contrary view. Once they recognize that s. 7 is engaged by the deportation of non-citizens, courts can address the real question: whether deportation is fundamentally just in individual cases. It is to this question that I devote my concluding remarks. While it may be true that the scope of fundamental justice – whether the deprivation of a non-citizen’s liberty or security of the person by the state is just or unjust – is influenced by the “immigration context,”203 including the power of the state, subject to international norms, to decide who it will admit to its territory, the impact of “context” on fundamental justice has its limits. Principles of fundamental justice “set out minimum requirements that a law that negatively impacts on a person’s life, liberty or security must meet”204 and are about “the basic values underpinning our constitutional order”: “The s. 7 analysis is concerned with capturing inherently bad laws: that is, laws that take away life, liberty or security of the person in a way that runs afoul of our basic values. The principles of fundamental justice are an attempt to capture those values.”205 2017 CanLIIDocs 175 In Bedford, the Court was concerned with the basic values against arbitrariness, overbreadth and gross disproportionality. Laws or state acts run afoul of these basic values when the means by which the state seeks to attain its objective is fundamentally flawed because its effects on s. 7 interests are not rationally connected, in whole or in part, or grossly disproportionate to their objective. While the immigration context may supply the various state objectives against which rational connection and proportionality are to be measured, it does not alter the basic values at play. The Court’s view that Parliament’s choice to deport a non-citizen convicted of a serious crime is not arbitrary206 may be defensible in light of the IRPA’s objective of maintaining the security of Canadian society.207 However, the conclusion that fundamental justice does not require consideration of any mitigating circumstances beyond the non-citizen’s conviction overlooks the fact that deportation may impact his s. 7 interests in a manner grossly disproportionate to the “even in relation to entry or residence, for example, when considerations of non-discrimination, prohibition of inhuman treatment and respect for family life arise”: Human Rights Committee, General Comment 15, The Position of Aliens under the Covenant (27th sess 1986), available in Compilation of General Comments and General Recommendations adopted by Human Rights Treaty Bodies (1994) UN Doc HRI/GEN/I/Rev. 9 at 189, paras 5, 7. 203 Chiarelli, supra note 46 at 733. 204 Bedford, supra note 34 at para 94. 205 Ibid at para 96. 206 Chiarelli, supra note 46 at 734. 207 IRPA, supra note 25 at s 3(h); Medovarksi, supra note 4 at paras 9–10. 356 UNBLJ RD UN-B [VOL/TOME 68 state’s purpose and violate a basic value underpinning Canada’s constitutional order.208 The impact of context on procedural norms of fundamental justice is similarly limited. The principles of fundamental justice demand, at a minimum, compliance with the common law duty of procedural fairness. The specific procedural safeguards they require depend on several factors linked to the “context of the statute involved and the rights affected.”209 The Supreme Court has stated, for example, that to conform to fundamental justice, security certificate procedures “must reflect the exigencies of the security context”210 and the need to protect information and evidence critical to national security,211 militating in favour of more limited disclosure to the named person. However, it has recognized that “the seriousness of the individual interests at stake”212 also form part of the contextual analysis and that the principles of fundamental justice cannot be reduced to the point where they “cease to provide the protection of due process that lies at the heart of s. 7 of the Charter.”213 Liberty and security of the person may be engaged where individuals are 2017 CanLIIDocs 175 subjected to the threat of detention or other statutory compulsions or to laws or government acts that adversely impact their physical and psychological well-being or interfere with inherently personal choices that go to the core of what it means to enjoy individual dignity and independence. The fact that the legal authority for such compulsions and government acts is found in the IRPA should have no bearing on whether s. 7 of the Charter is engaged. Canadian courts should recognize non- citizens’ full right to liberty and security of the person under the Charter and focus on the key question in immigration and refugee protection decision making: whether the state has interfered with those fundamental interests pursuant to a fair process and in a manner rationally connected and proportionate to the objectives of Canada’s immigration laws. 208 See Galloway & Lieu, supra note 91 at 659. 209 Suresh, supra note 23 at paras 113, 115. These include the nature of the decision made and the procedures followed in making it; the role of the particular decision within the statutory scheme, including the existence of an appeal; the importance of the decision to the individual affected; the legitimate expectations of the person challenging the decision where undertakings were made concerning the procedure to be followed; and the choice of procedure made by the agency itself. 210 Charkaoui, supra note 5 at para 27. 211 Canada (Citizenship and Immigration) v Harkat, 2014 SCC 37 at para 42, [2014] 2 SCR 33. 212 Charkaoui, supra note 5 at para 25. 213 Ibid at para 27. THE DUTY OF FAIRNESS IN THE INVESTIGATIVE STAGE OF ADMINISTRATIVE PROCEEDINGS Lucie LaBoissonnière* Introduction Administrative bodies and tribunals may be subject to procedural requirements within their statutory scheme and statutory powers procedures acts. In other respects, administrative bodies are generally found to be masters of their own procedures.1 This potentially allows for wide latitude with respect to process. However, this is always subject to a requirement to ensure the common law rules of natural justice are recognized and applied at the appropriate stages of any proceeding.2 This comment will address how the courts have interpreted the duty of 2017 CanLIIDocs 175 fairness as it relates to investigations in the context of administrative proceedings. At what stage does the duty of fairness arise? The common law duty of fairness ensures correct results are reached when public authorities, regulators and tribunal members make administrative decisions affecting the rights, interests and privileges of an individual. However, the duty is more than a means of encouraging better results. In fact, it has been defined as an independent, unqualified right that any party affected by an administrative decision is entitled to.3 Furthermore, the denial of a right to a fair process must always invalidate a decision, whether or not it may appear to a reviewing court that the hearing would likely have resulted in a different decision.4 The Baker case has confirmed that the requirements of the duty of procedural fairness in the context of administrative proceedings are determined by a fact-specific analysis and based on a number of well-known factors, namely: the nature of the decision and the process followed in making it; the nature of the statutory scheme; the importance of the decision to the person affected by it; the legitimate expectations of the parties and the procedure chosen by the tribunal.5 * Lawyer, McInnes Cooper, Moncton, NB. 1 Prassad v Canada (Minister of Employment and Immigration), [1989] 1 SCR 560 at para 16, 57 DLR (4th) 663; TWU v Telus Corp, 2004 CIRB 277 at paras 66–67. 2 Nova Scotia (Human Rights Commission) v Sam’s Place (2000), 185 NSR (2d) 372 at para 8, 3 CCEL (3d) 131. 3 Cardinal v Director of Kent Institution, [1985] 2 SCR 643 at para 23, 69 BCLR 255. 4 Ibid at para 23. 358 UNBLJ RD UN-B [VOL/TOME 68 But before an administrative decision is made, an investigator may be tasked with a fact-finding role, interviewing witnesses, gathering evidence and making conclusions, perhaps even recommendations to a decision-maker. While the duty of fairness may have limited application to certain non-final administrative decisions and investigations, there are compelling reasons for recognizing that in some cases, the duty of fairness should be broad. Individuals impacted by an administrative decision have the right to hear the other side and to respond to evidence and arguments of other participants and of the decision-maker. This right, an expression of audi alteram partem, exists at all times and is context-sensitive. In certain cases, those sensitive times occur in the investigative phase of the administrative process. In Irvine v Canada (Restrictive Trade Practices Commission), the Supreme Court of Canada found that courts should not intervene in procedural matters where an investigator is involved in the first stage of an investigation if there are sufficient safeguards later in the process to protect all of the parties involved.6 In that matter, the Court was asked to decide if a party being investigated had the right to have his counsel cross-examine witnesses in the course of a hearing inquiry. The Court, per 2017 CanLIIDocs 175 Estey J., found that there was a duty of fairness in the investigative phase, yet its extent depended on the context: Fairness is a flexible concept and its content varies depending on the nature of the inquiry and the consequences for the individuals involved. The characteristics of the proceeding, the nature of the resulting report and its circulation to the public, and the penalties which will result when events succeeding the report are put in train will determine the extent of the right to counsel and, where counsel is authorized by statute without further directive, the role of such counsel. The investigating body must control its own procedure. When that body has determinative powers, different considerations enter the process. The case against the investigated must be made known to him.7 Further, the Supreme Court explained that the duty of fairness will depend on the potential consequences and outcomes: Courts must, in the exercise of this discretion, remain alert to the danger of unduly burdening and complicating the law enforcement investigative process. Where that process is in embryonic form engaged in the gathering of the raw material for further consideration, the inclination of the courts is away from intervention. Where, on the other hand, the investigation is conducted by a body seized of powers to determine, in a final sense or in the sense that 5 Baker v Canada (Minister of Citizenship & Immigration), [1999] 2 SCR 817 at paras 210–216, 174 DLR (4th) 193. 6 Irvine v Canada (Restrictive Trade Practices Commission), [1987] 1 SCR 181 at para 78, 41 DLR (4th) 429. 