Tournament of Appeals : Granting Judicial Review in Canada [1 ed.] 9780774851022, 9780774810821

Citation preview

Tournament of Appeals

flemming1.p65

1

01/04/2004, 2:23 PM

Law and Society Series W. Wesley Pue, General Editor

Gender in the Legal Profession: Fitting or Breaking the Mould Joan Brockman Regulating Lives: Historical Essays on the State, Society, the Individual, and the Law Edited by John McLaren, Robert Menzies, and Dorothy E. Chunn Taxing Choices: The Intersection of Class, Gender, Parenthood, and the Law Rebecca Johnson Collective Insecurity: The Liberian Crisis, Unilateralism, and Global Order Ikechi Mgbeoji Murdering Holiness: The Trials of Franz Creffield and George Mitchell Jim Phillips and Rosemary Gartner Unnatural Law: Rethinking Canadian Environmental Law and Policy David R. Boyd People and Place: Historical Influences on Legal Culture Edited by Jonathan Swainger and Constance Backhouse Feminist Activism in the Supreme Court: Legal Mobilization and the Women’s Legal Education and Action Fund Christopher P. Manfredi Compulsory Compassion: A Critique of Restorative Justice Annalise Acorn The Heiress vs the Establishment: Mrs. Campbell’s Campaign for Legal Justice Constance Backhouse and Nancy L. Backhouse

flemming1.p65

2

01/04/2004, 2:23 PM

Roy B. Flemming

Tournament of Appeals: Granting Judicial Review in Canada

flemming1.p65

3

01/04/2004, 2:23 PM

© UBC Press 2004 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without prior written permission of the publisher, or, in Canada, in the case of photocopying or other reprographic copying, a licence from Access Copyright (Canadian Copyright Licensing Agency), www.accesscopyright.ca. 15 14 13 12 11 10 09 08 07 06 05 04 5 4 3 2 1 Printed in Canada on acid-free paper National Library of Canada Cataloguing in Publication Flemming, Roy B. Tournament of appeals: granting judicial review in Canada / Roy B. Flemming. (Law and society series, ISSN 1496-4953) Includes bibliographical references and index. ISBN 0-7748-1082-3 1. Canada. Supreme Court. 2. Judicial review – Canada. 3. Political questions and judicial power – Canada. 4. Judicial power – Canada. I. Title. II. Series: Law and society series (Vancouver, B.C.) KE8244.F54 2004 KF8764.ZA2F54 2004

347.71’035

C2004-900824-2

UBC Press gratefully acknowledges the financial support for our publishing program of the Government of Canada through the Book Publishing Industry Development Program (BPIDP), and of the Canada Council for the Arts, and the British Columbia Arts Council. This book has been published with the help of a grant from the Canadian Federation for the Humanities and Social Sciences, through the Aid to Scholarly Publications Programme, using funds provided by the Social Sciences and Humanities Research Council of Canada. UBC Press The University of British Columbia 2029 West Mall Vancouver, BC V6T 1Z2 604-822-5959 / Fax: 604-822-6083 www.ubcpress.ca

flemming1.p65

4

01/04/2004, 2:23 PM

To Toni ... In a way it all began in Canada, didn’t it?

flemming1.p65

5

01/04/2004, 2:23 PM

flemming1.p65

6

01/04/2004, 2:23 PM

Contents

Figures and Tables / ix Preface and Acknowledgments / xi 1 Judicial Review and Agenda Setting: American Accounts and the Canadian Setting / 1 2 Tournaments and Stratification of Canada’s Supreme Court Bar / 19 3 Litigants, Lawyers, and the Tournament of Appeals / 43 4 Tournament Rules and the Consequences of Institutional Choices / 61 5 The Court Game: Strategy and Consensus among the Tournament’s Judges / 77 6 Conclusion: Canadian Variations on American Themes / 98 Notes / 107 References / 110 Index / 116

flemming1.p65

7

01/04/2004, 2:23 PM

flemming1.p65

8

01/04/2004, 2:23 PM

Figures and Tables

Figures

1.1 Pathways to the agenda of the Supreme Court of Canada / 7 1.2 Steps in the leave to appeal process / 14 5.1 Ideological differences between justices in Charter cases: Libertarian versus communitarian dimension / 82 5.2 Leave to appeal decision tree / 84 5.3 Hypothetical agenda game with one panel dissenter / 86 5.4 Hypothetical probability distribution of leave outcomes / 89 Tables

2.1 Comparisons of national bar, leave attorneys, and lawyers who responded to the Survey of Leave to Appeal Attorneys, 1993-95 / 27 2.2 Lawyers’ litigation experience before Canada’s Supreme Court, 1993-95 / 30 2.3 Attorney characteristics and Supreme Court experience, 1993-95 / 31 2.4 Advice giving and prior Supreme Court experience, 1993-95 / 33 2.5 Variable definitions and characteristics / 34 2.6 A professional status model of leave activity / 38 3.1 Definitions and characteristics of variables used in this study / 47 3.2 Applicant success rates by party advantage, leaves to appeal, 1993-95 / 48 3.3 Applicant success rates by lawyer experience advantage, leaves to appeal, 1993-95 / 49 3.4 Applicant success rates by Queen’s Counsel advantage, leaves to appeal, 1993-95 / 50 3.5 Applicant success rates by attorney firm advantage, leaves to appeal, 1993-95 / 51 3.6 Logit model of leave to appeal outcomes, government and private litigants, 1993-95 / 52

flemming1.p65

9

01/04/2004, 2:23 PM

x

Figures and Tables

3.7 Litigant advantage model of leave to appeal outcomes, government and private litigants, 1993-95 / 54 3.8 Interactive logit model of leave to appeal outcomes, government and private litigants, 1993-95 / 56 3.9 Logit model of leave to appeal outcomes, private litigants only, 1993-95 / 57 4.1 Definitions and characteristics of major variables used in this study / 68 4.2 Alternative logit models of agenda setting in Canada / 69 4.3 Leave to appeal panels, 1990-96 / 74 4.4 Panel-by-panel model of leave to appeal decisions, 1993-95 / 75 5.1 Leave to appeal panels and “preferred voting partners” on the merits / 95

flemming1.p65

10

01/04/2004, 2:23 PM

Preface and Acknowledgments

High courts or courts of final appeal set their agendas when they have the authority to pick and choose appeals of lower court decisions and place them on their dockets for judicial review. There is a very large literature on agenda setting drawn from repeated studies of the Supreme Court of the United States, but little is known about this process in other countries. Not only is there a dearth of comparative research dealing with how high courts pick appeals for review, but what literature exists generally lacks any grounding in the American literature with its strong social science orientation. This study of the leave to appeal process in Canada is the first of its kind in that it builds on the US literature and adopts an explicit social science approach. While the Supreme Court Law Review published annually in Canada has included a chapter describing in general terms the preceding year’s leave to appeal activity before the Court, and despite successive biennial editions of the Crane and Brown handbook, Supreme Court of Canada Practice, which review the legal and procedural aspects, empirical research has not yet appeared that either probes the dynamics of how Canada’s Supreme Court sets its agenda or attempts to replicate prior research done in the United States in the Canadian context. Accounts by American political scientists that explain how the US Supreme Court chooses cases were therefore adopted to see whether they provide satisfactory explanations of the Canadian process. To place these accounts and the book’s analyses in a broader context, this study uses the notion of a “tournament” as a metaphor for the process and to provide a way of fitting the various pieces of the empirical puzzle together under a single rubric. My interest in the leave to appeal process began in 1993, when I spent a month in Ottawa with the support of a Canadian Studies Program grant, developing a course on law and politics in Canada. As there was

flemming1.p65

11

01/04/2004, 2:23 PM

xii Preface and Acknowledgments

no social science literature on this process, I began to think about a research project to fill this void. A second Canadian Studies Program grant supported the development of a proposal to the National Science Foundation in the United States that requested financial support to collect data on the leave to appeal process. The Foundation funded this proposal (SBR-9515025) and data collection in Ottawa commenced in May 1996. A Senior Fellowship Grant through the Canadian Studies Program in 1999 provided additional support for the completion of the project. During the spring of 2002, I began revising the manuscript and wrote a draft of Chapter 5 while I was a Fulbright Research Fellow at the Institute of Canadian Studies at the University of Ottawa. Canada Post provided the funding for my fellowship at the Institute. I am deeply indebted to the Canadian Embassy in Washington, DC, which administers the Canadian Studies Program; the National Science Foundation; Canada Post; and the Canada-US Fulbright Program for their support of this project. I also owe a great deal to many people without whom this project would not have seen the light of day. From my first visits to the Supreme Court in 1993 to the present, the Court’s staff in the Registrar’s Office and its staff in the Court’s library have been invariably courteous and unfailingly helpful. Barbara Kincaid, General Counsel for the Supreme Court, who supervises the Law Branch of the Court, which is responsible for reviewing leave applications and editing the Court’s judgments prior to release, provided valuable information on the operation of the leave process and has not hesitated to call attention to my errors. During the summer of 1996, Rita Banerjee, Marie Crowley, Anick Demers, and Lynn Marchildon collected a major portion of the data for this project before they returned to law school at the University of Ottawa in the fall. I should add that Ted Morton’s tip that I hire University of Ottawa law students when I visited him at the University of Calgary proved to be right on the money; his suggestion immeasurably improved the odds that this project would succeed. Last but not least, Kevin McGuire was kind enough to let me adapt for this study the questionnaire he prepared for his important book, The Supreme Court Bar, the first study to explore the world of the elite attorneys with litigation practices in the United States Supreme Court. When I returned to College Station, Texas, I depended on Kristi Campbell and Shamika Fulson to do the very tedious job of preparing the mailing list of attorneys for the survey that formed the second major part of the project. Combing the Canadian Law List for the addresses

flemming1.p65

12

01/04/2004, 2:23 PM

Preface and Acknowledgments

of the lawyers whose names we collected from the factums in the Supreme Court during the summer was no idle stroll through the park. Once the mailing list was compiled, Donna Dunlap, who worked with me after I became Director of Graduate Studies for the Political Science department at Texas A & M, applied her special administrative skills to organizing and monitoring the mail survey. (Marie Crowley translated the questionnaire into French.) Donna had her hands full during our four years together in the Graduate Office and I am very grateful for her help with the project. For coding the Supreme Court data we collected in Ottawa and from the questionnaires lawyers returned to us, plus the construction of variables, I relied heavily on Jennifer Renee Schwank and Glen S. Krutz, who at the time were graduate students in our department. They were deeply involved in the data analysis and co-authored with me conference papers based on their analyses. Glen subsequently became a coauthor with me on two publications that were precursors of Chapters 3 and 4. Because I could depend implicitly on their intelligence, resourcefulness, and eagerness to learn, our work was genuinely collaborative. During 2002 Quan Li has helped me tie up the manuscript’s loose ends with equal diligence and care. Bright and capable graduate students like Jennifer, Glen, and Quan make the world of academic research go around. A final acknowledgment must be made to the roughly thirty repeat player attorneys and several former clerks to Canada’s Supreme Court whom I interviewed over the course of a month or so in Toronto before my return to Texas in 1996. Their willingness to share their understandings of the leave to appeal process helped me form some preliminary notions of how these players in the tournament of appeals saw their roles and how they interpreted the process. What I have produced here, of course, is my own responsibility and I’m sure will surprise many of them. For all the support, cooperation, advice, and counsel I have received since the inception of the project, I alone am responsible for any factual and interpretative errors in this book.