7 Ibid. 2017] THE DUTY OF FAIRNESS IN THE INVESTIGATIVE STAGE 359 detrimental impact may be suffered by the individual, the courts are more inclined to intervene.8 In Ruffo v Conseil de la magistrature, the Supreme Court confirmed that the duty to act fairly and the audi alteram partem rule apply to preliminary stages of administrative proceedings. The scope of the requirements varies depending on the circumstances of each case.9 It also found relevant the fact that the alleged failure to ensure fairness at the preliminary stage would not have real consequences on the person, given the further steps provided in the process, including the appointment of an inquiry committee and the right to be heard from that committee. In Irvine, the Supreme Court emphasized that the remaining steps in the administrative process would cure any unfairness potentially caused by the inability to cross-examine witnesses in the investigative process. In Ruffo, the Supreme Court rather focused on the fact that the inability to be heard at the initial examination of the complaint did not have any real consequences on the person being investigated, due to a number of circumstances specific to that case. 2017 CanLIIDocs 175 In Swanson v Institute of Chartered Accountants (Professional Conduct Committee),10 the Manitoba Court of Queen’s Bench found that there was a duty of fairness owed to an accountant by a professional conduct committee. Further, it was found that the committee’s decision to recommend that a disciplinary hearing be held was a decision that was subject to judicial review. The Court concluded: In determining whether the decision of the professional conduct committee is subject to judicial review, the relevant focus is not on whether the committee has made a final determination. The focus is on the relationship between the committee and Mr. Swanson, with particular attention to whether he could be affected adversely by the committee’s decision.11 Some courts still decide the scope of the duty of fairness, not by considering the potential impact and the rights at stake, but by drawing a line between the investigative stage and the adjudicative stage. A recent example is MK Engineering v Assn of Professional Engineers and Geoscientists of Alberta.12 In that case, the Alberta Court of Appeal adjourned the appeal of the appeal board of a regulatory body as being 8 Ibid at para 87. 9 Ruffo v Conseil de la magistrature, [1995] 4 SCR 267 at para 89, 130 DLR (4th) 1. 10 Swanson v Institute of Chartered Accountants (Professional Conduct Committee), 2007 SKQB 480, 308 Sask R 32 [Swanson]. 11 Ibid at para 26. Also, in Mitten v College of Alberta Psychologists, 2010 ABCA 159, 26 Alta LR (5th) 102, it was found that a decision by a committee not to proceed to a hearing is a just extension of the investigative process. The decision was subject to judicial review to assess its fairness. 12 MK Engineering Inc v Assn of Professional Engineers and Geoscientists of Alberta, 2014 ABCA 58, [2014] AJ No 119 (QL). 360 UNBLJ RD UN-B [VOL/TOME 68 premature because the administrative proceedings were not yet completed. On the issue of procedural irregularities, the Court stated: “It is well established that duties of fairness at the investigative stage are qualitatively lower than duties owed at the adjudicative stage … that errors at the investigative stage can be corrected.”13 That distinction may not be appropriate since it assumes that errors in the investigative stage can be corrected. Whether procedural irregularities can be cured in the subsequent steps depends on the processes followed, which can vary significantly. Moreover, not all investigations are alike. An investigator’s role may amount to more than accumulating evidence and therefore attract a higher duty of fairness in an investigation, for instance where conclusions and recommendations are made. This issue was addressed in Provincial Health Services Authority v British Columbia (Information and Privacy Commissioner). In that case, the Court found that an investigator was exercising a quasi-judicial function when she investigated into a human rights complaint because she had to weigh and assess evidence as well as form an opinion on whether there had been a human rights violation.14 2017 CanLIIDocs 175 A contextual analysis is preferable, as was articulated in Hawrish v Cundall on the issue of procedural fairness in the investigative stage, where the court summarized as follows: From my reading of all these decisions and several others I draw the following: (1) It is of no value to go through the process of determining whether a statutory body or tribunal is judicial, quasi-judicial or administrative in deciding if it must act with fairness; (2) The duty to act fairly arises whenever such body has the power to make a decision which will affect the rights of an individual; (3) Whether a particular body has such power or function is to be ascertained from the whole of the legislative scheme.15 What is the content of the duty of procedural fairness in the context of administrative investigations and non-final decisions? An investigation is oftentimes the foundation of an ultimate administrative decision. Cracks in the foundation can be salvaged in subsequent steps in the administrative process but there are instances where those cracks will be fatal to the outcome and render a decision invalid. 13 Ibid at para 18. 14 Provincial Health Services Authority v British Columbia (Information and Privacy Commissioner), 2010 BCSC 931, 10 BCLR (5th) 175. 15 Hawrish v Cundall (1989), 76 Sask R 208 at para 28, 39 Admin LR 255, also cited in Swanson, supra note 10 at para 57. 2017] THE DUTY OF FAIRNESS IN THE INVESTIGATIVE STAGE 361 The hallmarks of a proper investigation are thoroughness and fairness. The circumstances of each case dictate the degree of thoroughness.16 A thorough investigation does not require that the investigator interview every person proposed by the complainant but the investigator must consider crucial evidence.17 In Tahmourpour v Canada (Solicitor General), the question before the court was whether the Human Rights Commission had been sufficiently thorough to be procedurally fair.18 The Federal Court of Appeal concluded that the Commission’s investigator had failed to consider and assess relevant data and that he had failed to interview relevant witnesses. Therefore, the investigation did not meet the required thoroughness prescribed by the case law. Thus, the investigator’s decision to dismiss the human rights complaint was set aside as being in breach of the duty of fairness. Another example of inadequacy in an investigation which does not meet the required fairness is the failure to advise the person affected of the case they have to meet.19 As previously stated, the content of the duty of fairness will depend on the context. In Swanson v Institute of Chartered Accountants, the duty of fairness at the investigative stage was described as a limited duty of fairness, not the broad duty of 2017 CanLIIDocs 175 fairness that is typically associated with rights to complete disclosure and to a full hearing.20 However, this limited duty of fairness does not lessen the duty for the affected party to be made aware of the case to meet and their right to be heard from the committee conducting the investigation.21 In Re Abel et al and Advisory Review Board, the Court concluded that a duty to act fairly generally arises where a tribunal's decision affects "the rights, interests, property, privileges or liberties of any person" or where an investigative board investigates and makes a report that may result in a person being subjected to “pains or penalties ... or in some such way adversely affected by the investigation and report”.22 The particular administrative board under consideration in that case played an advisory role regarding the adverse measures ultimately taken against a person's rights or interests. The question was whether and to what extent this role attracted a duty to act fairly. The Court found that in the reality of the operation of that board, the recommendations of the advisory board were virtually almost always 16 Slattery v Canada (Human Rights Commission), [1994] 2 FCR 574 at para 55, aff’d (1996), 205 NR 383. 17 Ibid at para 69. 18 Tahmourpour v Canada (Solicitor General), 2005 FCA 113, [2005] FCJ No 543. 19 Mooney v Canadian Society for Immigration Consultants, 2011 FC 496 at para 164, 9 Imm LR (3d) 198. 20 Swanson, supra note 10 at para 65. 21 Ibid at paras 75 and 78. 22 Re Abel et al and Advisory Review Board (1980), 119 DLR (3d) 101, 31 OR (2d) 520 (Ont CA). 362 UNBLJ RD UN-B [VOL/TOME 68 accepted. Therefore, their impact was great and attracted a duty of fairness. The proximity of the investigative or recommendation stage to the final decision is relevant; the closer the proximity to the final decision, the greater the content of the duty of fairness ought to be. It is well established that where the potential consequences of an administrative decision affects a person’s livelihood, a high standard of procedural fairness exists.23 Particularly, but not exclusively, in those situations, procedural unfairness in the preliminary stages may invalidate the decision. Where procedural errors occur at the preliminary stages of an administrative process, the characteristics of the remaining stages will determine whether there is a reasonable opportunity to cure the errors which were made initially. As the majority of the Court found in Khan v Ottawa (University of): Curing errors made at first instance depends on the seriousness of the initial error, the procedures followed by the appellate body, the powers of the appellate body, the way these powers were exercised and the weight the appellate body attributes to the initial decision. The closer the appeal is 2017 CanLIIDocs 175 to a complete reconsideration, with fair procedures, by a body that does not attribute significance to the initial decision, the more likely the defects will be cured.24 While all administrative decisions will require that the individual be afforded an opportunity to be heard, it does not follow that a full oral hearing would be necessary to meet the duty of fairness. However, where there are issues of credibility to be determined by an administrative body, fundamental justice requires that credibility be determined on the basis of an oral hearing.25 Even in cases where there are no credibility issues, the absence of an oral hearing in the later stages of the administrative proceeding may impact on the extent of the duty of procedural fairness in the investigative phase. Where there is no subsequent oral hearing or opportunity to address errors made in the investigative stage, there are no safeguards in place to avoid substantive errors and omissions in the decision, which stem from an investigation. In effect, this would remove the affected party’s fundamental right to be heard and to respond. Take, for instance, the New Brunswick Court of Appeal decision in Province of New Brunswick v Comeau. In that case, the Court of Appeal was critical of the procedural framework followed by the Department of Social Development conducting investigations pursuant to that province’s Family Services Act. The Department’s investigative protocols did not allow the persons affected by the investigation to receive disclosure of the complete investigation conclusion report, 23 Kane v Bd of Govenors of UBC, [1980] 1 SCR 1105, 18 BCLR 124; Henderson v College of Physicians and Surgeons of Ontario (2003), 65 OR (3d) 146, 228 DLR (4th) 598. 24 Khan v Ottawa (University of) (1997), 34 OR (3d) 535, 148 DLR (4th) 577 (Ont CA). 25 Ibid; Singh v Canada (Minister of Employment and Immigration), [1985] 1 SCR 177 at 213–214, 17 DLR (4th) 422. 2017] THE DUTY OF FAIRNESS IN THE INVESTIGATIVE STAGE 363 along with witness statements, or to be heard before the decision-maker prior to a decision affecting their rights being made. The Court of Appeal, per Bell J.A. concluded: Given that the stakes for future employment and access to programs and services offered by the Minister are so high, the Comeaus are entitled to be provided with a copy of all the evidence and an opportunity to respond fully. That response would obviously be made to the ultimate decision- maker[.]26 The Court of Appeal therefore concluded that the affected party was entitled to fulsome disclosure and an opportunity to present their response to the decision- maker. The process was contrary to the principle of audi alteram partem. Without these safeguards in place, there was no way to cure any substantive or procedural errors that may have occurred in the investigative stage. Generally, where a person or a panel exercise both investigative and 27 adjudicative functions, it will result in a reasonable apprehension of bias. In 2017 CanLIIDocs 175 Aylward v McMaster University, the court concluded that members of the deciding board or committee should not play the role of both prosecutor and decision-maker.28 The division of the investigative process and the administrative process is associated with a higher degree of procedural fairness. However, in principle, an affected person is always entitled to have an impartial and independent decision-maker, regardless of the content of duty owed. An inordinate delay in an investigation may potentially also amount to procedural failure. The administrative body may breach its own time limit, which in turn may be a sufficient ground for a court to set aside or stay the proceedings of an administrative decision. Where no time limits are set by the administrative body’s procedural framework, undue delay in administrative proceedings may constitute a breach of natural justice and can warrant a stay. This was recognized by the Supreme Court of Canada in Blencoe v British Columbia (Human Rights Commission).29 In that case, the Supreme Court identified two types of prejudice that may flow from delay and may result in a stay of proceedings. First, there is a delay that compromises a fair hearing. Second, there is a delay, which amounts to an abuse of process in that it brings the administrative process into disrepute. Under the second type of prejudice, it has been recognized that delay, in itself, may be sufficient to warrant a stay of proceedings. The Supreme Court found that to amount to an abuse of process, the delay must be unacceptable. 26 Province of New Brunswick v Comeau, 2013 NBCA 41 at paras 30–32, 406 NBR (2d) 269. 27 Gardner v Ontario Civilian Commission on Police Services, (2005) 72 OR (3d) 285, 18 Admin LR (4th) 191; Québec Inc v Québec (Régie des permis d'alcool), [1996] 3 SCR 919, 140 DLR (4th) 577. 28 Aylward v McMaster University (1991), 79 DLR (4th) 119 at para 31, 47 Admin LR 198 (Ont SCDC). 29 Blencoe v British Columbia (Human Rights Commission), 2000 SCC 44, [2000] 2 SCR 307. 364 UNBLJ RD UN-B [VOL/TOME 68 Conclusion Aside from fulfilling the duty to exercise their delegated powers in accordance with the law, why is fairness in the investigative process a concern for administrative bodies? For one, administrative bodies face a certain scope of liability for failing to meet the required standards, including regulatory negligence. In 2007, the Supreme Court of Canada recognized the tort of negligent investigation in Hill v Hamilton.30 Participants in all stages of administrative functions may face liability. The cases illustrate that administrative decision-making may be the result of a variety of various processes. Those processes will dictate whether it’s possible to cure flaws in the investigate stage with subsequent steps. The more a subsequent stage incorporates a procedure with the broad content of procedural fairness resembling a trial de novo, the more likely a court will find the procedural errors in the investigation may be cured. However, as the cases demonstrate, some irregularities in investigations will not be cured, where the proximity to the decision doesn’t allow for a meaningful way for the affected party to be heard before the decision-maker. To avoid injustice and a finding of procedural unfairness, it would 2017 CanLIIDocs 175 be good practice for administrative bodies to ensure that a process is well handled from the outset by having well-trained persons in charge of the initial investigations. Further, administrative bodies should review and revise their own procedural framework, where possible, to ensure they do not systematically deprive an affected party of their rights to procedural fairness. 30 Hill v Hamilton Wentworth Regional Police Services, 2007 SCC 41, [2007] 3 SCR 129. MARITIME LAW: SOVEREIGNTY IN THE ARCTIC KYLE MERCER* I. The Changing Arctic The Arctic region is changing literally and figuratively. In the literal sense, natural temperature variability and rising temperatures linked to global warming continue to cause the Arctic ice pack and Arctic sea ice to decrease at a concerning rate.1 Since 1979, NASA satellites have monitored and produced a constant record of the Arctic ice levels, substantiating this decrease.2 The period between 2007 and 2016 heightened global concern as the record indicates that the average annual decrease has become much more dramatic than that of decreases from previous years. The Arctic ice pack and Arctic sea ice has decreased by 40% since the satellite 2017 CanLIIDocs 175 monitoring began in 1979. This period has seen the lowest levels of the Arctic ice pack and Arctic sea ice levels, unmatched in recent human history.3 A release by the National Snow & Ice Data Center has shown that the Arctic sea ice extent in 2017 has been the lowest since the satellite records began in 1979.4 Credible projections indicate that this trend will continue, and that it is likely to become more drastic. It is becoming widely known and increasingly noticeable that the once ice-entombed Arctic is melting. In the figurative sense, the Arctic will become a significant economic and geopolitical region of the world. The decrease of the Arctic ice pack and the Arctic sea ice will reveal untouched land and seabed, rich in oil and minerals, and expose previously unnavigable waters and shipping lanes during the summer months. The * J.D., University of New Brunswick Law 2017, hailing from St. John’s, NL. I acknowledge and thank Gabrielle Chowne, J.D., for her comments on an earlier draft, and the editoral assistance of Ashley Godfrey, J.D. I would also like to thank my parents, Ric and Cindy, and Captain Charles Anonsen for their continued support and inspiration. 1 United States, National Snow and Ice Data Center, “Another Record Low for Arctic Sea Ice Maximum Winter Extent”, (Colorado, National Snow and Ice Data Center, 24 March 2016), online: 2 United States, National Aeronautics and Space Administration Earth Observatory, “Sea Ice”, (Colorado, NASA Earth Observatory (16 September 2016), online: 3 Council on Foreign Relations, “The Emerging Arctic” (2015), Council on Foreign Relations (blog), online: 4 United States, National Snow and Ice Data Centre, “2017 Ushers in Record Low Extent”, (Colorado, National Snow and Ice Data Center, 7 February 2017), online: 366 UNBLJ RD UN-B [VOL/TOME 68 newly navigable waters will be vital to the ability of states to control and economically benefit from the Arctic region. The 1982 United Nations Convention on the Law of the Sea (1982 UNCLOS) defines the rights and responsibilities of states with respect to their use of the world’s oceans.5 Yet, this convention is principally unable to resolve which states have territorial claims to some of the navigable waters in the Arctic region. Uncertainties surrounding territorial claims remain a preliminary question in the application of the 1982 UNCLOS in the Arctic region. However, the historical inaccessibility of the Arctic waters and the lack of activity in the region created disinterest in adjudicating territorial claims at the International Tribunal for the Law of the Sea (ITLOS) and International Court of Justice (ICJ). To date, differing opinions regarding claims to these waters have formed the basis of the conversation between the Arctic states.6 However, the recent7 turning of the capitalistic tide in the United States of America (U.S.) and the ambition of the Government of China will undoubtedly turn what has been a moderate disagreement into a more intense dispute, as states seek to exercise their rights to these waters. 2017 CanLIIDocs 175 II. Claim to the Northwest Passage: An Exercise of Sovereignty As an Arctic state and founding member of the Arctic Council,8 Canada will be involved in the global dispute regarding territorial claims and the rights and responsibilities of states in the Arctic region.9 Canada is currently a principal party of 5 United Nations Convention on the Law of the Sea, 10 December 1982 at Part III (entered in force 16 November 1994) [UNCLOS]. 6 Ted L McDorman, Salt Water Neighbours: International Ocean Law Relations between the United States and Canada, (New York: Oxford University Press, 2009) at 225-230. The reference to Arctic States includes all members of the Arctic Council: Canada, the Kingdom of Denmark (including Greenland/the Faroe Islands), Finland, Iceland, Norway, Russian Federation, Sweden, the United States of America and six international organizations representing Arctic Indigenous Peoples. 