flemming1.p65

13

01/04/2004, 2:23 PM

xiii

flemming1.p65

14

01/04/2004, 2:23 PM

Tournament of Appeals

flemming1.p65

15

01/04/2004, 2:23 PM

flemming1.p65

16

01/04/2004, 2:23 PM

1 Judicial Review and Agenda Setting: American Accounts and the Canadian Setting

In December 2002 the provincial court of appeals for Newfoundland and Labrador released a blistering opinion aimed directly at the Supreme Court of Canada.1 The case involved a pay equity suit filed by women employed by the province after the government, facing a budget crunch, reneged on a settlement to which it had agreed earlier. The judgment upholding the government ran to more than 230 pages. The opinion in lengthy and considerable detail accused the Supreme Court of “undue incursions” into policy domains rightfully under the prerogatives of Parliament and of the provincial assemblies. It specifically challenged the method developed by the justices and used by them to interpret the Canadian Charter of Rights and Freedoms. Feelings often run high about judicial review and the power courts display when they exercise this authority. For many observers of Canada’s Supreme Court, judicial activism has become the sine qua non of judicial review since the Charter of Rights and Freedoms was constitutionally entrenched in 1982. For example, in 1999 the Globe and Mail announced, “Ottawa to Enshrine Same-Sex Rights.”2 The Supreme Court set the stage for the government’s actions in Egan v. Canada (1995), a case in which Canada’s Old Age Security Act was challenged on the grounds that its allocation of benefits favoured heterosexual married couples. Although the Court rejected the challenge by a 5-4 vote, it nevertheless ruled for the first time that the Charter of Rights and Freedoms prohibited discrimination based on sexual orientation. Egan v. Canada encouraged numerous lawsuits, including one that challenged fifty-eight federal statutes. In February, a month after the Globe and Mail story, the Reform Party of Canada held its “United Alternative” convention, which led to the formation of the Canadian Alliance. Criticism of the Supreme Court crackled throughout the convention as delegates decried “judicial activism.” The National Post reported: “No issue touched the emotions

flemming2.p65

1

01/04/2004, 2:24 PM

2

Judicial Review and Agenda Setting

and united the delegates ... more than calls to rein in the power of Canada’s judges and human-rights bureaucrats.”3 The Supreme Court’s selection of cases and the issues raised by them clearly has far-reaching effects. Which cases are heard mould the development of the law, but equally important, as Egan v. Canada shows, the choice of cases and the Court’s emphasis on particular areas of the law can lead to major public policy changes. When placing appeals for judicial review on its plenary docket, the Court also creates winners and losers, as some lower court decisions are left to stand while others become subject to judicial review and possibly to being overturned. At the same time, the decision to hear some cases and not others imposes the Court’s priorities on the politics of the country and its government. As the news stories from the “United Alternative” convention suggest, many politically active Canadian citizens do not share the Supreme Court’s priorities. By setting its agenda, then, high courts elevate the public visibility of issues of concern to some groups while downplaying issues of interest to other groups. For the Supreme Court in Canada, an important ancillary step toward the Court’s independence in interpreting the Charter’s provisions came seven years earlier, when Parliament expanded the Court’s discretion in selecting cases for judicial review. Parliament’s only stipulation was that the Court’s decisions would be governed by whether the appeals involved issues of “public importance.” The content of this standard was left in the hands of the justices to determine, however, as well as the procedures they would follow when exercising this new grant of authority. As the provincial appellate judges in Newfoundland and Labrador made very clear in their complaints about the Supreme Court, how courts make decisions is often as important as what decisions they make. Indeed, the power to decide what to decide, which is at the heart of setting a court’s agenda, carries with it in most instances the freedom to choose how to choose. In other words, the ways in which judges on high courts or courts of final appeal arrange their decision-making processes may fundamentally affect the array of choices they consider and perhaps the nature of the choices itself. There is a very limited literature that explains how high courts other than the United States Supreme Court grant judicial review and construct their agendas. What literature exists outside the United States is formalistic and descriptive. This book is the first systematic investigation of agenda setting in Canada’s Supreme Court and perhaps the only study to understand agenda setting in any high court that empirically tests American findings to determine whether they can be extended to explain the behaviour of

flemming2.p65

2

01/04/2004, 2:24 PM

American Accounts and the Canadian Setting

high courts in other countries. Despite the Court’s emergence as an influential policy maker, little attention has been paid to the way in which the Court selects cases for review. The first survey of the Supreme Court’s application for leave to appeal process appeared in 1982 in the Supreme Court Law Review. The Review continues to publish these annual surveys, but they are primarily “practice notes” addressed to the legal profession and are of limited analytical value. At present, social science research on agenda setting in Canada’s Supreme Court does not exist. The rest of this chapter outlines and compares the processes in Canada and the United States as a prelude to the subsequent chapters. American Accounts of Granting Judicial Review

How can agenda setting in Canada be explained? What goes into the decision to grant judicial review to one appeal but not another? Tentative answers to this question can be found in the literature on the writ of certiorari process in the United States Supreme Court.4 Certiorari, or “cert,” is the means by which the Supreme Court grants judicial review to applicants appealing lower court decisions. The American literature offers three accounts or general hypotheses to explain how the Court exercises its discretion when granting judicial review. The first and most developed account is litigant-centred. In the United States, the type of litigant – that is, whether “upperdog” or “underdog” – is related to the voting behaviour of justices in the agenda-setting process. Higher-status litigants, the upperdogs, more often gain access to the Supreme Court’s docket than underdog or low-status litigants, although this relationship depends in part on the Court’s ideological preferences (Ulmer 1978, 1981). Other studies show that the presence of interest groups supporting the parties influences the selection of cases for judicial review (Caldeira and Wright 1988, 1990a). The US Supreme Court grants certiorari more often to parties backed by organized interests than to parties lacking this support. The growth in interest group participation in litigation before the US Supreme Court and its impact on the Court’s certiorari process is well documented (Epstein 1991; Caldeira and Wright 1988, 1990a; McGuire 1993, 1994). A similar development has occurred in Canada, where the Supreme Court has relaxed its rules on standing (Bogart 1994; Gertner 1988; Sossin 1999) and rarely declines requests by groups to intervene at the merits stage when the Court reviews appeals (Brodie 2002; Lavine 1992; Welch 1985). Organized interests are thus increasingly common participants in Supreme Court cases and have received financial support from the federal government to encourage their activities (Brodie 2002; Epp 1998). Finally,

flemming2.p65

3

01/04/2004, 2:24 PM

3

4

Judicial Review and Agenda Setting

parties represented by lawyers who are repeat players with experience and expertise before the US Supreme Court have an edge over opponents with counsel who are less experienced in the Court’s agenda-setting decisions (McGuire and Caldeira 1993; McGuire 1995a, 1995b). Experienced lawyers may also recruit interest group support (McGuire 1994). A jurisprudential account comprises the second approach. In this account, justices apply legal considerations to requests for judicial review, and legal factors weigh heavily in their deliberations (Perry 1991; Provine 1980; Stern et al. 1993; Ulmer 1984). This hypothesis does not mean that the process is mechanical or rigidly rule-dominated. Instead, legal factors prompt justices to give certiorari applications a second look in a process otherwise strongly governed by the presumption that few requests for review warrant approval. Perry (1991) offers the most thorough exploration of this account based on his interviews with several Supreme Court justices and over fifty clerks in the Court. His itemization of the various factors included in the jurisprudential account is the source for this book’s test of the account in Canada. The third account focuses on the strategic choice behaviour of the justices as they construct their agenda. In broad terms, this hypothesis suggests that justices anticipate the likely outcomes of cases if they are heard on the merits; the justices then cast their votes on whether to grant judicial review according to whether these expected outcomes coincide with their policy views. Recent elaborations of this hypothesis argue that justices pursue particular strategies, such as “aggressive grants” or “defensive denials” (Boucher and Segal 1995; Perry 1991; Caldeira et al. 1999). The question that immediately comes to mind, of course, is whether Canada’s justices play a similar “leave to appeal game.” While individual pieces of the literature in the US gravitate toward one or the other of the three accounts, the three hypotheses are not mutually exclusive. For example, Perry (1991, 278) argues that certiorari decisions may reflect both jurisprudential and strategic considerations. Moreover, the “indices” and “signals” that he says lead the justices to take closer looks at petitions for judicial review include elements from the litigant-resources hypothesis such as the identities of the litigants or their counsel (Perry 1991, 113-39). A comprehensive understanding of agenda setting in Canada’s Supreme Court and how it picks appeals from the lower courts for judicial review clearly requires consideration of all three hypotheses. The value or utility of these three accounts undoubtedly varies with the institutional characteristics of particular high courts. Canada’s Supreme

flemming2.p65

4

01/04/2004, 2:24 PM

American Accounts and the Canadian Setting

Court shares many similar features with the US Supreme Court but there are important differences that will be noted later in this chapter. As a consequence, this comparison between the two courts can test the generality of the American accounts while providing new perspectives on agenda setting in Canada’s Supreme Court. On one very important matter the two high courts are almost identical; both have wide sway over the appeals they choose to hear. In 1975 Canada’s Parliament amended the Supreme Court Act to limit the right to appeal in civil cases and, with a couple of exceptions, in criminal cases as well. The 1975 amendment, analogous to the “Judges’ Bill of 1925” that gave the US Supreme Court discretion over what to hear, freed the hands of Canada’s justices so that they could select appeals for judicial review according to their “public importance.”5 When Parliament limited the right of appeal, it made the discretionary leave to appeal process the main avenue to judicial review. For the first time, Canada’s Supreme Court held substantial control in its hands over the kinds of cases it wished to hear.6 The 1975 reform gave Canada’s Supreme Court wide latitude, declaring that the decision to grant leave to appeal rested on the Court’s determination of the “public importance” of issues raised by an application. According to Section 40(1) of the Supreme Court Act, applications would be granted leave if The Supreme Court is of the opinion that any question involved therein is, by reason of its public importance or the importance of any issue of law or any issue of mixed law and fact involved in such question, one that ought to be decided by the Supreme Court or is, for any other reason, of such a nature or significance as to warrant decision by it.

This section does not define “public importance,” and it is as vaguely phrased as Rule 10 governing writ of certiorari decisions in the United States. Certiorari, or “cert” in the vernacular, is an order to the lower court to forward the record of the case to the Court. Rule 10, written by the Supreme Court to its own specifications, reads as follows: Review on writ of certiorari is not a matter of right, but of judicial discretion. A petition for a writ of certiorari will be granted only for compelling reasons. The following, while neither controlling nor fully measuring the Court’s discretion, indicate the character of reasons the Court considers:

flemming2.p65

5

01/04/2004, 2:24 PM

5

6

Judicial Review and Agenda Setting

(a) A United States court of appeals has entered a decision in conflict with the decision of another United States court of appeals on the same important matter; or has decided an important federal question in a way that conflicts with a state court of last resort; or has so far departed from the accepted and usual course of judicial proceedings, or sanctioned such a departure by a lower court, as to call for an exercise of this Court’s power of supervision. (b) When a state court of last resort has decided an important federal question in a way that conflicts with the decision of another state court of last resort or of a United States court of appeals. (c) When a state court or a United States court of appeals has decided an important question of federal law which has not been, but should be, settled by this Court, or has decided an important federal question in a way that conflicts with applicable decisions of this Court. A petition for a writ of certiorari is rarely granted when the asserted error consists of erroneous factual findings or the misapplication of a properly stated rule of law.7

Rule 10 underscores and emphasizes the US Supreme Court’s discretion. While it appears to stress conflicting lower court decisions as the chief criterion governing certiorari decisions, for instance, this interpretation is immediately undercut by the declaration that the criterion, and indeed even Rule 10 taken as a whole, is neither controlling nor a full measure of the Court’s discretion. Like their Canadian peers, the American justices do not identify what “compelling reasons” would lead to grants of certiorari, nor did they define “compelling” or “important.” As one eminent scholar of the certiorari process put it: “In short, the rule is almost a tautology: cases are important enough to be reviewed by the justices when justices think they are important” (Perry 1992).8 One Docket, Three Agendas: Pathways to the Supreme Court

An agenda focuses attention on particular issues or concerns to the exclusion of others. At a minimum, the ability to place issues before decision makers means that other issues may go unattended, perhaps against the preferences of the decision makers themselves. Institutions influence the agenda access of actors. In contrast to the situation in the United States, Canada’s Supreme Court does not have complete control over the appeals or issues it hears. Instead, it is useful to think of the Court’s docket as being made up of three agendas, only one of which is under full control of the justices. The first agenda consists of cases over

flemming2.p65

6

01/04/2004, 2:24 PM

American Accounts and the Canadian Setting

Figure 1.1 Pathways to the agenda of the Supreme Court of Canada

Applications for leave to appeal

Appeals as of right in criminal cases

Decision to grant or deny leave to appeal

Leave granted

Oral arguments on appeals and references

Judgments from bench after oral arguments

Leave denied

Reference questions (primarily) constitutional

Judgments reserved after oral arguments

Formal judgments and opinions

Source: Revised and adapted from Supreme Court of Canada 2001.