7 On April 28, 2017, an executive order titled “America-First Offshore Energy Strategy” was released by the President of the United States, Donald Trump, to allow oil exploration and drilling in the Arctic region. To carry out the policy of “encourag[ing] energy exploration and production, including on the Outer Continental Shelf” this executive order offers in Section 11: Review of Offshore Arctic Drilling Rule that “The Secretary of the Interior shall immediately take all steps necessary to review the Final Rule entitled "Oil and Gas and Sulfur Operations on the Outer Continental Shelf—Requirements for Exploratory Drilling on the Arctic Outer Continental Shelf," 81 Fed. Reg. 46478 (15 July 2016), and, if appropriate, shall, as soon as practicable and consistent with law, publish for notice and comment a proposed rule suspending, revising, or rescinding this rule.” Presidential Executive Order Implementing an America-First Offshore Energy Strategy (USA White House Executive Order), pursuant under Outer Continental Shelf Lands Act, 43 USC 1331 (28 April 2017), online: 8 The Arctic Council is the leading intergovernmental forum promoting cooperation, coordination and interaction among the Arctic States, Arctic indigenous communities and other Arctic inhabitants on common Arctic issues, in particular on issues of sustainable development and environmental protection in the Arctic. Arctic Council, “The Arctic Council: A backgrounder” (2015), online: 9 Ibid. 2017] SOVEREIGNTY IN THE ARCTIC 367 a long-standing but dormant dispute regarding territorial claims; more specifically, a dispute regarding the international legal status of the Northwest Passage (NWP). The NWP is an ocean route comprised of a series of straits/sea lanes through the Canadian Arctic Archipelago. This Archipelago was assigned to Canada from Britain in 1880.10 Indeed, it is not the route nor the archipelagic land that is the subject of the international dispute, but the NWP’s international legal status under the 1982 UNCLOS.11 Canada claims that the waters within the NWP have the international legal status of Canadian internal waters. The NWP dispute is critical to Canada’s ability to assert their sovereignty in the Arctic. Recognition of this status would accord Canada the authority to deny all foreign vessels, both commercial and military, the use of the NWP.12 Many states, predominantly the U.S., do not recognize Canada’s assertion that the NWP is Canadian internal waters, as they seek to have access to, and use of, the NWP, with as much freedom or rights during vessel passage as legally attainable. Thus, other states, namely the U.S., consider the waters to be an International Strait.13 Ultimately, the NWP will not have a definitive legal status unless this matter is either settled between the principal states, or is adjudicated by the ITLOS or the ICJ. 2017 CanLIIDocs 175 It is the value of the transit of the NWP for commercial and military vessels that propels the dispute. It is estimated that the Arctic contains 13% (≈90 billion barrels) of the world’s undiscovered conventional oil resources and 30% of the undiscovered natural gas resources.14 One-third of the undiscovered oil lies within the Arctic territory of the U.S. Through the implementation of the recent executive order titled “America-First Offshore Energy Strategy,” it is foreseeable that these resources will be realized in the near future.15 Extraction and transport of these resources will increase the need and value of transit through the NWP. The NWP offers a route through the Arctic Sea, between the Pacific and Atlantic Oceans. There exist five major routes that can be used depending on a number of factors: the size of the vessel, the vessels ice-breaking capability, the ice conditions, and the adequacy of current navigational information.16 Three of the five routes have been identified as 10 Michael Byers, Who Owns the Arctic? Understanding Sovereignty Disputes in the North. (Madeira Park, BC: Douglas & McIntyre, 2009) at 6. 11 Alan Vaughan Lowe & Robin Rolf Churchill, The Law of the Sea (Manchester: Manchester University Press, 1999) at 102. 12 Ibid at 61. 13 McDorman, supra note 6 at 225. 14 United States Energy Information Administration, “Arctic Oil and Natural Gas Resources”, (Washington DC, USEIA, 20 January 2012), online: 15 Christina Nunez, “What Happens When Oil Spills in the Arctic?”, National Geographic (24 April 2014), online: 16 McDorman, supra note 6 at 229-230. 368 UNBLJ RD UN-B [VOL/TOME 68 having the greatest commercial viability and are predicted to be ice-free from July until September by the end of this century – at the latest.17 If the predictions about the usability of the NWP are correct, ships will be attracted to the beneficial usage of Canadian Arctic waters. Use of the NWP for transit from Eastern Asia and Europe is approximately 7,000 km shorter than the current route through the Panama Canal – this will save time and reduce the cost of transit. The Danish-operated Nordic Orion became the first bulk carrier to traverse the NWP in September 2013, saving approximately $80,000 in fuel.18 This transit was registered with, and received permission from the Canadian Coast Guard. Thereby, it did not undermine Canada’s legal position regarding the NWP.19 It is uncertain whether other states and their vessels will respect Canadian authority in this manner or seek to transit the NWP without permission. However, following the successful transit of the Nordic Orion, it is certain that other states and their vessels are planning to use the NWP. In 2016, the Government of China publicly announced their ambition and encouragement of the use of the NWP for commercial transit. This announcement was accompanied by the publication of the Arctic Navigation Guide (Northwest Passage). This Guide was published without consultation with Canada and offers extensive information 20 regarding the transit of the NWP. Further, the NWP can also accommodate super- 2017 CanLIIDocs 175 tankers and container ships that are too large for the Panama Canal. Naval aircraft carriers are also too large for the Panama Canal and may be attracted to an ice-free NWP.21 Indeed, the importance of the NWP is evident from the many beneficial usages of the passage. The international strait status, asserted by the U.S., provides the highest degree of freedom for vessel passage, allowing states to take advantage of the economic and geographic/transportation benefits listed above.22 Canada reserves the potential to gain substantial economic benefits from the use of the NWP. However, Canada must comprehensively consider a balance of the benefits and the risks associated with the use of the NWP before transit activity increases. As the NWP is a route through the Canadian Arctic Archipelago, it is Canada’s duty to first consider the detrimental effects of the use of the NWP on the Canadian Arctic environment and the Arctic peoples. There are many Inuit communities along the NWP that should not be discounted by the size of their population. Communities such as Qausuittuq (Resolute), Uqsuqtuuq (Gjoa Haven), Iqaluktuuttiaq (Cambridge Bay), Qurluktuk (Kugluktuk), and Ulukhaqtuuq 17 NASA Earth Observatory, supra note 2. 18 Whitney Lackenbauer & Adam Lajeunesse, “More Ships in the Northwest Passage will boost our Arctic claim”, The Globe and Mail (5 January 2015), online: 19 Michael Byers, “Canada’s Arctic Nightmare Just Came True: The Northwest Passage is Commercial”, Globe and Mail (20 September 2013), online: 20 Nathan VanderKlippe, “China Reveals Plans to Ship Cargo across Canada’s Northwest Passage”, Globe and Mail (20 April 2016), online: 21 Byers, supra note 10 at 40. 22 Lowe & Churchill, supra note 11 at 105. 2017] SOVEREIGNTY IN THE ARCTIC 369 (Ulukhaktok) are located on the three routes identified as having the greatest commercial viability. It is important to note that these communities may benefit from a reduction in the cost of supplies and unemployment if activity in the region increases. However, these communities are likely to experience the most severe negative impacts. In particular, oil extraction can significantly benefit the Canadian economy. However, oil spills, caused by drilling, extraction and transportation, have more severe effects in the Arctic region when compared to spills in more temperate climates. In high-latitude, cold ocean environments, oil persists for longer periods.23 Low temperatures and insufficient sunlight result in low rates of natural oil evaporation and decomposition. This is substantiated by the effects of the 1989 Valdez oil spill that occurred in the sub-Arctic region. The Exxon Valdez spilled approximately 42 million litres of crude oil, resulting in contamination of 1,990 kilometers of shoreline. It has been estimated that approximately 302 harbor seals, 2,000 sea otters, and 250,000 seabirds died in the days following the spill.24 Fish populations and larger marine mammals will suffer long-term ill effects from this spill and it will take up to 30 years for shoreline habitats such as mussel beds to fully 2017 CanLIIDocs 175 recover.25 More than a decade after this event, researchers report that a significant amount of oil persists in this region.26 If the oil becomes trapped in ice, the long-term impacts of Arctic oil spills can be more devastating than previously thought.27 The National Academy of Sciences have concluded that “no current cleanup methods remove more than a small fraction of oil spilled in marine waters, especially in the presence of broken ice.”28 The once ice-entombed Arctic is vulnerable to deliberate and residual pollution caused by increases in transport, industrial development and mineral, oil and natural gas extraction.29 Indeed, the environmental risks associated with the use of the NWP are significant. 23 Mark Nuttal, “The Arctic is Changing” (2000) Stefansson Arctic Institute/European Union Raphael Programme, online: 24 Sarah Graham, “Environmental Effects of Exxon Valdez Spill Still Being Felt”, Scientific American (19 December 2003) online: 25 Ibid. 26 Adam Hadhazy, “20 Years After the Exxon Valdez: Preventing and Preparing for the Next Oil Spill Disaster”, Scientific American (23 March 2009), online: 27 Earth Observatory, NASA, supra note 2. 28 Joel K Bourne Jr, “As Arctic Melts, a Race to Test Oil Spill Cleanup Technology”, National Geographic (14 September 2013), online: 29 World Wildlife Foundation Canada, “Balancing Shipping Opportunities with a Healthy Arctic Future” (2017), online: 370 UNBLJ RD UN-B [VOL/TOME 68 It has become apparent that the states who assert legal access to, and use of, the NWP are not fully considering the vulnerability of the Arctic environment and the Arctic peoples. Thus, the international legal status should not be left to uncertainty. The Valdez spill sheds light on the devastating effects of oil on the Arctic environment. A spill in the NWP would have devastating consequences on a population that for the past 4000 years, has relied on the hunting of seal and caribou, and fishing for Arctic char. 30 Should Canada win this dispute, they will have the authority to require foreign vessels to obtain their permission to transit the NWP. With exclusive access and control over the NWP, Canada will have the opportunity and the responsibility to instate stringent environmental regulations as well as the authority to foster strong international cooperation in the protection of the Arctic environment and the Arctic peoples. III. Focus of Report: International Legal Statuses & Associated Rights of Passage The resulting legal status of the NWP will have implications on Canada’s authority 2017 CanLIIDocs 175 to regulate, manage, and control the passage. This is a concern to other states, who seek to minimize the amount of control Canada is authorized to exercise over the passage. Simply, three main international legal statuses exist under the 1982 UNCLOS: 1) internal waters – which does not allow a foreign vessel the right of passage; 2) international strait – which allows a foreign vessel the right of ‘transit passage’; 3) territorial sea – which allows foreign vessels the right of ‘Innocent Passage’.31 These rights will be explained further in this report as their implication on foreign vessel passage greatly differs. The significance of the dispute is increasing as the Arctic ice pack and Arctic sea ice melt. If Canada is to assert their sovereignty in the Arctic, this dispute must be settled by the principal states such that the international legal status of the NWP is determined and internationally recognized. If the international legal status dispute is resolved in Canada’s favour, Canada will be better able to economically benefit from the Arctic, and protect the Arctic environment and Inuit peoples from threat. As Canada is a party to UNCLOS, this report will further explore the three international legal statuses offered under the convention, and the rights of passage that each grant to foreign, commercial and military vessels. Then, the uncertainty over the international legal status of the NWP will be approached through the lens of the Canada-U.S. disagreement. Assessing the cogency of their claims will shed light on the likelihood of the possible outcomes of an international legal status for the NWP. 30 Canadian Museum of History, “The Dorset People” (2017) Government of Canada (Blog), online: 31 McDorman, supra note 6 at 212. 2017] SOVEREIGNTY IN THE ARCTIC 371 IV. Introduction of Analysis The technical definition of a ‘strait’ is a narrow, natural passage or arm of water connecting two larger bodies of water.32 While the NWP fits this physical definition, it is largely irrelevant at International Law. The physical properties of the passage do not lead to a clear determination of its legal status. The absence of a clear definition offers the starting point for an analysis as to the international legal statuses that may apply to the NWP. If brought before the ITLOS or ICJ for determination of this question, three international legal statuses could be mandated to the waters of the NWP: internal waters, international strait, or territorial sea. As previously mentioned, the legal status that is ultimately ascribed to the NWP will determine Canada’s control over its regulation and other countries’ ability to navigate it. What follows is an outline of the three international legal statuses that could be given to the NWP. Indeed, it is this status and not its geographical features that determine the rights sought after by the U.S. and other states. V. Internal Waters 2017 CanLIIDocs 175 Internal waters are not part of a state’s maritime territory or territorial sea; they are legally assimilated to the terrestrial territory. A state enjoys full and exclusive sovereign authority over their own internal waters and, as mentioned above, has the right to deny all foreign vessels, both commercial and military, from passage through these waters.33 Without the permission to enter another state’s internal waters, the act of entering is a breach of law. When a vessel enters another state’s internal waters, that vessel becomes subject to all domestic laws of that state. The 1982 UNCLOS does not offer a right of passage to foreign vessels nor does it provide for the regulation of conduct within internal waters.34 This omission is likely due to the principle that the state has the sovereign authority to deal with its internal waters as it so chooses. The 1985 NWP voyage of the US Coast Guard Ice-Breaker, the Polar Sea, was the catalyst for the Canadian Government’s public assertion that the NWP was internal Canadian waters. Although, it may not have been the intent of the U.S., the transit caused a nationwide public outcry as Canadian nationalists saw this as a serious violation of Canada’s sovereignty in the Arctic region. In response, Canada drew straight baselines around the outmost islands of the Arctic Archipelago later that year.35 Arguably, these straight baselines were drawn in accordance with the 1982 UNCLOS regulations.36 Article 8 of Part II of the 1982 UNCLOS states that 32 Lowe & Churchill, supra note 11 at 102. 33 Ibid at 61. 34 Ibid. 35 Byers, supra note 10 at 51-52. 36 McDorman, supra note 6 at 238-340. 372 UNBLJ RD UN-B [VOL/TOME 68 internal waters are: “waters on the landward side of the baseline of the territorial sea form part of the internal waters of the State.” Based on Article 8, these straight baselines would prima facie give the NWP the status of internal waters. The U.S. and other countries do not agree with the Canadian drawn straight baselines, as highlighted by statements such as: “Canada was not justified in stating that all waters between Canadian islands in the Arctic were internal Canadian waters.” 37 Further, from an international law standpoint, they claim there is no basis to support justification of the straight baselines. However, Canada supports their claim that the NWP is within Canadian internal waters regardless of the legitimacy of the refuted straight baselines. This is based on an international legal distinction that is reliant on historic usage: “If the waters are internal by reason of historic title, no right of passage applies and the strait would not be capable of becoming an international strait.”38 This limitation set out by the 1982 UNCLOS offers Canada an expanded claim to strengthen the justification of the NWP’s international legal status. In 1962, the United Nations Secretariat Study determined that historic claims were necessary 2017 CanLIIDocs 175 to maintain a state’s title to areas that otherwise escape the scope of codified international law of the sea.39 Therefore, claims of historic internal waters are based on customary international law. This was determined by the ICJ in the 1982 Libya- Tunisia Case concerning the delimitation of the continental shelf between the adjacent states.40 Canada’s claim is further strengthened by ICJ precedence from the 1951 Anglo-Norwegian Fisheries Case. This case was brought before the ICJ to determine whether historical usage of a territory would act to include such territory within the exclusive bounds, offering Norwegian vessels exclusive rights to use such waters. The judgement is monumental as it provides that, although waters may not have the characteristics of internal waters at international law, they will be treated as such if there is evidence of historical title.41 This reasoning is a grand representation of a component of Canada’s claim regarding the international legal status of the NWP. As outlined in the 1962 United Nations Secretariat Study, there are three criteria for assessing whether the NWP is, in fact, historic waters: 1) the authority exercised over the area by the state claiming it as ‘historic water’; 2) the continuity of such exercise of authority of which occurs over a considerable amount of time; and 3) the attitude of foreign states.42 The 37 Rob Huebert, “Polar Vision or Tunnel Vision: The Making of Canadian Arctic Waters Policy”, (1995) 19:4 Marine Policy 343. 38 Donat Pharand, Canada’s Arctic Waters in International Law (Cambridge: Cambridge University Press, 1988) at 223. 39 McDorman, supra note 6 at 214-215. 40 Case Concerning the Continental Shelf (Tunisia/Libyan Arab Jamahiriya), [1982] ICJ Rep 18 at para 100. 41 Fisheries Case (United Kingdom v Norway), [1951] ICJ Rep 116 at 131. 42 McDorman, supra note 6 at 216. 2017] SOVEREIGNTY IN THE ARCTIC 373 burden of proof to satisfy these criteria is very high and the onus would be placed on Canada if the dispute was brought before ITLOS or the ICJ. With application of Canada’s claim to the above three-part test, the validity of the historic internal waters claim will be analyzed further in this report.43 VI. International Strait The second international legal status is the international strait. This status was created during the expansion of the territorial sea from three to twelve nautical miles in the 1982 UNCLOS. This resulted in many narrow waterways, once known as ‘High Seas,’ falling within the territorial sea of the adjacent states. This was a concern for many states, as areas of water were now ‘captured’ in the territory of certain states, thus threatening to impede navigation and detrimentally affect strategic and economic interests.44 The international strait status was the result of a need for a navigational-friendly right of passage through these narrow waterways. Canada’s support for the resolution to create this international legal status was conditional on its inapplicability to the NWP. The status was first codified in Article 2017 CanLIIDocs 175 16(4) of the 1958 Territorial Sea and Contiguous Zone Convention, granting the right of passage through international straits if such passage was for international navigation. This status carried with it the Non-Suspendable right of ‘Innocent Passage.’ This right of passage conferred states with a higher degree of unimpeded navigation and introduced a right that could not be suspended. However, the remaining restriction on the freedom of navigation, within the Non-Suspendable right of ‘Innocent Passage’ was a catalyst for the codification of a new right: the right of ‘transit passage’. Article 37 of the 1982 UNCLOS transit passage reads: “This section applies to straits which are used for international navigation between one part of the high seas or an exclusive economic zone and another part of the high seas or an exclusive economic zone.”45 The added emphasis identifies the distinction between a Non-Suspendable Right of ‘Innocent Passage’ and ‘transit passage’. A Non-Suspendable Right of ‘Innocent Passage’ exists for international navigation only when the route connects one part of the high seas, or an exclusive economic zone and the territorial sea of a foreign state. It was previously mentioned that the absence of a definition of what constitutes a ‘strait’ at international law is problematic to the NWP dispute. This is also true for the absence of those defining factors that establish a strait as an “international strait” or ‘Strait used for International Navigation’ – terms found in the 1958 Territorial Sea and Contiguous Zone Convention and the 1982 UNCLOS. Thus, International law relies on the 1949 ICJ decision in the Corfu Channel Case to 43 Pharand, supra note 38 at 223. 44 McDorman, supra note 6 at 212. 45 UNCLOS, supra note 5 (emphasis added). 