which the Court has discretion to grant leave to appeal. The second agenda is made up of residual “appeals as of right” in criminal cases, over which the Court has less control. The third includes reference questions put to the Court by the federal government or the Governor-inCouncil. As a general matter, the justices feel compelled to hear references because of their importance, although the Court may decline to address some issues if they are too vaguely phrased or too hypothetical. This mix of agendas means that the Court impresses its priorities on only a portion of its docket. Figure 1.1 diagrams the three streams of cases or questions for which the Court renders judgments and opinions. Requests for leave to appeal and those granted leave have been the wellspring for judicial review since 1975. Parliament’s amendment dramatically changed the balance between the discretionary and nondiscretionary sides of the Court’s docket. During the term prior to 1975, cases granted leave to appeal accounted for 29 percent of the Court’s

flemming2.p65

7

01/04/2004, 2:24 PM

7

8

Judicial Review and Agenda Setting

docketed caseload. One year later, the proportion rose to 60 percent, and by 1980-81 the proportion was 74 percent (Bushnell 1982). During the 1990s, applications for leave to appeal ranged from a low of 445 in 1991 to a high of 637 in 1997; the annual volume averaged 526 applications per year (Supreme Court of Canada 2001). Cases appealed as a matter of right constitute a smaller but nonetheless sizable part of the Court’s docket; the number of notices to appeal as of right averaged forty-five a year during the 1990s, although it dropped by over half in the last three years of the decade (Supreme Court of Canada 2001). Because many of these notices fall by the wayside, the actual size of this agenda shrinks to about twenty-five appeals as of right per year, still a significant portion of the justices’ workload. For example, 29 of the 110 opinions published by the Supreme Court in 1994 were appeals as of right placed directly on the Court’s docket by criminal defendants or by the Crown. In general, appellants in criminal cases whose acquittals at trial are overturned by appellate courts as a result of appeals by Crown prosecutors have an automatic right of appeal. Similarly, the Crown has a right of appeal if the appellate court overturns a trial conviction. Appeals as of right also occur when an appellate judge dissents on any question of law. Since these appeals force the Supreme Court to review cases it might otherwise ignore or leave undisturbed, the justices often render brief judgments from the bench after frequently truncated oral arguments. The following two cases are typical of the brief judgments made in other appeals as of right. In the first case, the justices issued a judgment from the bench in a case from Nova Scotia that dismissed an appeal of a conviction for second-degree murder. A dissenting opinion opened the Supreme Court’s doors for this appeal. This is an appeal as of right. The basis of the dissent of Jones J.A. in the Court of Appeal is whether a warning pursuant to Vetrovec v. The Queen, [1982] 1 S.C.R. 811, was appropriate in the circumstances of this case. This [appeal] does not raise a question of law although failure to give a warning in some circumstances may constitute a miscarriage of justice. The appellant, therefore, does not have an appeal as of right to this Court. Having heard the appellant on the merits, however, we are of the view that, even if the Court had jurisdiction to entertain the appeal, we would dismiss it. In the circumstances of this case, failure to give a Vetrovec warning to the jury did not result in a miscarriage of justice. The appeal is, therefore, dismissed.

flemming2.p65

8

01/04/2004, 2:24 PM

American Accounts and the Canadian Setting

Another appeal as of right came from the New Brunswick Court of Appeal when the majority on the bench overturned a trial conviction, allowing the Crown to bring the case to the Supreme Court for review. The Supreme Court’s bench opinion included the following comment. The comment also shows that appeals as of right are not necessarily futile efforts for the parties who take advantage of them, which explains why attempts to eliminate them entirely from the Court’s docket have been unsuccessful. [A]lthough the trial judge erred with regard to the requisite intent required for the section, the error benefited the respondent [the trial defendant] in that it was more onerous than required. We respectfully disagree with the majority of the Court of Appeal [citation deleted] regarding the issue of the burden of proof. The trial judge merely placed an evidentiary burden on the respondent. The ultimate burden remained on the appellant throughout. The trial judge carefully reviewed the evidence and properly concluded that the respondent should be convicted. The appeal is therefore allowed. The order of the Court of Appeal is set aside and the conviction is restored.

The Court generally handles these appeals in an expeditious manner. These appeals nonetheless take up time and attention the justices could devote to matters of greater public importance. The Supreme Court Act requires Canada’s high court to give advisory opinions on constitutional questions put to it by the federal government (Hogg 1996, 209). Advisory opinions are consistent with Canada’s Westminster parliamentary system, which does not separate legislative, executive, and judicial powers in the same ways as in the United States. Canada’s courts are considered to have non-judicial functions, and reference questions are seen as part of the Supreme Court’s executive function. It is worth noting that when Canada’s Court sat for the first time in 1876, the question before it was a reference from the Senate asking the Court’s opinion on a private bill dealing with a divisive religious issue (Crane and Brown 2002, 394). The Court usually receives one or two references a year. A steady flow of references over the years since Confederation has contributed substantially to Canada’s constitutional development.9 Between 1867 and 1966, references amounted to roughly 35 percent of all constitutional cases; more recently, the proportion has dropped to about 15 percent for the period 1967-86 (Hogg 1996, 210). The small number

flemming2.p65

9

01/04/2004, 2:24 PM

9

10

Judicial Review and Agenda Setting

of references the Court hears in a year disguises the fact that they are among the most politically volatile and potentially explosive decisions the Court makes and that they often put the Court in the middle of political controversies it might avoid if it could. Few high courts have been asked to rule, for example, on whether a state, province, or region has the constitutional right to secede from a country, as Canada’s Supreme Court did in the Quebec Secession Reference in 1998, which established Quebec’s right to separate from Canada as well as the ground rules for the process. The Lamer Court and Its Political Context

This book focuses on the years 1993-95, when Chief Justice Antonio Lamer led Canada’s Supreme Court. Justice Lamer was appointed in 1980 by Liberal Prime Minister Pierre Elliott Trudeau. After defeating the Liberals in 1984, the Progressive Conservatives won one more election in 1988 before their massive losses in 1993 nearly swept them completely out of Parliament. By 1990, when Prime Minister Brian Mulroney promoted Lamer to chief justice, Lamer was the lone surviving Liberal appointment on the bench. Lamer retired in 2000. During his ten years as chief justice, changes in the Court’s composition continued. Mulroney made two appointments before the Tories’ 1993 electoral rout; subsequently, Liberal Prime Minister Jean Chrétien made three. The departure and replacement of justices created six intervals or periods when the composition of the bench was stable. One of these “natural courts” lasted roughly seventy months, from November 1992 until September 1997; the next longest survived about twenty months. The study period of this book extends from January 1993 through December 1995, the first three years of the longest natural court. Since the number of leave applications averaged around 500 a year during the early half of the decade, a natural court that lasted more than two years was important so that the same set of justices reviewed enough applications to generate a large enough dataset for statistical analyses. Another advantage of focusing on one natural court is that it ruled out changes in the mix of justices as a factor in the decision-making process. The ten years of the Lamer Court were a time of political upheaval and intense constitutional turmoil in Canada. The 1993 election saw the demise of the Progressive Conservatives and the weakening of the New Democratic Party, the emergence of the Bloc Québécois and the Reform Party, and the first of three election victories by the Liberals under the leadership of Prime Minister Chrétien. For scholars of Canadian

flemming2.p65

10

01/04/2004, 2:24 PM

American Accounts and the Canadian Setting

politics, the 1993 election marked the end of a party system that had lasted nearly thirty years and the beginning of a new “fourth party system” (Carty et al. 2000). This new party system lacks the national focus of its predecessor as support for the parties has become heavily regionalized. Even the Liberal Party, which claims to be a national party, has won only plurality victories at the polls and relies greatly on Ontario for roughly two-thirds of its seats in Parliament. These political changes were entwined with the constitutional crises that centred on Quebec, its position within Confederation, and its demand for sovereignty and secession from Canada. On 23 June 1990, roughly two weeks before Lamer became chief justice, the deadline for ratifying the Meech Lake Accord, which among other things would have recognized Quebec as a “distinct society,” passed without the approval of two provinces, and the accord was defeated. Mulroney, keen on solidifying the Progressive Conservatives’ electoral position in Quebec and eager to resolve the constitutional anomaly of Quebec’s self-imposed exclusion from the constitutional deal of 1982, made what he referred to as a “roll of the dice” (Monahan 1991, 236) but his luck failed him. Mulroney nevertheless persisted in his efforts and two years later, in October 1992, after extensive public hearings and consultations across the country, Canadians went to the polls to resoundingly reject the more sweeping constitutional proposals known as the Charlottetown Accord (Johnston et al. 1996). These debacles in “mega-constitutional” politics (Russell 1993) in no small measure led to the near-destruction of the Progressive Conservatives in 1993. They also provoked the 1995 referendum in Quebec that brought Canada to the brink of dissolution. If 30,000 voters out of more than 4,671,000 (nearly 94 percent of eligible voters turned out) had cast oui instead of non ballots, the proposal would have been approved (Young 1998). Ottawa, stung by the closeness of the vote, instituted what became known as Plan A to accommodate some of Quebec’s demands, while Plan B laid out a more confrontational approach in dealing with the ruling provincial Parti Quebecois. A central stratagem of Plan B was the Quebec Secession Reference, heard by the Supreme Court in 1998. This decision established Quebec’s democratic right to secede while spelling out the conditions for what would be a legitimate referendum and the federal government’s role in the process. The governments and citizens of Quebec and the rest of Canada, weary and wary of further constitutional imbroglios, increasingly turned their attention to other, more mundane political questions as the nineties drew to a close.

flemming2.p65

11

01/04/2004, 2:24 PM

11

12

Judicial Review and Agenda Setting

Granting Judicial Review in Canada

During Lamer’s decade as chief justice, the number of applications for leave to appeal rose more or less steadily, from 424 in 1990 to 642 in 2000. At the same time, the success rate of applications declined from 22 percent at the beginning of the decade, which was unusually high compared with preceding years, to 13 percent by its conclusion. The Court granted leave to appeal to an average of 15 percent of applicants. In these two respects the Canadian Supreme Court differs from the US Supreme Court. For one thing, the volume of leave applications is considerably smaller than the over 9,000 writs of certiorari reviewed by the US Supreme Court in 2001 (Epstein et al. 2003, 71). A second and more telling difference is that the chances for judicial review in Canada are much higher than in the United States. Canada’s Supreme Court grants leave to appeal to an average of 15 percent, or one out of six or seven, of the applications it receives, while the United States Supreme Court grants certiorari to fewer than one out of a hundred of all the writs it receives. Canada’s justices not only consider far fewer requests for judicial review but also place a much higher proportion of the leave applications on their docket, which means that their agenda is much more accessible to litigants than that of the US Supreme Court. Shortly after Lamer became chief justice, he commented during an interview with Law Times, a Canadian professional weekly newspaper, that Charter appeals had overshadowed commercial law cases, with the consequence that provincial courts were resolving legal issues of concern to the business community.10 Lamer, claiming that several justices agreed with him, thought the occasion had arrived to “reaffirm a national perspective in private law.” According to the Law Times, the chief justice added: It is important that there be some national uniformity in how principles of commercial law are interpreted. I’m mainly aware of these concerns because I’ve frequently had lawyers up to my chamber for a coffee to ask them about how we could improve things. This kept coming up.