374 UNBLJ RD UN-B [VOL/TOME 68 address these definitional gaps. The Corfu Channel Case ultimately outlined two heads of criteria that exist when determining whether a ‘Strait’ is an international strait: 1) Geographic and 2) Functional.46 Although these criteria shed light on what constitutes an international strait, they are not without their problems. The criteria are broad and lack official specification. Thus, if ITLOS or the ICJ were to adjudicate the legal status of the NWP, reliance on the criteria to assist on this determination would likely prove unhelpful. The geographical criterion, as noted above, is outlined in Part III, Section 2 Article 37 of the 1982 UNCLOS: “This section applies to straits which are used for international navigation between one part of the high seas or an exclusive economic zone and another part of the high seas or an exclusive economic zone.”47 In the event that Canada is not able to justify that the NWP are Canadian historic internal waters, it is certain that the NWP satisfies the above geographic criterion. The Functional Criterion, as outlined in the Corfu findings, is not dependent on the volume of vessel traffic through a Strait.48 There was no “use” threshold bearing on whether a ‘Strait’ is an international strait. This would discredit a Canadian argument that the NWP, due to lack of use for international navigation, is 2017 CanLIIDocs 175 not an “international strait.” However, after the Corfu Channel Case, more emphasis has been placed on the historical acknowledgment of a route as a “useful” route for international navigation.49 The right of passage through an international strait, known as a transit passage, is unique at international law. Article 38 of the 1982 UNCLOS outlines the right of ‘transit passage’ which reads: 1. In straits referred to in article 37, all ships and aircraft enjoy the right of transit passage, which shall not be impeded; except that, if the strait is formed by an island of a State bordering the strait and its mainland, transit passage shall not apply if there exists seaward of the island a route through the high seas or through an exclusive economic zone of similar convenience with respect to navigational and hydrographical characteristics. 2. Transit passage means the exercise in accordance with this Part of the freedom of navigation and overflight solely for the purpose of continuous and expeditious transit of the strait between one part of the high seas or an exclusive economic zone and another part of the high seas or an exclusive economic zone. However, the requirement of continuous and expeditious transit does not preclude passage through the strait for the purpose of entering, leaving or returning from a State bordering the strait, subject to the conditions of entry to that State. 46 The Corfu Channel Case, Judgment of April 9th, 1949: [1949] ICJ Rep 28-29 at para 4 [Corfu Channel Case] 47 UNCLOS, supra note 5 at part III (emphasis added). 48 Corfu Channel Case, supra note 46 at 28. 49 McDorman, supra note 6 at 243. 2017] SOVEREIGNTY IN THE ARCTIC 375 3. Any activity which is not an exercise of the right of transit passage through a strait remains subject to the other applicable provisions of this Convention.50 ‘Transit passage’ does not offer the same freedom of navigation that exists on the high seas. However, when read together, the emphasis in article 38(1) and 38(2) above grants vessels, both commercial and military, the highest degree of freedom of unimpeded navigation that exists under the 1982 UNCLOS. This status is not subject to a threshold of ‘Innocence’ that must be satisfied during transit.51 Traditionally, within the territorial sea of a state, passage of a foreign vessel is required to comply with Article 19 of the 1982 UNCLOS: 1. Passage is innocent so long as it is not prejudicial to the peace, good order or security of the coastal State. However, foreign vessels are required to refrain from making threats or use of force. This is not a condition of the right of transit passage but an ancillary obligation. An activity which threatens a state has the effect of classifying the vessel under the 2017 CanLIIDocs 175 general regime of innocent passage, in which case passage can be denied for want of innocence. The right of ‘transit passage’ does have limitations as it only permits vessels to conduct activities incidental to the normal mode of “continuous and expeditious transit” unless violations are rendered necessary by force majeure or distress. 52 It is this right and the freedom that it grants to vessels, both commercial and military, that the U.S. and other states seek to obtain when transiting the NWP. The motivations of foreign states are clear in relation to their assertion that the NWP has the international legal status of an international strait. Canada, on the other hand, considers the international legal status of an international strait to be a serious concern. If the NWP is considered an international strait, Canada will not be able to deny a vessel’s passage unless the transit violates other provisions of the 1982 UNCLOS. VII. Territorial Sea The third international legal status is that of a territorial sea. This status is not a subject of the NWP dispute. Each island within the archipelago is entitled to its own baseline, drawn accordingly to the normal principles.53 Within Canadian territorial sea, foreign vessels, both commercial and military, enjoy the right of ‘Innocent Passage’. Under Article 18 of the 1982 UNCLOS – Meaning of Passage states: 50 UNCLOS, supra note 5 at part III (emphasis added). 51 Lowe & Churchill, supra note 11 at 107. 52 Ibid at 110. 53 Ibid at 123 (Archipelagic Baslines). 376 UNBLJ RD UN-B [VOL/TOME 68 Passage means navigation through the territorial sea for the purpose of: traversing that sea without entering internal waters or calling at a roadstead or port facility outside internal waters; or proceeding to or from internal waters or a call at such roadstead or port facility. Passage shall be continuous and expeditious. However, passage includes stopping and anchoring, but only in so far as the same are incidental to ordinary navigation or are rendered necessary by force majeure or distress or for the purpose of rendering assistance to persons, ships or aircraft in danger or distress.54 Article 19 of 1982 UNCLOS states ‘Innocent Passage’ is innocent so long as it is not prejudicial to the peace, good order or security of the coastal state. If ‘innocent’, foreign commercial and military vessels would have more rights during passage within the territorial sea. The emphasis in Article 18 denotes “stopping and anchoring” but only so far as the same are incidental to ordinary navigation. The reason that Canada has concern about the right of ‘Innocent Passage’ can be found within Article 24 of the 1982 UNCLOS. This Article states: 2017 CanLIIDocs 175 1. The coastal State shall not hamper the innocent passage of foreign ships through the territorial sea except in accordance with this Convention. In particular, in the application of this Convention or of any laws or regulations adopted in conformity with this Convention, the coastal State shall not: (a) impose requirements on foreign ships which have the practical effect of denying or impairing the right of innocent passage; or (b) discriminate in form or in fact against the ships of any State or against ships carrying cargoes to, from or on behalf of any State. 2. The coastal State shall give appropriate publicity to any danger to navigation, of which it has knowledge, within its territorial sea.55 If the NWP is considered Canadian Territorial Waters, Canada would have jurisdiction over the waters but would not be permitted to deny, impair or discriminate against the ‘Innocent Passage’ of foreign vessels, whether military or commercial. However, “Innocent Passage” offers Canada the ability, in the event that a foreign vessel violates Article 25 of the 1982 UNCLOS, to prevent passage of such vessel. Specifically, Article 25(1) and Article 25(3) – Rights of Protection of the Coastal State reads: 1. The coastal State may take the necessary steps in its territorial sea to prevent passage which is not innocent. … 3. The coastal State may, without discrimination in form or in fact among foreign ships, suspend temporarily in specified areas of its 54 UNCLOS, supra note 5 at Part II (emphasis added). 55 Ibid (emphasis added). 2017] SOVEREIGNTY IN THE ARCTIC 377 territorial sea the innocent passage of foreign ships if such suspension is essential for the protection of its security, including weapons exercises. Such suspension shall take effect only after having been duly published.56 While these provisions offer Canada some ability to prevent passage through the NWP, this is not the international legal status that Canada is asserting to exist within the waters of the NWP. In a time in which the Arctic is becoming a critical economic and geopolitical region of the world, the limited rights that the status of Territorial Waters affords is not sufficient. In Canada’s opinion, this status does not give enough authority to regulate, manage and control the NWP. Therefore, the status of territorial sea should be considered a compromise to the NWP dispute. It must briefly be noted that Article 234: Ice-covered areas of the 1982 UNCLOS does not grant Canada indirect jurisdiction over foreign vessel navigation through the NWP. The U.S. consider Article 234 as binding customary law but oppose the attempts of states to assert jurisdiction over navigation through this authority.57 Article 234 does grant Canada limited authority to enforce more 58 stringent environmental standards, like the Arctic Waters Pollution Prevention Act , 2017 CanLIIDocs 175 to reduce and control pollution caused by foreign vessels that navigate in Canadian “ice-covered” waters. However, it would be detrimental to Canada’s legal claim to invoke Article 234 regarding the NWP. Invoking Article 234 would concede the claim that the NWP is within Canadian internal waters as foreign vessels navigating the NWP should be subject to the jurisdiction of Canadian domestic law – not a multilateral treaty. If Canada continues to assert sovereignty in the NWP, the authority granted under Article 234 should be disregarded. VIII. The Canada-U.S. Disagreement As mentioned, Canada is the principal party to the long-standing but dormant dispute regarding the correct international legal status of the NWP. As the dispute is approaching its pinnacle of importance, the uncertainty is best approached through the lens of the opposing claims of Canada and the U.S. Canada asserts the NWP is Canadian internal waters whereas the U.S. asserts that the NWP is an international strait. Canada should consider this dispute to be a national issue – a test of sovereignty burdened with emotional freight as many Canadians consider the Arctic to be Canadian. The U.S. considers this dispute to be a global issue and a military concern, as it will hinder the ability of states to transit and economically benefit from 56 Ibid. 57 Kristin Bartenstein, “The “Arctic Exception” in the Law of the Sea Convention: A Contribution to Safer Navigation in the Northwest Passage?” (2011) 42 Ocean Dev & Intl Law 22 at 27, online: 58 Arctic Waters Pollution Prevention Act, RSC 1985, c A-12. 