The chief justice’s concerns were not enough to overcome the inertia of the process. Moreover, despite the concerns of some lawyers, public law matters had not squeezed commercial matters off the Court’s docket. By one reckoning, the mix of applications granted leave by the justices during the nineties divided along the following lines: civil or private law cases accounted for about half of all the applications granted leave; criminal appeals (which included both Charter and non-Charter issues)

flemming2.p65

12

01/04/2004, 2:24 PM

American Accounts and the Canadian Setting

made up roughly one-third of the docket, with non-criminal constitutional appeals comprising the remaining one-sixth of cases. These proportions varied from year to year during the decade but there was no inversion of the Court’s long-standing priorities. Another close observer of the Court who classified the cases differently also found more signs of continuity than change during the Lamer years compared with the Court’s priorities during the preceding decade (McCormick 2000, 131). Lamer’s comment about the commercial bar’s concerns about the Court’s docket being dominated by Charter cases raises an important contrast between the agenda processes in Canada and the United States. Interest group participation in litigation before the United States Supreme Court and the presence of organized interests in the Court’s agenda-setting process are well-documented facts (Caldeira and Wright 1988, 1990a; McGuire 1994). The amicus curiae or “friend of the court” brief provides groups or associations that are not direct parties to the cases with a way to express their views about the issues being raised. In Canada, organized interests file motions with the Court to become interveners in cases, which is directly analogous to the American amici curiae. The justices rarely decline these motions (Brodie 2002). Nevertheless, interest group participation is not as widespread as in the United States (Flemming 2000). Not only are interveners typically few in number, but their involvement focuses exclusively on the stage after leave applications are granted. The Court’s rules do not forbid or preclude such briefs at the leave stage but they are rare. To the extent that amicus curiae briefs provide the US Supreme Court with some independent measures of public interest in cases, the justices in Canada have set aside interveners as a possible aid in their deliberations. Unless they are direct parties in the cases, interest group interveners are conspicuously absent in the Canadian agenda-setting process. The implications of this difference for the Canadian Supreme Court will be explored later in this book. Selecting cases for judicial review is constrained in some measure by how the Court allocates its time among its various tasks and by how many cases the justices feel they can or should decide over the course of a year. Their time is not their own, however, as appeals as of right and reference questions force themselves onto the Court’s docket, which means they reduce the time the justices might give to other cases. Canada’s justices typically set aside two weeks a month from October through June to hear oral arguments. Within this weekly block of time, they normally hear one or two cases a morning, four days of the week. During 2001, for instance, the Court scheduled ninety-six oral arguments over sixty-three days, or roughly a case and one-half for each day. Figure 1.2

flemming2.p65

13

01/04/2004, 2:24 PM

13

14

Judicial Review and Agenda Setting

Figure 1.2 Steps in the leave to appeal process

Parties and attorneys decide to file applications for leave to appeal

Legal Services* reviews applications and recommends whether to grant or deny leave

Panels decide to grant leave: decisions go on “B” list (grant), “C” list (delay), or “D” list (deny)

Justices sitting en banc in conference decide whether to accept panel recommendations to deny or grant leave

* In 1999, Legal Services was amalgamated with the Reports Branch to form the Law Branch, which now prepares the objective summaries and recommendations for leave applications.

steps back from these details to outline the steps of the agenda-setting process in Canada, which also provide the sequence of chapters in this book. The first step in the process, of course, is the decision to file an application for leave to appeal. Surprisingly little research focuses on this decision in either the United States or Canada (cf. Songer et al. 1995). Chapter 2 attempts to fill this gap by constructing a statistical model based on the notion that attorneys can improve their professional standing or reputation through the leave process. This “status-seeking” model claims that, setting aside whatever legal issues or other concerns may affect decisions to participate in the leave process, professional ambition is an independent factor prompting involvement in the agendasetting process. The applications are filed with the Process Clerk of the Registrar’s Office, which certifies whether the applications conform to Court rules. This is a ministerial function of little substantive import and is directly analogous to the procedure in the United States where the Clerk of the

flemming2.p65

14

01/04/2004, 2:24 PM

American Accounts and the Canadian Setting

Court reviews writs of certiorari to see that they meet the Court’s rules. The second step, however, is strikingly different from the American process and is a major institutional difference between the two courts. In Canada, leave applications meeting the format requirements are forwarded to the Law Branch (Registrar’s Legal Services before 1999) section. The section’s attorneys prepare “objective summaries” of the applications. Each summary provides a history of the case, outlines the facts, and lists the legal issues raised by the applicant. Since 1994-95, the summaries also include the staff attorneys’ recommendations on whether leave should be granted or denied. This stage is roughly similar in function to the “cert pool” in the United States, in which the clerks for all but one justice share the task of writing short assessments of the writs; these memoranda are then circulated among the justices and other clerks. The cert pool in essence is a division of labour intended to cope efficiently with the thousands of writs that flow into the US Supreme Court every year. The memos are also used by the chief justice to prepare a “discuss list” of writs he feels the Court should review when all nine justices meet in conference to vote on whether to grant or deny certiorari. This list helps reduce the workload of the justices, who can, however, add cases to the list should they choose to do so. Canada does things differently. It has decentralized the review process through the use of three-justice panels. The supervisor of Legal Services distributes the applications and summaries among the panels. The panels receive roughly equal numbers of applications with some exceptions. Applications from Quebec are sent to the justices from Quebec because of the province’s civil code, while those raising special issues are sometimes forwarded to a panel with a justice with expertise in the area. In the United States, all writs on the discuss list or those added to it by the associate justices are reviewed by the nine justices sitting en banc in conference. In Canada, the primary responsibility for deciding whether to grant or deny leave rests in the hands of the three panel justices. The panels do not meet face-to-face to make their decisions. Instead they vote by written memorandum (Crane and Brown 2002, 398). Although only two votes are needed for a decision, the published votes in the Supreme Court’s Bulletin of Proceedings are almost always unanimous; a mere thirty judgments coded for this study had a dissent. The panels notify the other justices of their decisions prior to conference, the next step in the process. Panels place applications on one of three lists after their review. If a panel wishes to defer an application because the Court is currently hearing a case raising similar issues, the application is placed on a “C” list

flemming2.p65

15

01/04/2004, 2:24 PM

15

16

Judicial Review and Agenda Setting

pending the outcome of the case. Applications approved for leave are placed on a “B” list for consideration at conference. This list is similar to the American discuss list and reflects the decentralized deliberations of the panels. If the panel decides to dismiss an applicant and deny leave, the panel notifies the other members of the Court of its decision. If a panel member dissents, the justice may place the application on the “D” list for reconsideration at conference. The “D” list is comparable to what was once called the “dead list” in the United States. If a justice who is not on the panel disagrees with a dismissal, the decision is deferred until the conference can reconsider it. A similar procedure takes place on the US Supreme Court when a justice asks that a certiorari petition not on the discuss list be reviewed by the justices when they meet in conference. As in the United States, conference votes in Canada are not public information. Unlike their American peers, however, Canadian justices do not keep records of conference votes, or at least they have not released them to the public when they retire or resign from the bench. This means that the panel and conference votes are conflated in this study; no case-level data are available that show in a systematic fashion whether votes in conference alter or reverse panel recommendations. Nor is it possible to determine how often voting divisions occur in conference even if the divisions do not change panel recommendations. Although conference discussions reportedly “can get quite spirited” (Crane and Brown 2002, 398), the common view is that the panel recommendations are the final word. In sum, the selection of appeals for judicial review in Canada moves through a process that differs markedly from the one in the United States. The Canadian process is more decentralized as leave decisions rest in the hands of the justices on the panels. This delegation of responsibility may lead to different standards of review or criteria for leave unless the justices share a common perspective of what cases should be granted access to the Court’s agenda. It is also possible the panels may grant leave at different rates. The conference may correct or modify panel decisions as other justices voice their opinions and urge panels to change their minds. Tournament of Appeals as a Metaphor for the Process

This study uses the notion of a tournament as a metaphor for the process of granting judicial review. The metaphor provides a way of fitting the various pieces of the empirical puzzle together under a single rubric. Tournaments pit numerous contestants against one another in a series

flemming2.p65

16

01/04/2004, 2:24 PM

American Accounts and the Canadian Setting

of competitive trials or rounds of play. Access to the Supreme Court’s agenda is limited; it is the prize for leave applicants and their attorneys in this particular tournament. The rules of tournaments, of course, dictate the nature of the contest and presumably the outcomes of the contest. And, of course, there are referees to enforce the rules and make sure they are followed. The tournament of appeals differs from the standard tournament, however, in that the “referees” in the leave tournament are the justices, who are not neutral enforcers of the rules since they decide based on their own interpretation of the rules whether the applicant or the respondent “wins” a round. Tournaments put contestants competing for prizes through a series of rounds of play. Repeated play of any kind over time may produce a hierarchy of players or teams. Each application for leave to appeal can be considered a separate round; typically there are roughly several hundred rounds of leave applications a year in Canada. The Supreme Court’s decision to grant or deny leave is a reward or penalty (depending on the party) that has both tangible and intangible payoffs for the lawyers. The rules of this tournament are uncertain and kept deliberately vague by the justices. When all is said and done, “public importance” is a term of art that conceals as much as it reveals. If attorneys have only vague intuitive notions of the rules the Court follows when picking lower court decisions for judicial review, repeated play by some attorneys may give them an advantage vis-à-vis the Court. Since the attorneys must make guesses about these rules and how they apply to their cases, the likelihood of mistakes when they draft their arguments for leave is probably higher because of their relative ignorance of what the justices want. As there is no a priori reason to expect that the lawyers’ guesses are not randomly distributed, the lawyers’ chances of winning may also be randomly distributed. The metaphor is intended to portray the dynamics of the leave process but it also goes beyond this purpose in that, like other tournaments, this one is expected to generate both patterns of outcomes and hierarchies of players. As the metaphor is adapted to the empirical realities of the leave process, other distinctions will emerge that will underscore the uncertain and contingent nature of this particular tournament for the players. Data Sources and Outline of the Book

A brief description of the empirical foundation of this book is in order at this point. This study relies on two major datasets. The first dataset drew information about the applications for leave for 1993-95 from the

flemming2.p65

17

01/04/2004, 2:24 PM

17

18

Judicial Review and Agenda Setting

Supreme Court of Canada Bulletin of Proceedings, which is the Court’s official record of its actions and decisions. Every application filed with the Court for this period was included; a total of slightly more than 1,250 applications with judgments as to the Court’s leave decisions are part of this dataset. Information from the Bulletin was supplemented with information from the factums or briefs filed by the applicant attorneys; the arguments in the factums were coded, plus the identities of the lawyers involved in the leave process, since the latter cannot be found in the Bulletin. The Supreme Court Reports, the official publication of the Court’s decisions in cases for which it grants judicial review, were combed for the outcomes of cases decided during 1975-92 plus the identities of the attorneys in these cases; the latter information was used to calculate the frequency with which attorneys argued the merits of cases before the Court. The second dataset was generated through a bilingual questionnaire mailed to the attorneys involved in the leave process during the study period. The purpose of the questionnaire was to collect information about the backgrounds and professional status of the attorneys. A total of 552 attorneys returned completed questionnaires. As the study unfolds, further details about the research design and data sources will be discussed. The following chapters follow the steps in the Canadian sequence of granting judicial review and are organized according to the three agendasetting accounts drawn from the American literature on the writ of certiorari. Chapter 2 concentrates on two concerns of the litigant-centred account. One is whether there is an identifiable “Supreme Court Bar” in Canada similar to the one McGuire found in the United States. The second issue, as mentioned earlier, is whether requests for judicial review reflect the professional ambitions of attorneys. Chapter 3 elaborates on the litigant-centred account by turning to the question of “who wins?” to see whether repeat players and litigants with superior status or resources are more likely to be granted leave than one-shot players or litigants with inferior status or capabilities. Chapter 4 takes up the jurisprudential account and develops a statistical model of the factors associated with the Supreme Court’s decisions to grant leave. It takes into account the decentralized nature of Canada’s process by looking at the individual panel decisions to determine the consistency of the criteria used by the panels. Chapter 5 develops a strategic account to gain an understanding of why leave to appeal decisions by the justices are almost always unanimous. The concluding chapter places these findings in a comparative and theoretical context.

flemming2.p65

18

01/04/2004, 2:24 PM

2 Tournaments and Stratification of Canada’s Supreme Court Bar

Introduction: A Lawyer-Centred Perspective on Appeals

Located in the leafy suburban northwest section of Washington, DC, near the District’s border with Maryland, is the boutique law firm of Goldstein and Howe.1 Founded in 1999, the firm practises exclusively in the US Supreme Court. Despite its Lilliputian size, Goldstein and Howe, during the early part of the 2002-2003 term, was lead counsel in three pending cases before the Supreme Court, counsel in seven petitions for certiorari and three amicus curiae briefs on the merits, and a consultant on two other petitions filed by other firms. The firm keeps close tabs on petitions for certiorari and how the Court’s calendar takes shape over the course of the term; it is almost as though the firm were reading the Dow Jones Industrial Averages or checking leading economic indicators to plot the direction of the stock market. The firm prepares and posts on its website informal, chatty reports about the cases orally argued that day before the justices and what cert petitions the Court granted at the most recent conference, and highlights what significant petitions will be reviewed at upcoming conferences, when cases are relisted, and how the timing and movement of cases in the lower courts may affect the Court’s calendar and when cases are scheduled. Across town, in downtown Washington, are the more traditional tournament players, namely, large firms with strong appellate practices. A few blocks from the White House, situated among the offices along the K Street Corridor where the Capitol’s vast army of lobbyists is encamped, are two of America’s larger law firms. Sidley Austin Brown & Wood, with over 1,400 lawyers in several offices in the United States and abroad, established a Supreme Court specialization within its appellate division in 1985, and since then its lawyers have briefed over seventy-five cases on the merits and argued over fifty. The head of the division clerked for a court of appeals judge and for Chief Justice Warren Burger of the

flemming2.p65

19

01/04/2004, 2:24 PM

20

Tournaments and Stratification of Canada’s Supreme Court Bar

Supreme Court, served in the Office of the Solicitor General, and has argued thirty-four cases before the Court. West of Farragut Square on K Street are the offices of Mayer, Brown, Rowe, and Maw, with 1,300 lawyers worldwide; its Washington office employs 115 lawyers. The appellate division is chockablock with former Supreme Court clerks, attorneys who once worked in the Office of the Solicitor General, as well as counsel attracted from the departments or regulatory agencies of the US government. Mayer, Brown’s current appellate attorneys have argued a total of over 180 Supreme Court cases. Their experiences and insights undoubtedly find their way into the “bible” for lawyers who expect to appear before the justices, the authoritative Supreme Court Practice, now in its eighth edition, which is edited by two senior Mayer, Brown partners.2 Attorneys from these three firms (others could easily be added) form an informal grouping of lawyers whose practices involve a significant amount of Supreme Court litigation and who routinely play leading roles in the American appeals tournament. Tony Mauro, a seasoned legal reporter, recently remarked that the Supreme Court was “more than ever taking on the feel of a club, where everyone knows everyone else and newcomers are tolerated but not necessarily welcomed.”3 Roughly ten years before Mauro’s observations, Kevin McGuire (1993) published a systematic study of this “Supreme Court Bar” in which he traced the contours of America’s most exclusive appellate bar. With survey data and archival information, McGuire plotted the topography of a legal elite, centred largely in Washington, DC, that formed the core of a professional community of repeat players with the legal skills and reputations needed to gain access to the Supreme Court. An ample number of studies have probed the American legal profession’s stratification. A contemporary classic is the work by Heinz and his colleagues, who found that the social structure of Chicago lawyers reflected the contrasting economic statuses of their clients as well as the ethnicity and political inclinations of the lawyers themselves (Heinz and Laumann 1982; Heinz and Laumann 1997; Heinz et al. 1998). Until McGuire’s book, however, no attention had been paid to how attorneys and firms coalesce around appellate courts, although there are studies with this focus dealing with the “courthouse communities” that spring up in urban trial courts (e.g., Flemming et al. 1992; Heinz and Manikas 1992). Comparable analyses of high court legal communities are especially rare. One possible and important exception is Epp’s (1998) structural and institutional analysis of the mobilization of lawyers in rights litigation in four countries, including Canada and the United States.

flemming2.p65

20

01/04/2004, 2:24 PM

Tournaments and Stratification of Canada’s Supreme Court Bar

McGuire’s investigation of the American Supreme Court Bar thus remains one of a kind. This chapter seeks to determine whether there is a cluster of attorneys centred on Canada’s Supreme Court that resembles what McGuire found in the United States. Is there any reason to think that a legal elite exists in Canada that is similar to and parallels in most respects what McGuire found in the United States? There are reasons to think so, but also reasons for doubt. For example, in downtown Toronto, the financial capital of Canada, the headquarters of the “Seven Sisters,” the country’s leading law firms, are all located in the heart of the city’s banking district. Within walking distance of the intersection of King and Bay Streets, which has given rise to the term “Bay Street lawyer,” these firms are large, have multiple offices, and represent major corporations and businesses. McCarthy Tetrault, the largest of the Seven Sisters, occupies the top floors of a building in the Toronto Dominion Centre, a glass and steel complex of buildings by the modernist architect Mies van der Rohe that affords the attorneys dramatic views of Lake Ontario and the Toronto Islands. These firms, like those in Washington, have strong political ties. John Turner, a former Liberal prime minister of Canada, is a member of Miller Thomson, for example. More recently, Mike Harris, Ontario’s Progressive Conservative premier for seven years, joined Goodmans, another leading firm, as a “senior business advisor” after retiring from political office. These firms have active appellate practices that attract non-business litigation. Lawyers from Torys, another of the Seven Sisters, have represented the Canadian Civil Liberties Union in several Canadian Charter of Rights and Freedoms cases before the Supreme Court, most recently as an intervener in the highly controversial case R. v. Sharpe, which involved child pornography. A boutique firm comparable to Goldstein and Howe in Washington is the Toronto firm of Eberts Symes Street & Corbett. This firm’s practice, however, is not as intently focused on Canada’s Supreme Court, although its members have been important players in Charter litigation before the Court. Mary Eberts, a feminist lawyer, a founder of the Women’s Legal Education and Action Fund (LEAF), and at one time a lawyer with Torys, has been actively involved as an intervener and advocate in major Charter cases centring on the equality rights of women (Simpson 1993). Her partner, Beth Symes, has represented the Canadian Abortion Rights Action League (CARAL) in a case before the Supreme Court that involved the medical incarceration of a pregnant Aboriginal woman as

flemming2.p65

21

01/04/2004, 2:24 PM

21

22

Tournaments and Stratification of Canada’s Supreme Court Bar

well as in other cases. David Corbett has been counsel in leading gay rights cases, such as Vriend v. Alberta. This firm clearly has a causeoriented practice and is a repeat player in civil rights and equality cases. The US Supreme Court Bar is located principally in Washington. One factor encouraging the formation of an Ottawa-based bar is that all parties to cases before the Canadian Supreme Court, even if they have counsel, must appoint a local Ottawa agent after leave to appeal is granted. Provincial and territorial attorneys general and ministers of justice also are represented by firms in Ottawa. The agents perform mostly administrative services although they also advise the parties and often conduct more substantive work, but their primary responsibility is as a liaison between the Court and the parties or their lawyers (Crane and Brown 2002, 241). The agents reduce the travel costs and lessen the communication problems that would arise if parties and their local attorneys from across Canada’s sprawling landmass had to be in Ottawa to handle directly the details of litigation. This Court-imposed rule creates a market for agents in Ottawa. Roughly thirty attorneys in twenty-one firms, including branch offices of the Seven Sisters, are currently licensed as agents. These Ottawa firms also have sections or divisions that specialize in litigation before the Supreme Court. For example, Lang Michener’s Ottawa office (the firm also has an office on Bay Street in Toronto) organized a “Supreme Court Group” of six attorneys who provide a wide array of services, including drafting leave applications, motions, and briefs as well as offering strategic advice to other attorneys. The head of this group was once the “executive legal officer” to the chief justice of the Court and has written extensively about the Court, the Charter, and Canada’s Constitution. He also is the lead author of Supreme Court of Canada Manual: Practice and Advocacy. One member of the group clerked for a Supreme Court justice. At another firm, Gowling Lafleur Henderson, the senior partner in the advocacy department has been a ministerial advisor to several federal departments in addition to his involvement with the Court. He and another senior member of the group are the authors of Supreme Court of Canada Practice, which was first published in 1991 and is now in its fifth edition.4 While the list of regulars before Canada’s Court can be extended, there are considerations unique to Canada that may prevent the formation of a Supreme Court Bar along American lines. For one thing, Ottawa for a very long time was not considered an attractive place in which to practise law. Shortly after the Supreme Court was established in 1875, there was talk of moving it to Toronto since Ottawa was scarcely more than a

flemming2.p65

22

01/04/2004, 2:24 PM

Tournaments and Stratification of Canada’s Supreme Court Bar

“backwoods frontier town” (Bushnell 1992, 114). The bill establishing the Supreme Court included the requirement that justices had to live in Ottawa or within forty kilometres of the capital. This requirement reflected the city’s lacklustre reputation as well as the fear that the justices would reflect the views of their home cities and provinces if they continued to return to them after concluding the Court’s business. In light of the tensions between Francophones and Anglophones, this was no small consideration. Indeed, the rivalry and struggle between Quebec and Ontario is one reason why Queen Victoria, as a compromise between Montreal and Toronto, picked the site on the Ottawa River for Canada’s capital. The residency rule nevertheless made it difficult to find good lawyers who were happy to give up the amenities and comforts of Quebec City, Montreal, or Toronto for a seat on the high court bench in Ottawa (Snell and Vaughan 1985). Ottawa’s attractiveness in 1875 was comparable to Washington’s when it became the capital of the United States nearly a hundred years earlier. With time the city’s professional and residential appeal has grown, especially since the emergence of the administrative state in Canada after the Second World War. Canada’s highly decentralized form of federalism also may deter the development of an Ottawa-focused Supreme Court Bar. The provinces are constitutionally empowered to act in social and economic matters either to the exclusion of the federal government or with substantial authority of their own, which they jealously and vigorously guard. To take one example, the provinces and not the federal government regulate securities markets in Canada, which heightens the importance of the provincial courts. The provincial appellate courts gain further significance because they hear substantially more of the country’s appellate caseload than Canada’s federal court system, which is much smaller and less important than in the United States. One last but not final consideration affecting the size and concentration of a Supreme Court bar in Ottawa are the differences in the volume or number of intervener briefs filed in Canada compared with amicus curiae briefs in the United States. The US Supreme Court allows amicus curiae briefs to be filed with cert petitions; this greatly expands the market for the services provided by the US Supreme Court Bar. In addition, the vast majority of cases heard on the merits by the Court are accompanied by amici briefs; a further boost to demand for the Bar’s attorneys. The Canadian Supreme Court does not allow interveners to file briefs before leave to appeal is granted. As a typical year may see 500 to 600 leave applications submitted to the Court, this prohibition cuts substantially into the need for the involvement of the Court’s repeat

flemming2.p65

23

01/04/2004, 2:24 PM

23

24

Tournaments and Stratification of Canada’s Supreme Court Bar

players. Moreover, while interveners often appear on one side or the other when cases are heard by the Court, they appear less frequently than amici in the United States. This further reduces the demand for legal services that a Supreme Court bar centred in Ottawa could provide. This difference between the two countries is another important reason not to expect that if Canada has a Supreme Court bar it will be based in Ottawa. Furthermore, in concert with the factors just mentioned, if a bar exists, it may be more dispersed and less centralized, which may weaken the forces that have stratified the Supreme Court Bar in the United States. The first aim of this chapter is to determine whether there is a cohesive bar made up of repeat player attorneys in Canada. The second aim of this chapter, however, is to construct a statistical model to test whether lawyers’ involvement in the agenda process of Canada’s Supreme Court leads to stratification. This lawyer-centred perspective of agenda setting in high courts is unusual as work on “deciding to decide” generally focuses on how high courts select cases for judicial review and not on the processes prior to this stage, when parties and their lawyers decide whether or not to appeal lower court decisions (cf. Barclay 1999; Songer et al. 1995). This chapter therefore lays the foundation for the next chapter, which deals directly with the impact of lawyers on the agenda process. Identifying the Tournament’s Players

Tournament rules usually specify who is eligible to play in the game; an appeals tournament is no different. For example, an attorney who wants to argue a case before the US Supreme Court must be a member of its Bar. Indeed, except for the rare pro se litigant, every party in a case before the Supreme Court must be represented by at least one member in good standing of the Bar. If the US Supreme Court tightly controlled admission to the Bar, keeping its membership small and exclusive, it in effect would automatically create a cadre of elite tournament players. Admission standards are minimal, however: an attorney must be free of any disciplinary sanctions and be admitted to practice before the highest court of a state, commonwealth, territory, or possession, or of the District of Columbia, for three years before asking for admission to the Supreme Court’s Bar (Stern et al. 1993, 735). These criteria are easily met and since 1970, when the Court allowed lawyers to submit their applications by mail, the Court annually admits roughly 5,000 attorneys to the Bar (Stern et al. 1993, 733). For most of the Bar’s members, the only benefit of membership is the certificate