378 UNBLJ RD UN-B [VOL/TOME 68 the Arctic region.59 They are ultimately concerned about the level of authority Canada will have in the regulation, management and control of their own (foreign) vessels transiting the NWP. A verdict regarding the international legal status of the NWP will determine whether Canada has the sovereign authority to require foreign vessels, both commercial and military, to obtain their permission before passage through the NWP. This is an interesting dispute as both parties continually avoid legal resolution and actual political confrontation, often expressing a commitment to cooperation regarding the NWP.60 It is apparent that both parties are uncertain as to what the outcome of an ITLOS or ICJ adjudication would be. However, both have well-founded legal arguments regarding their assertion as to the international legal status of the NWP. IX. The Canadian Legal Claim The Canadian claim finds its roots in a response to the actions of the U.S. As previously mentioned, the 1985 NWP voyage of the U.S. Coast Guard Ice-Breaker, the Polar Sea, was the catalyst for the Canadian Government’s public assertion that 2017 CanLIIDocs 175 the NWP was within Canadian internal waters. This assertion is contested by other states, as the validity of the straight baselines drawn around the outmost islands of the Arctic Archipelago is not certain.61 The Canadian government will need to substantiate that the 1985 straight baselines drawn around the Arctic Archipelago are consistent with international legal criteria and guidelines. According to Donat Pharand’s research, these baselines meet the “flexible” geographic criteria and guidelines at international law. Donat Pharand is a well-recognized expert with regards to the NWP dispute. However, other scholars do not share this opinion.62 As mentioned, the U.S. and other states claim that from an international law standpoint, there is no basis to support justification of the straight baselines. The European Community, an organization focusing on economic integration between members of the European Union, claim that these straight baselines violate Article 7(3) of the 1982 UNCLOS which reads: 3. The drawing of straight baselines must not depart to any appreciable extent from the general direction of the coast, and the sea areas lying within the lines must be sufficiently closely linked to the land domain to be subject to the regime of internal waters.63 They advocate that the baselines around the Arctic Archipelago utilize “inappropriate base-points, are excessive in length and depart from the general direction of the coast in areas such as Lancaster Sound, Amundsen Gulf and 59 McDorman, supra note 6 at 225. 60 Ibid (The 1988 Canada-U.S. Arctic Cooperation Agreement). 61 Ibid at 238-340. 62 Ibid at 238; Pharand, supra note 38. 63 Byers, supra note 10 at 53 (emphasis added). 2017] SOVEREIGNTY IN THE ARCTIC 379 McClure Strait.”64 However, application of the ‘flexible criteria’ referred to by Donat Pharand, is favorable to the Canadian claim. Based on the ruling in the 1951 Anglo- Norwegian Fisheries Case, Canada can contend that the length of the baselines are justified. In this judgement, it states that consistent and sufficiently long “peaceful usage” may be considered in drawing baselines that legitimize the extension of baselines under customary international law.65 Through this reasoning, Inuit use and habitation of the Arctic and Arctic Waters can substantiate the extended Arctic straight baselines.66 It remains an unanswered question whether the Arctic straight baselines satisfy the “flexible” geographic criteria and guidelines at international law. If so, the Arctic straight baselines capture the waters of NWP as Canadian internal waters. However, the status of internal waters does not automatically accord Canada the authority that it should be asserting; which is, the right to deny all foreign vessels, both commercial and military, from passage through these waters. This is due to Article 8(2) – internal waters of the 1982 UNCLOS: 2. Where the establishment of a straight baseline in accordance 2017 CanLIIDocs 175 with the method set forth in article 7 has the effect of enclosing as internal waters areas which had not previously been considered as such, a right of innocent passage as provided in this Convention shall exist in those waters.67 However, there is a cogent argument that allows Canada to avoid this treaty obligation. When Canada drew the straight baselines around the Arctic Archipelago in 1985 it was not a party to the 1982 UNCLOS – it was a signatory. Canada ratified the 1982 UNCLOS in 2003. Thus, Canada would have been subject to customary international law as set out in the 1951 Anglo-Norwegian Fisheries Case, which provided that no right of innocent passage exists in waters enclosed by straight baselines.68 To assert that otherwise would be a clear violation of the fundamental principle of non-retroactivity of treaties outlined in Article 28 of the Vienna Convention on the Law of Treaties.69 Ultimately, if the straight baselines are considered to satisfy the “flexible” geographic criteria and guidelines at international 64 Ibid; McDorman, supra note 6 at 238. 65 Ruwanthika Gunaratne, “Anglo-Norwegian Fisheries Case” (2008), Public International Law (blog), online: 66 Byers, supra note 10 at 54. 67 UNCLOS, supra note 45 at part II (emphasis added). 68 Pharand, supra note 38 at 228. 69 Kristin Bartenstein, “Use it or lose it”: An appropriate and wise slogan?”, Policy Options (1 July 2010), online: 380 UNBLJ RD UN-B [VOL/TOME 68 law, then Canada would have a cogent argument that grants the authority they seek in the NWP. This claim is strongly contested and has been denounced by the U.S and other states. In the event that the claim fails, Canada will be required to rely on the claim that the NWP is within Canadian historic internal waters. This claim will be difficult to substantiate and will likely be met with similar assertions that the NWP is an international strait. Nonetheless, if Canada is successful it will result in the NWP gaining the internal waters status. In 2007, Donat Pharand concluded that Canada would not succeed in establishing that the NWP had the status of Canadian Historic internal waters.70 His conclusion was based on the difficulty faced by Canada in discharging the heavy burden of proof that it had exercised jurisdiction over the Arctic waters for a sufficiently long period of time, without protest from the U.S. and other states. The argument that Canada has exercised jurisdiction over the NWP is exceptionally complex. This is due, in part, to piecemeal Canadian assertions of such jurisdiction, from the implicit assertion made in 1907 by Canadian Senator Pascal Poirier that Canada owned everything within a pie-shaped sector extending from the continental coastline to the geographic North Pole to Prime Minister Stephen 2017 CanLIIDocs 175 Harper’s continual admonishment of states that do not recognize Canadian sovereignty over the NWP.71 Under international law, Canada can assert that they have exercised exclusive authority over the NWP; however, factually based demonstration of this authority is necessary. As the Arctic Archipelago was assigned to Canada from Britain in 1880, the Canadian claim relies partly on British exploration and mapping, dating back to the voyages of Sir Martin Frobisher in 1576. Canada also enacted its first legislation regarding the Arctic waters in 1926. 72 As argued by Donat Pharand, there exists a fatal flaw in Canada’s historic argument caused by the lack of alignment between historical activity in the Canadian Arctic waters and the assertion of Canadian sovereignty in such waters. Assertions of the NWP as Canadian internal waters have predominantly occurred in the 20th century and have always been met by opposition from states that do not recognize the claim. 73 If Canada is to succeed in the NWP dispute they will be required to partner with the Inuit peoples and base the claim on the Inuit’s historical occupation and use of the NWP. The Inuit peoples were assimilated into Canada as citizens under the Indian Act in 1924.74 Following the 1975 ICJ decision on the rights of Nomadic peoples in Western Sahara, the Canadian-Inuit connection in the claim for Canadian 70 Pharand, supra note 38 at 237-238. 71 Byers, supra note 10 at 43. 72 Ibid at 49. 73 Ibid at 50. 74 Canada, Indian and Northern Affairs Canada, Canada's Relationship with Inuit: A History of Policy and Program Development (Ottawa: Public History Inc, 2006), online: 2017] SOVEREIGNTY IN THE ARCTIC 381 Historic internal waters has become paramount in establishing continuity of the use of the NWP. The Western Sahara Advisory Opinion determined that Nomadic peoples can acquire and transfer sovereign rights over land.75 Thus, this judgement could be argued to apply to the Inuit, as they have hunted, fished, travelled and lived as Nomadic peoples in the Canadian Arctic for millennia.76 As a result, Canada could refer to the 1993 Nunavut Land Claims Agreement. At section 2.7.1 of the Agreement, the Inuit release and surrender all of their sovereign rights. The section reads as follows: In consideration of the rights and benefits provided to Inuit by the Agreement, Inuit hereby: cede, release and surrender to Her Majesty The Queen in Right of Canada, all their aboriginal claims, rights, title and interests, if any, in and to lands and waters anywhere within Canada and adjacent offshore areas within the sovereignty or jurisdiction of Canada; and agree, on their behalf, and on behalf of their heirs, descendants and successors not to assert any cause of action, action for a declaration, claim or demand of whatever kind or nature which they ever had, now have or may hereafter have against Her Majesty The Queen in Right of Canada or 2017 CanLIIDocs 175 any province, the government of any territory or any person based on any aboriginal claims, rights, title or interests in and to lands and waters described in Sub-section (a).77 Section 2.7.1(a) and 2.7.1(b) are very clear in portraying that the Inuit have ceded, released and surrendered all of their rights to the Canadian government. This will be met with controversy due to the historical relationship between Canada and the Inuit peoples. This claim will fail if Canada and the Inuit peoples do not establish a Canadian-Inuit partnership in as it is vital to the factors of this line of reasoning. If this occurs, Canada, prima facie, has a strong and cogent legal claim. X. The U.S. Legal Claim Although the U.S. have not been aggressive opponents in the NWP dispute, they remain steadfast in their position that Canada has no basis at international law to support the claim of internal waters.78 The U.S. Department of State released a statement with regard to the Canadian claim over the NWP in 1970 that sheds light on the underlying reason as to why the U.S. claims that the NWP is an international strait. The statement reads: “We cannot accept the assertion of a Canadian claim that 75 Western Sahara, Advisory Opinion, [1975] ICJ Rep 12 at 56, online: 76 Byers, supra note 10 at 50. 