flemming2.p65

24

01/04/2004, 2:24 PM

Tournaments and Stratification of Canada’s Supreme Court Bar

they receive from the Court, a certificate described in the authoritative Supreme Court Practice as “suitable for framing.” Furthermore, as this bible of Supreme Court litigation notes, the Court has remarked on at least one occasion that membership in the Bar is a “relatively uninformative fact” regarding an attorney’s professional skills and has warned that it is “at least bad taste” to suggest otherwise in advertisements (Stern et al. 1993, 733). The Canadian Supreme Court does not have a formal bar. This means that even the diaphanous boundary between lawyers based on whether or not they are members of the US Supreme Court Bar does not exist in Canada. As the authoritative Supreme Court of Canada Practice states, “Any member of the bar in Canada may appear as counsel in the Court” (Crane and Brown 2002, 62). The Supreme Court extensively revamped its rules recently but this rule remained unchanged. As long as lawyers who wish to appear before the Court are licensed by the provinces’ law societies, they may litigate in Canada’s highest court. Because the boundary of the US Supreme Court Bar is so porous, actual involvement in litigation before the Court becomes the crucial criterion for separating the active Bar from those lawyers who do little more with their Bar certificates than hang them on their office walls. McGuire (1993, 6) therefore defined the American Supreme Court Bar as including those lawyers admitted to practice before the Supreme Court but who also had participated in litigation before the Court, including the writ of certiorari stage. The latter criterion is extremely important since it means that McGuire included attorneys who participated in the certiorari process but who may not have argued the merits of cases. By doing so, he recognized the importance of the agenda-setting stage to the Court and to the lawyers, expanded the pool of eligible lawyers, and thus avoided drawing an arbitrarily tight boundary around the Court’s “real” Bar. This study followed McGuire’s lead and identified lawyers who actually participated in the leave to appeal process. Unlike McGuire, however, it excludes lawyers employed by the federal or provincial governments in Canada because their inclusion might effectively determine the outcome to the question being asked. An overview of the Canadian legal profession claims that lawyers in non-private or public settings are likely to have distinctive subcultures that set them apart from private practitioners, and that publicly employed lawyers are more likely to be specialists with skills suited to their work environment (Arthurs et al. 1987, 511-12). The nature of government litigation is such that public attorneys are very likely repeat players in their own right who form cohesive

flemming2.p65

25

01/04/2004, 2:24 PM

25

26

Tournaments and Stratification of Canada’s Supreme Court Bar

groups specific to and reflective of their governmental positions and agencies. For example, the premier repeat player lawyers in the United States are the roughly twenty-five attorneys who work in the Office of the Solicitor General and represent the federal government in all cases before the Supreme Court. To a lesser degree, the same thing can be said of the career attorneys working as legal counsel in the various departments or administrative agencies of the federal government. Propinquity, specialization, or relationship to the Supreme Court fosters litigation cultures or communities that differ from those that emerge from the working environments of private lawyers. The outcome of this analysis becomes less predictable by not including lawyers employed by the provincial or federal governments. The factums filed by the applicants requesting leave to appeal were used to identify the attorneys for both parties involved in the proceedings. Do the leave attorneys, who are being treated for the moment as Canada’s “Supreme Court bar,” contrast greatly with Canada’s “national bar” or private attorneys more generally? Information regarding the latter was drawn from the Canadian Law List, a comprehensive directory of lawyers and firms used for the mail survey of leave attorneys. The questionnaire used by McGuire was adopted and modified slightly for the Canadian context, including making it bilingual, as part of the effort to replicate the major American approaches to agenda setting.5 The questionnaires returned by the attorneys generated the information used in this chapter. Table 2.1 presents the provincial and urban locations of the national bar, the leave attorneys, and the attorneys who responded to the survey. (Comparisons with the latter group provide some measure of the representativeness of the respondents to the survey questionnaire.) In 1995, there were 51,237 non-governmental lawyers in Canada, according to the Canadian Law List 1996. Ontario, the country’s most populous province, had the largest proportion of lawyers, with 38.6 percent of the national bar. The second largest province, Quebec, ranked second with 18.6 percent of Canada’s private lawyers (excluding the province’s notaries). British Columbia and Alberta, next in order of population size, had the third and fourth largest proportions, of 15.2 percent and 10.2 percent, respectively. Alberta, with 5,201 lawyers in 1995, had more than three times the number of lawyers in Manitoba, the next most populous province, which had 1,573 attorneys, or 3.0 percent. Manitoba and the remaining provinces and territories accounted for 17.4 percent of the national bar. Despite population shifts and market changes, these proportions resemble those in 1981 (Arthurs et al. 1988, 166). It is worth noting that relative to their proportion of Canada’s

flemming2.p65

26

01/04/2004, 2:24 PM

Tournaments and Stratification of Canada’s Supreme Court Bar

Table 2.1 Comparisons of national bar, leave attorneys, and lawyers who responded to the Survey of Leave to Appeal Attorneys, 1993-95 National bar (%)

Leave attorneys (%)

Respondents (%)

38.6 18.6 15.2 10.2 17.4 100.0

27.8 25.4 16.3 10.6 19.9 100.0

25.2 25.6 15.7 11.6 21.9 100.0

51,237

1,113

540

City Ottawa Toronto Montreal Vancouver Others Total

3.1 24.3 9.3 9.4 53.9 100.0

5.7 18.4 18.0 12.3 45.6 100.0

3.5 18.6 17.0 11.1 49.8 100.0

Number

3,954

1,096

540

Province Ontario Quebec British Columbia Alberta Others Total Number

Note: Only non-governmental attorneys are included in this table. Approximately 20 percent of the sampled attorneys listed in the Canadian Law List (1996) were public or government lawyers; roughly 10 percent of the leave attorneys were public attorneys; and 6.5 percent of the sampled leave attorneys were non-private attorneys (universities, law schools, legal aid, etc.). The tally for the national bar of 51,237 excludes notaries in Quebec. Sources: Provincial locations of the “national bar” are from the Canadian Law List (1996) (Antoniadis 1996, A-7). The urban locations of the national bar come from a 10 percent sample of attorneys in the Canadian Law List. Comparable information for the leave attorneys comes from the Canada Leave to Appeal Database. Data for the respondent attorneys come from the Survey of Leave to Appeal Attorneys. .

population, the country’s three largest provinces, Ontario, Quebec, and British Columbia, have disproportionately more lawyers than the smaller provinces. The provincial locations of the lawyers involved in the leave process in 1993-95 differed from the national bar. In the case of Ontario, the difference was substantial. Lawyers practising in Ontario were less involved in the leave process than might be expected. Nearly an eleven-point difference (–10.8) separates the proportion of the national bar in Ontario and the proportion of lawyers from this province involved in leave to appeal applications. Quebec lawyers, in contrast, were more active than might be expected; the percentage difference is +6.8 compared with the province’s share of the national bar. There are slightly higher

flemming2.p65

27

01/04/2004, 2:24 PM

27

28

Tournaments and Stratification of Canada’s Supreme Court Bar

than expected proportions of leave attorneys from Alberta (+0.4), British Columbia (+1.1), and the other provinces and territories (+2.5). A notable distinction, at least when compared with the United States, is the heightened concentration of lawyers in Canada’s largest cities. McGuire (1993, 38) reported that 7.3 percent of the national bar in the United States and 8.3 percent of its Supreme Court Bar practised in New York. (These proportions include publicly employed and government attorneys in the United States, which inflates the percentages compared with those for Canada.) Toronto, Canada’s largest city, contains about 10 percent of the nation’s population but 24 percent of its lawyers (see also Arthurs et al. 1988, 126). The proportion of the national bar located in Montreal is also higher than in any large American city. Canada’s capital, Ottawa, has a smaller proportion of the national bar compared with Washington, DC, however. Lawyers located in Ottawa were nevertheless disproportionately involved in the leave process compared with their proportion of the national bar (there is a difference of 2.6 percentage points between the two proportions), although compared with the bars in other cities, Ottawa’s attorneys constituted a smaller proportion of the active agenda-setters. Lawyers in Montreal and in Vancouver participated more often in the leave process than might be expected given their shares of the national bar. For Montreal, there is a substantial difference of nearly 9 percentage points (+8.7) between the proportion of the national bar practising in the city and the proportion of the city’s attorneys involved in the leave process. This means that the proportion of leave attorneys with practices in Montreal was nearly double the city’s share of the national bar, much like Ottawa but on a larger scale. The difference in Vancouver was similar to that in Ottawa (+2.9), except that in Ottawa this difference means that their lawyers were twice as likely to file leave applications while in Vancouver they were only a quarter again as likely. Toronto lawyers were underrepresented among the leave attorneys, with a –5.9 percentage point difference between Toronto’s share of the national bar and the proportion of attorneys from Toronto who participated in the leave process. Table 2.1 provides mixed evidence for concluding that lawyers who might be part of a Supreme Court bar in Canada are geographically concentrated in particular provinces or major cities. Attorneys who participated in the leave to appeal process came for the most part from provinces and major cities in proportions that roughly resembled Canada’s national bar. One notable exception is that lawyers from Ontario, the country’s wealthiest and largest province, and Toronto, the nation’s

flemming2.p65

28

01/04/2004, 2:24 PM

Tournaments and Stratification of Canada’s Supreme Court Bar

business centre, did not participate in the leave process as often as might be expected. Lawyers in Ottawa, Vancouver, and Montreal, though, were disproportionately more active in the leave process relative to their share of the national bar. The geography of appellate litigation in Canada, then, does not resemble the pattern McGuire found the United States.6 Rounds of Play and Stratification of Canada’s Bar

A relatively small number of attorneys make up the core of the US Supreme Court Bar. Experienced, with many cases under their belts, these attorneys dominate the Bar and know one another quite well. The American appeals tournament is more often a jousting match between regulars than a coincidental, accidental pairing of one-time players. In this clubby setting, Justice Antonin Scalia is reported to have interrupted a lawyer during oral argument after he referred to his opposing counsel as “my adversary.” The justice corrected the lawyer by pointing out that while the attorneys’ clients may be adversaries, the opposing attorney was “not your adversary ... He’s your friend.”7 A tournament in which each round is a contest between two amateurs who have never played the game before and will not play it again will by definition not produce a hierarchy or stable social structure among the attorneys who play the game regularly. Two contrasting tournaments are conceivable. One is that the Canadian bar could be composed of first-time or one-shot lawyers for whom the leave to appeal application is their first and only involvement with the Supreme Court. The composition of the bar over time would be inherently unstable and continually changing; its membership would be no more than a transitory, temporary artifact of fleeting participation in the process. The other alternative, of course, is the American case, in which the bar has a stable membership and is made up of lawyers with considerable experience in both the agenda-setting process and the merits stage; membership in this instance would have genuine meaning and consequences. The distribution of prior experience among attorneys needs investigation in order to determine where Canada’s bar falls on this continuum. Three categories of prior experience were created. The first category includes lawyers with no prior litigation experience before the Supreme Court, while the second includes attorneys with experience in the leave process but no involvement at the merits stage. The third category consists of lawyers with both leave and merits experience. Table 2.2 presents the data for these categories. Nearly 30 percent of the attorneys had no prior litigation experience before Canada’s Supreme Court. A slightly smaller proportion (28.6

flemming2.p65

29

01/04/2004, 2:24 PM

29

30

Tournaments and Stratification of Canada’s Supreme Court Bar

Table 2.2 Lawyers’ litigation experience before Canada’s Supreme Court , 1993-95 Supreme Court experience None (%)

Leave only (%)

Merits and leave (%)

29.5 (n = 163)

28.6 (n = 158)

41.8 (n = 231)

Source: Survey of Leave to Appeal Attorneys.

percent) had participated in the leave process but not at the merits stage. Roughly 42 percent of the lawyers had experience in both phases of Supreme Court litigation. The total for the latter two categories is 70.4 percent, which is slightly less than the comparable proportion of about 75 percent that McGuire (1993, 109) reports for the American Supreme Court Bar. It is important to note, however, that the category of most experienced lawyers, those with both leave and merits experience, is twice the proportion McGuire found for private attorneys in the United States. The upshot is that Canadian lawyers are much more likely to have won a previous round in the leave tournament than US attorneys in the certiorari game. What this means is that the merits stage in Canada is more accessible than in the United States. Between 1970 and 1990, Canada’s Supreme Court granted leave to 25-35 percent of the applications submitted to it, and since 1990, as the previous chapter noted, the rate has hovered around 15 percent. In contrast, the US Supreme Court’s acceptance rate in “paid” or predominantly civil cases fell from 10 percent to 3-4 percent. The rate for “unpaid” criminal cases, which account for the vast majority of petitions, is dramatically lower at between 0.002 and 0.004 percent. Not only have acceptance rates in the United States declined but the number of cases that are heard has fallen dramatically. Signed opinions by the Court, for instance, slumped from 146 in 1985 to 75 ten years later (Epstein et al. 2003, 76). Canada’s higher acceptance rate increases the chances of winning a round in the tournament, relaxes the boundaries of the bar, and makes it less exclusive than the American Bar. While Canada’s bar is stratified on the basis of attorney experience, Table 2.2 shows that it is neither predominantly composed of first-time players nor dominated by repeat players. It is important to know, however, how experienced the repeat players may be and whether the bar’s stratification is associated with other forms of differentiation among the lawyers. Are experienced attorneys most likely to work in large firms,

flemming2.p65

30

01/04/2004, 2:24 PM

Tournaments and Stratification of Canada’s Supreme Court Bar

Table 2.3 Attorney characteristics and Supreme Court experience, 1993-95 Supreme Court experience Attorney characteristic Prior experience Number of leave applications (mean) Number of cases on merits (mean) Professional reputation Nominated for expertise (%) Queen’s Counsel (%) Professional tenure Years of legal practice (mean) Type of practice Member of firm (%) Firm size (mean number of lawyers) Number of specializations (mean) Appellate specialty (%) Rights practice (%) Current leave to appeal client Business (%)

None Leave only Merits and leave – –

1.2 –

4.8 2.9

‡

2.5 4.9

9.0 6.5

27.6 33.0

‡

15.1

18.8

24.9

‡

68.7 42.5 2.3 5.5 24.5

83.5 48.2 2.5 15.8 29.1

82.3 (n.s.) 51.0 (n.s.) 2.8 * 24.8 ‡ 40.7 †

25.7

18.5

27.4 (n.s.)

‡

* p < .05, † p < .01, ‡ p < .001. Sources: Survey of Leave to Appeal Attorneys; Canada Leave to Appeal Database.

the Seven Sisters firms, like McCarthy Tetrault? If so, then the Court’s agenda would be more accessible to lawyers who can tap the plentiful resources such firms make available to them when preparing cases, and the tournament of appeals would be biased against solo practitioners or members of small firms. Table 2.3 compares the experience levels of the attorneys according to various characteristics and indicates whether differences between the groups are statistically significant. The first two rows in Table 2.3 show that lawyers with merits experience were typically four times as likely to be involved in the leave process than attorneys without merits experience. On the average, these experienced lawyers were also involved in 2.9 cases at the merits stage, which is higher than the 2.3 to 2.6 cases reported by experienced firm and corporate counsel in the United States (McGuire 1993, 167). This higher rate is consistent with Canada’s more accessible plenary docket. In other words, not only are experienced attorneys more likely to win a round in the leave tournament and thus argue the merits of cases more often than their less experienced peers in Canada, but they also do so more frequently than experienced attorneys in the United States. Attorneys with both leave and merits experience practised law for an average of nearly 25 years compared with about 19 years for the group

flemming2.p65

31

01/04/2004, 2:24 PM

31

32

Tournaments and Stratification of Canada’s Supreme Court Bar

with prior leave involvement but no merits experience and 15 years for those attorneys with no prior experience in the Supreme Court. Lawyers with merits or leave experience worked in firms more often than attorneys without either experience, who tended to be solo practitioners. While the typical size of the firm varied with the average level of experience, the differences were not statistically significant. The attorneys reported significant differences, however, in the number of legal areas in which they specialized, with the mean number of specialties rising with prior experience; in other words, experienced lawyers were more often generalists than those with less experience. Experienced attorneys, moreover, tended to have an appellate practice or a human or civil rights specialty in addition to their other specialties. There were no statistically significant differences between the groups with regard to whether they represented individuals or businesses in the leave process. Finally, stratification in terms of professional prestige and giving advice paralleled differences in experience. Lawyers with merits experience were more likely to be Queen’s Counsel, although this relationship may not be quite what it seems. Queen’s Counsel (or King’s Counsel) was imported from Great Britain when Canada was one of its colonies. The award has become largely honorific, and the appointments have not lacked for controversy. Perhaps the award’s nadir was reached in 1896, when its value as patronage became evident during the waning days of the Conservative government of Prime Minister Sir Charles Tupper, who seized the chance to nominate 173 lawyers for the award.8 In 1993, the Government of Canada ceased making federal Queen’s Counsel appointments, which followed Quebec’s abolition of the practice in 1976 and Ontario’s discontinuance in 1985. Manitoba more recently cocked a critical eye at the practice. The upshot is that standards for the award are not onerously stringent and that being a Queen’s Counsel often reflects the recipients’ political ties (Stager and Arthurs 1990, 201-4). The following table casts a different perspective on the bar’s stratification. As it indicates, asking for help or counsel followed the lines of status within the bar. Lawyers with both leave and merits experience reported that they counselled other attorneys more often than attorneys with less experience. Experienced attorneys offered advice on leave to appeal applications, the preparation of briefs on the merits, and oral arguments, as well as help to lawyers who received permission from the Supreme Court to act as interveners in cases. It appears that the market for legal advice both generates and reflects the stratification of attorneys according to their Supreme Court experience.

flemming2.p65

32

01/04/2004, 2:24 PM

Tournaments and Stratification of Canada’s Supreme Court Bar

Table 2.4 Advice giving and prior Supreme Court experience, 1993-95 Supreme Court Experience Type of advice given Leave application Prepare brief on merits Oral argument Help attorney for an intervener

None (%)

Leave only (%)

Merits and leave (%)

Phi

22.7 14.7 10.4 3.7

38.6 33.5 22.8 7.6

44.6 39.0 28.6 15.6

.19* .22* .18* .17*

* p < .001. Source: Survey of Leave to Appeal Attorneys.

In sum, the Canadian bar is stratified according to the distribution of experience and its associated characteristics. Attorneys who previously argued the merits of cases had longer professional tenures, had general practices that included appellate or rights work, and had attained visible stature in the profession. Lawyers for whom the current leave application was their first involvement with the Supreme Court had shorter tenures, were more specialized, and were professionally less visible. For the most part, lawyers with only prior leave experience occupied a middle ground between these two groups. The leave tournament, in other words, is played by three groups of attorneys that differ systematically with regard to status-related factors. Status Seeking as an Incentive for Tournament Play

Current perspectives of agenda setting in high courts focus on the criteria used by the courts or on the voting behaviour of the justices when selecting cases for judicial review. Little attention has been paid to determining the factors that affect lawyers’ involvement in the process. The players in the tournament seemingly appear out of the blue to represent their clients’ interests as though there were no professional stakes or personal concerns prompting their involvement. If, as McGuire concluded, attorneys function as gatekeepers to courts, it is important to know more about what influences lawyers’ participation in the agendasetting process. More specifically, in light of previous research on the stratification of lawyers, it makes sense to consider how professional status affects their decisions to request judicial review. The following discussion expands the traditional focus to consider these incentives by constructing a status-seeking model of leave activity in Canada. The model’s basic assumption is that involvement in the leave process reflects in part the lawyers’ stakes in establishing a reputation and gaining visibility

flemming2.p65

33

01/04/2004, 2:24 PM

33

34

Tournaments and Stratification of Canada’s Supreme Court Bar

within the litigating community as a way to enhance their professional reputations. The model’s dependent variable is the lawyers’ self-reported number of leave to appeal applications in which they have been involved over the course of their professional careers. The reliability of this information was checked against a roster of attorneys identified in the Supreme Court Reports that was used to determine the number of merits appearances made by lawyers during 1975-92. The correlation between these appearances and the lawyers’ reported merits appearances is quite high, although a handful of respondents were removed for apparent overreporting. In other words, the self-reported number of leave applications generates reasonably reliable data for this analysis. Table 2.5 lists and describes the variables in the model. Lawyers seeking higher status in their profession can promote their aim by filing leave to appeal applications and having their appeals Table 2.5 Variable definitions and characteristics Variable Leave activity by lawyer Number of times as applicant or respondent Reputation Nominations for expertise Queen’s Counsel Experience Supreme Court clerk Former federal attorney Success on merits Resources Number of attorneys in firm Years as lawyer Advice giving General Intervener Legal specialty Appellate Rights Location of practice Ottawa Toronto Montreal Vancouver

Value or range

Mean

N

0-35

2.20

541

0-82 1, 0

.72 .17

546 544

1, 0 1, 0 1, 0

.02 .06 .32

552 552 550

1-550 1-48

47.70 20.10

548 477

1, 0 1, 0

.33 .10

493 498

1, 0 1, 0

.16 .33

552 552

1, 0 1, 0 1, 0 1, 0

.04 .19 .17 .11

547 547 547 540

Source: Survey of Leave to Appeal Attorneys.

flemming2.p65

34

01/04/2004, 2:24 PM

Tournaments and Stratification of Canada’s Supreme Court Bar

successfully reviewed by the Supreme Court. Appearances before the Supreme Court raise the professional visibility of attorneys, which also attracts clients.9 For example, lawyers with both leave and merits experience were significantly more likely to cite their professional reputations as an explanation for why they filed their current leave application than attorneys with only prior leave experience or none at all. And, of course, as the number of clients grows, so do opportunities for appellate work and further participation in the leave tournament. The lawyers’ professional stature was measured in two ways. First, the survey questionnaire asked attorneys to identify lawyers whom they considered to be particularly skillful Supreme Court advocates plus those to whom they would turn for advice. The lawyers received scores according to the number of times they were singled out for their skill and whether they would be asked for advice. The two tallies were combined to create a numerical measure of an attorney’s professional reputation. The second indicator was whether an attorney was a Queen’s Counsel (1 = yes; 0 = no). Even if this award is honorific, it reflects a lawyer’s visibility within the legal profession. Prior success on the merits undoubtedly encourages further leave activity. Lawyers who succeed in the Supreme Court reap more than the immediate benefits of enhancing their professional stature. They also gain a sense of the ins-and-outs of the litigation process that bolsters their confidence and sustains the ambition for future work before the Court. Thus, lawyers who reported victories on the merits are expected to seek further involvement in the leave process. The measure is a dummy variable indicating whether or not a lawyer claimed one or more successes on the merits in the Supreme Court (1 = yes; 0 = no). In the United States, lawyers who have clerked for the Supreme Court or worked in the Office of the Solicitor General are especially active before the Court (McGuire 1993, 167). For example, most members of Sidley Austin Brown & Wood’s appellate division in Washington fit this pattern. Canadian attorneys with similar kinds of experiences might be similar to their American counterparts and be more involved in the leave to appeal process than other lawyers. Experience in these positions, especially clerking, generally occurs early in a career and gives young lawyers the chance to learn about the Court, valuable lessons that encourage participation in the appeals process after they enter private practice. Canada does not have, however, a direct equivalent of the US Office of the Solicitor General; the closest parallel might be the Litigation Committee of the Department of Justice, composed of deputy ministers and

flemming2.p65

35

01/04/2004, 2:24 PM

35

36

Tournaments and Stratification of Canada’s Supreme Court Bar

counsel for the department’s various divisions. The committee reviews cases for possible appeal to the Supreme Court and appoints counsel to represent the government, but it lacks its own staff of lawyers. There are senior lawyers in Justice who specialize in Supreme Court litigation, but there is no designated section focusing exclusively on the Supreme Court. Other federal departments have staff counsel that litigate in Canada’s courts, including the Solicitor General’s Department, which has law enforcement responsibilities, including the Royal Canadian Mounted Police and intelligence and security functions, and operates the country’s prisons. All this being said, however, no office in Canada’s government has the same function and characteristics of the US Solicitor General. Two dummy variables provide indicators of an attorney’s experience as a clerk or government attorney, with 1 = yes and 0 = no depending on whether the individual had been a clerk or attorney in the federal government. The latter indicator is broader and less specific as a reflection of the Canadian context. Expertise and resources facilitate litigation efforts. Professional tenure or the number of years the attorney has practised law offer a crude indicator of an attorney’s expertise and legal acumen. Moreover, as time passes, other things being equal, attorneys have more opportunities to participate in appeals. Tenure, therefore, is also a very important control variable in the model. As a control variable, it provides assurance that if the other variables in the model are found to be statistically significant, the results cannot be dismissed as merely the artifacts of time or the lengths of careers. Resources also matter. Attorneys in large firms presumably have advantages over solo practitioners or lawyers in smaller firms because of the depth of expertise attorneys in large firms can tap as they prepare their cases. The model’s indicator for attorney resources uses the actual number of lawyers in the lawyer’s firm in 1995-96, with a score of 1 assigned to solo practitioners. The priorities of Canada’s Supreme Court may share a varying, reciprocal relationship with attorneys whose legal specialties make them more or less eligible to play in the tournament as a reflection of the kinds of appeals the justices are most keen to hear. An attorney’s opportunities for appellate work depend in part on the priorities of the high court as it sets its agenda. Of course, they also depend on whether the lawyers have strong appellate practices. Lawyers with appellate practices, regardless of their area of legal expertise, should have more chances to argue cases before the Supreme Court than attorneys primarily involved in trial work. An attorney’s area or areas of legal expertise may matter more, though, because the Court’s agenda developed a strong public law orientation