77 Nunavut Land Claims Agreement Act, SC 1993 c 29 (emphasis added). 78 Huebert, supra note 37. 382 UNBLJ RD UN-B [VOL/TOME 68 the Arctic water are internal waters . . . Such acceptance would jeopardize the freedom of navigational essential for the United States naval activities worldwide.”79 The U.S. have historically challenged the legality of states’ sovereign authority if such authority affects U.S. interests.80 As mentioned, they are seeking the international status that grants their vessels, both commercial and military, the highest degree of freedom of unimpeded navigation. As Ted McDorman states the “U.S position regarding the NWP may be best described as benign neglect.”81 This is a very telling statement regarding the U.S.’s attitude toward NWP dispute. However, the U.S.’ claim is more straightforward than Canada’s and, prima facie, legally well- founded. The U.S. claim that the NWP is an international straits rests on a plain meaning interpretation of Article 37 of the 1982 UNCLOS which states that international strait s are straits which are used for international navigation between one part of the high seas or an exclusive economic zone and another part of the high seas or an exclusive economic zone.82 It is clear that the NWP satisfies this definition. However, referring back to the Corfu Case, there are two heads of criteria 2017 CanLIIDocs 175 that exist when determining whether a ‘Strait’ is an international strait: 1) geographic and 2) functional. The U.S. attests that satisfaction of the geographic criterion is enough to determine whether the NWP is an international strait at international law. However, as was mentioned, more emphasis has been placed on the functional criterion since the Corfu Channel Case. There have been very few NWP transits to date.83 An increase in international transiting through the NWP will be required if the functional criterion is to be satisfied. On this point, attention must be drawn to the wording of Article 37 of The 1982 UNCLOS - transit passage. It reads: “This section applies to straits which are used for international navigation between one part of the high seas or an exclusive economic zone and another part of the high seas or an exclusive economic zone.”84 Emphasis has been placed on “which are used for” as this can be interpreted as requiring the designation of an international strait to be conditional on actual use. International lawyers and academics agree with this interpretation. For example, Harvard international law professor Richard Baxter puts forth that, “International waterways must be considered to be those rivers, canals, and straits which are used to a substantial extent by the commercial shipping or warships.”85 Ultimately, if the 79 McDorman, supra note 6 at 236. 80 David Welch, “Disputology: The US and East Asia’s Sovereignty Disputes” (June 2015), Balsillie School of International Affairs (blog), online: 81 McDorman, supra note 6 at 228. 82 UNCLOS, supra note 5 at part III. 83 Byers, supra note 10 at 5. 84 UNCLOS, supra note 5 at part III (emphasis added). 85 Byers, supra note 10 at 55. 2017] SOVEREIGNTY IN THE ARCTIC 383 U.S. can satisfy both criteria outlined in the Corfu Channel Case, then they would have a cogent argument that they enjoy the right of passage that is sought; which is, the right of ‘transit passage’ – which shall not be impeded.86 XI. Concluding Opinion The Arctic region is undoubtedly changing. Arctic ice pack and Arctic sea ice will continue to decrease while states gradually recognize the significance of their economic and geopolitical interests in the Arctic region. This report has focused on the NWP dispute regarding the sought after international legal statuses that could exist in its waters. It is evident that each of these status’ are defensible when approached through the lens of the Canada-U.S. dispute. The global importance of this dispute will continue to increase as the many beneficial uses of the NWP continue to be discovered. This dispute will not remain dormant. Canada’s strongest claim is based on the waters within the NWP having the legal status of Canadian historic internal waters. This status would accord Canada the authority to deny all 87 foreign vessels, both commercial and military, the use of the NWP. This is the 2017 CanLIIDocs 175 reason that many states, the U.S. in particular, do not recognize Canada’s claim. As the economic, strategic, and military benefits that stem from the transit of the NWP increase, more states are going to seek as much freedom and rights during vessel passage as legally attainable. This has been made clear by the recent public display of ambition from China through the publication of the Guide regarding the transit of the NWP.88 A complication to the NWP dispute is that the U.S. is not a party to the 1982 UNCLOS, although they frequently state that they consider most of UNCLOS’s provisions to be customary international law.89 This discrepancy has led to the avoidance of both Canada and the U.S. from applying to the ITLOS or ICJ for adjudication of their claims. In the event of a hypothetical ITLOS or ICJ adjudication, the U.S. claim would likely be favoured. This report agrees, in part, with the opinion expressed by Donat Pharand: Canada is not in the position to discharge the heavy burden of proof that they have exercised jurisdiction over the Arctic waters for a sufficiently long period of time and without protest from the U.S. or other states.90 Another problem that exists with the Canadian case is the argument that the Arctic straight baselines are not subject to Article 8(2) – internal waters of the 1982 UNCLOS. At international law, a state that signs a treaty is obliged to refrain, in good faith, from 86 UNCLOS, supra note 5 at part III. 87 Lowe & Churchill, supra note 11 at 61. 88 VanderKlippe, supra note 20. 89 James L Malone, “The United States and the Law of the Sea After UNCLOS III” (1983) 46:2 Law & Contemp Probs 29. 90 Pharand, supra note 38 at 223. 384 UNBLJ RD UN-B [VOL/TOME 68 acts that would defeat the object and purpose of the treaty. Signature alone does not impose on the state obligations under the treaty.91 Canada should only be subject to the principles of customary international law; however, past action of drawing straight baselines around the Arctic Archipelago may be considered an act that defeats the object and purpose of the 1982 UNCLOS. The purpose of the implementation of the 1985 straight baselines was to assert that there exists no right of passage within the Canadian Arctic Archipelago. This act is inconsistent with the 1982 UNCLOS. Further, Canada may have difficulty discrediting the U.S. claim as it is undisputable that the NWP can be used for international navigation. It is undeniable that the NWP connects a part of the high seas or an exclusive economic zone with another part of the high seas or an exclusive economic zone. Based on the Vienna Convention on the Law of Treaties – Article 31 – General Rule of Interpretation, of which Canada is a party: 1. A treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose.92 2017 CanLIIDocs 175 Thus, if interpreted in its ordinary meaning, the NWP is an international strait. As outlined in the Corfu Channel Case, in the event that the geographic criterion is satisfied for an international strait, the functional criterion must also be satisfied. Due to the current conditions in the Arctic, frequency of vessels transiting the NWP is low. However, this is gradually changing. Canada needs to be wary of an increase in international shipping activity in the Arctic. If the frequency of foreign vessels transiting the NWP increases, then the claim that the NWP is an international strait satisfies the second criterion and gains legitimacy. Once the international status of an international strait is globally recognized, Canada will lose their claim permanently. Adding to the concern is the current lack of Canadian presence and legal capacity in the Arctic to regulate, manage and control the NWP. Canada will have difficulty establishing the merits of their internal waters claim if examined by the judiciary. This is not due to the legitimacy of the claim, but to the mismanagement of the dispute itself. Canada’s piecemeal public assertions of jurisdiction and lack of consistency as to the status they assert will weakens their claim. The Canadian line of reasoning will require inferences from a judiciary to establish the evidence needed for historical title. If Canada is to succeed in the NWP dispute, their success will require a genuine partnership with the Inuit peoples to strengthen the historic internal waters claim. This claim has flaws, but it is here that this report has a disagreement with Donat Pharand. Canada is in a position to discharge the heavy burden of proof that they have exercised jurisdiction over the NWP if they stand with the Inuit Nation as one. There is no principled reason to distinguish the Inuit peoples from the Nomadic peoples of Western Sahara. The Inuit Nation can acquire and transfer sovereign rights and have done so as stated in 1993 91 Vienna Convention on the Law of Treaties, 23 May 1969, arts 10, 18 (entered into force 27 January 1980), online: 92 Ibid. 2017] SOVEREIGNTY IN THE ARCTIC 385 Nunavut Land Claims Agreement. Together, all parts of the claim can potentially be affirmed, granting Canada a much stronger claim than the U.S. The two claims assessed in this report are cogent, but the question remains: what will become of the NWP? Ice is present for most of the calendar year, and the NWP’s viability is far from being able to accommodate large-scale commercial and military shipping. In the interest of Canada and the U.S., the next step that should be taken relates to the current Canada-U.S. agreement - The 1988 Arctic Cooperation Agreement. Both states should compromise to update this agreement before another state aggressively joins this dispute. The turning of the capitalistic tide in the U.S. will take notice of China’s ambition in the use of the NWP. This will increase the likelihood of adjudication of a dispute that was thought to be dormant. Greater attention to this dispute will likely motivate Canada to assert their sovereignty in the Arctic region. It will also decrease Canada’s vulnerability to gradually losing strands of sovereignty – until the NWP is recognized as an international strait – due to increases in vessel traffic in the Arctic. In any event, Canada has the opportunity to lead the global conservation regarding the NWP, Arctic environment and its peoples and to build relations that allow for cooperation. The NWP dispute is one of the most 2017 CanLIIDocs 175 important disputes in Canadian history and should be considered of paramount importance in the coming years. The outcome of this dispute is more than a determination of the international legal status of the NWP – it is a critical test of Canada’s ability to assert their sovereignty as an Arctic state, and more importantly, as a world power.