flemming2.p65

36

01/04/2004, 2:24 PM

Tournaments and Stratification of Canada’s Supreme Court Bar

after the justices gained discretionary control over their docket and after the entrenchment of the Charter of Rights and Freedoms. Lawyers whose work deals with equality claims under the Charter, human rights, or criminal law could be more involved in the leave process than attorneys with private law specialties like torts, property, or insurance. Indeed, attorneys in large corporate law firms may be disadvantaged under this regime as most of their appellate work may not correspond to the Court’s public law priorities. Two dummy variables indicate whether attorneys reported that they had an appellate practice (1 = yes; 0 = no) and whether they had a “rights” specialty (1 = yes; 0 = no). Attorneys turn to one another for advice about cases, and patterns of advice asking and giving may reflect the esteem in which lawyers hold their colleagues. They may ask to be briefed by other lawyers about preparing leave applications and how to craft convincing arguments, or they may seek counsel regarding oral arguments. Every facet or stage in the litigation process may prompt the need to talk to more experienced or highly esteemed colleagues. Two measures of advice giving will be used in the model. One measure combines the responses of attorneys who reported that they gave advice on leave applications, the preparation of briefs, and oral arguments because the three indicators were highly correlated with one another. The indicator is simply a dummy variable indicating whether or not the attorney offered to another lawyer advice on any of these matters (1 = yes; 0 = no). As it happens, whether an attorney helped another lawyer who was representing an intervener is not highly related to these indicators of advice giving. A separate dummy indicator for this form of advice giving is used (1 = yes; 0 = no). McGuire found that the most active members of the Supreme Court Bar in the United States were concentrated in the District of Columbia. For these litigators, close proximity to the Supreme Court, other federal courts, and government agencies, combined with reasonably quick transportation to New York City and other East Coast cities plus a network of former and current colleagues, made Washington, DC, an advantageous location. In the model, dummy variables (1 = yes; 0 = no) indicate whether lawyers were located in Ottawa, Toronto, Montreal, or Vancouver. If location affects involvement in the agenda-setting process, these variables should pick up this influence on the lawyers’ behaviour. Table 2.6 presents the results for the regression model specified in the preceding discussion. The model is statistically significant and has an adjusted R2 of .304. The significance tests for the individual variables are one-tailed because the hypotheses are directional; most of the signs for the variables are in the expected direction.

flemming2.p65

37

01/04/2004, 2:24 PM

37

38

Tournaments and Stratification of Canada’s Supreme Court Bar

Table 2.6 A professional status model of leave activity Independent variable Reputation Nomination for expertise Queen’s Counsel Experience Supreme Court clerk Former federal attorney Success on merits Resources Size of firm Professional tenure Advice giving General Intervener Legal specialty Appellate Rights Location Ottawa Toronto Montreal Vancouver Constant Number of observations F (15,325) R2 Adj R2

b

SE

Beta

t

.008 .853

.034 .456

.107 .094

2.328 ‡ 1.871 †

–.679 .397 1.991

1.114 .711 .375

–.026 .025 .263

–.610 .558 5.310 §

.008 .002

.002 .021

–.022 .173

–.490 3.332 §

–.278 .953

.337 .514

–.037 .084

–.823 1.854 †

1.518 .966

.423 .347

.162 .127

3.588 § 2.782 §

.604 .303 –.169 .637

.862 .427 .487 .478

.033 .033 –.016 .060

.702 .710 –.346 1.333 *

–.929

.460

–2.021

398 12.575§ .330 .304

* p < .10 (1- tailed), † p < .05 (1- tailed), ‡ p < .01 (1- tailed), § p < .005 (1- tailed).

The model suggests that involvement in the tournament of appeals depends on the professional stature or reputation of the attorneys; lawyers perceived as skillful by their peers played more rounds than lawyers lacking this reputation for expertise. Moreover, being Queen’s Counsel was a positive factor in promoting leave activity. Especially striking is the impact of merits success on leave activity; reputation matters but so do tangible results. Attorneys who can point to victories in the Supreme Court participate more often in the leave to appeal process, and they have solid incentives to continue this activity. Clerking for a Supreme Court justice or experience as a government attorney apparently has little impact on leave activity. A clerkship may not be related to leave activity because the Court did not hire law clerks

flemming2.p65

38

01/04/2004, 2:24 PM

Tournaments and Stratification of Canada’s Supreme Court Bar

until 1968 (Snell and Vaughan 1985, 223), when each justice was allotted one clerk. In 1985, the number of clerks was increased to two per justice. The Supreme Court now employs twenty-seven clerks for one year, with three for each of the nine justices (Sossin 1996). The prestige of clerking has probably risen over the years as the Court’s prominence within Canada has grown. In the early years, before the mid-1970s or early 1980s, clerking may not have carried the cachet it does now. The experience may also have been less useful, as the character of the Court’s docket changed from fairly routine private law questions that the Court was forced to hear to more complicated public law questions that created new laws. Moreover, law clerks in Canada come to the Court straight out of law school; they do not have the seasoning that American clerks gain in the lower appellate courts and, more recently, from a year or two in a major law firm before they clerk for a Supreme Court justice. Finally, only recently has a sense of self-identity begun to emerge among the former clerks; the first reunion of Canadian Supreme Court clerks was held in 1994 (Sossin 1996). The Department of Justice and other government agencies also do not serve as a training ground for future Supreme Court litigators, nor do they have the same close, reciprocal relationship with Canada’s Supreme Court that the Office of the Solicitor General has with the United States Supreme Court (Salokar 1992). Moreover, the Supreme Court’s attractiveness as an arena where young lawyers seek to develop their careers began to increase only during the past twenty years with the Court’s emergence as a major policy maker, and may have suffered in the past compared with the provincial courts in Ontario or other provinces. The upshot is that the prominence of former law clerks and former Assistant Solicitors General in the US Supreme Court Bar reflects the institutional evolution of the Supreme Court and the professionalization of litigation in this Court. These developments have only recently begun in Canada. The length of time lawyers have practised law is significantly and positively related to leave involvement, either because of the maturation of their careers or because lawyers with more years of practice have more opportunities for involvement in leave applications. It bears repeating that this obvious relationship functions as a control for the other findings; significant results for other variables after years of professional experience are taken into account mean that they have an impact on the leave activity of attorneys that is separate from the length of attorney careers. In other words, the model taps status-seeking activity by the attorneys through the leave tournament. Participation in the tournament

flemming2.p65

39

01/04/2004, 2:24 PM

39

40

Tournaments and Stratification of Canada’s Supreme Court Bar

is not simply a function of lengthy careers; the accretion or accumulation of tournament play is not simply a product of lengthy careers. Firm size, a measure of resources, did not foster leave activity. Although inexperienced lawyers were likely to be solo practitioners, Table 2.6 shows no statistically significant positive relationship between firm size and leave activity. It is clear, though, that attorneys with an appellate practice or rights specialty participate more often in leave to appeal applications than lawyers with practices in other specialties. With respect to rights specialties, the emergence of Canada’s Court as a forum for human rights and Charter controversies encourages attorneys who specialize in these areas to participate more frequently in the leave process. At the same time, the Court’s public law emphasis skews the market in favour of lawyers with practices that match the Court’s priorities and encourages these attorneys to enter the leave tournament. Lawyers who counsel attorneys who represent interveners in cases were more often involved in leave activity than others. It may be that giving advice to interveners is linked to leave activity because of the special nature of the intervention process in Canada. Parties wishing to intervene are granted access only to the merits stage, and despite increasing involvement, interveners are still relatively infrequent participants compared with the nearly universal appearance of amici curiae in litigation before the United States Supreme Court. Unlike other forms of advice giving, then, providing counsel on applications to intervene is not commonplace. Nor does the Court treat the applications as mere formalities (Crane and Brown 2002). Attorneys who are asked for advice by interveners may have special skills that are linked to greater leave activity. The model’s results in Table 2.6 fail to support the hypothesis that location is a statistically significant variable in shaping the leave activity of attorneys in Canada, with the possible exception of Vancouver. The dummy variable for Vancouver was significant at the .10 level and positively related to the number of leave applications reported by attorneys in this city. A practice located in Ottawa, however, unlike Washington, DC, in the United States, does not link a lawyer to the Supreme Court’s agenda-setting process. Geography at most only marginally affects the incentive to enter the leave tournament as a way of seeking status within the Canadian legal profession. Conclusion: Status Seeking as Warming Up for Tournament Play

McGuire concluded that the Supreme Court Bar in the US has all the earmarks of a discrete professional community. He limned the contours

flemming2.p65

40

01/04/2004, 2:24 PM

Tournaments and Stratification of Canada’s Supreme Court Bar

of a litigation community that is both cohesive and exclusive. Many of its members have clerked for the justices of the Supreme Court or worked in the Office of the Solicitor General. They interact professionally, often in the Supreme Court, and many have practices in Washington, DC. Their credentials and reputations set them apart from other lawyers, who in turn hold them in very high esteem. The Bar’s significance for understanding the dynamics of litigation in high courts is McGuire’s finding that Bar members play an important role in setting the US Supreme Court’s agenda. Again, as McGuire (1993, 202) put it, they act as “gatekeepers to the Court.” Does Canada have a Supreme Court bar? Canadian lawyers are concentrated in the country’s largest cities, a concentration that promotes professional ties and interactions among them. Mandatory membership in the provincial law societies fosters these relationships, as well as membership in the provincial branches of the Canadian Bar Association or more specialized groups such as the Criminal Lawyers’ Association. In contrast to the United States, though, attorneys in Ottawa or Toronto, respectively the political and financial capitals of Canada, are not more likely to be involved in the tournament of appeals than lawyers from other places in Canada. No small group of lawyers based in Ottawa funnels leave applications to Canada’s Supreme Court in the same way that the Washington-based attorneys do with regard to petitions for certiorari in the United States. It should also be remembered that the proportion of attorneys with merits experience in Canada is greater than in the United States, which suggests a less exclusive elite, a larger inner circle than the one McGuire found in Washington. The inner circle of Canadian lawyers with merits experience, nevertheless, have practised law longer, occupy more prestigious niches within their profession, are less specialized, and often include appellate or rights practices among their specialties. These seasoned lawyers, moreover, have both the incentives and opportunities to maintain their professional stature by involving themselves in the leave to appeal process, as the regression model of leave activity suggests. In one sense, the model’s results are consistent with a theoretical story that follows what might be the normal sorting process that occurs in professional markets. It is tempting to view this process in terms of a professional life cycle in which tenure or experience is the primary factor; like an escalator, their number of years in the profession lifts attorneys to a higher status as a simple function of professional aging. Recall, though, that many of the status-related variables attained statistical significance with tenure as a control variable. Time matters, but it is not

flemming2.p65

41

01/04/2004, 2:24 PM

41

42

Tournaments and Stratification of Canada’s Supreme Court Bar

the only factor. Litigation communities become stratified for various other reasons, and all combine to affect the agenda-setting activity of the lawyers in these communities. This is the first attempt to weigh these influences on the agenda-setting process. The status-seeking model suggests that lawyers are motivated to participate in the leave process in order to improve their professional standing. The question now turns to whether stratification among attorneys affects the outcome of the tournament for access to the Supreme Court’s plenary agenda.

flemming2.p65

42

01/04/2004, 2:24 PM

3 Litigants, Lawyers, and the Tournament of Appeals

Introduction: A Litigant-Centred Account of Agenda Setting

A tournament that routinely concludes with victories for those with superior resources may be fine for sporting events but it raises serious questions about fairness and bias when it involves the law. The evolution of law and the priorities placed on socio-legal issues in Canada reflect the Supreme Court’s decisions to selectively review lower court cases. If cases and their issues are placed on its plenary docket through a process that is shaped or influenced by the political and economic inequalities of Canadian society, then those parties in positions of power and privilege will have a hand in directing the path of the law. Parties with strategic interests and the resources to achieve them will seek out attorneys who can improve the chances that their cases will be granted leave to appeal or, alternatively, stymie their opponents’ efforts to get the court to hear the appeals. The possible impact of the parties’ superior status or resources and the role of experienced, skilled attorneys in affecting the outcomes of litigation raises critical concerns about the Supreme Court’s prominence in shaping public policies. Chapter 2 showed that the attorneys who filed leave applications in Canada are stratified in terms of experience and other factors. In the United States, McGuire (1993) claims that the elite Bar of Washingtonbased attorneys has a significant impact on the selection of cases for judicial review as these highly regarded and experienced lawyers have an edge over neophyte lawyers. Attorneys with extensive litigation backgrounds have the credibility needed to overcome the skepticism of justices and clerks who are pressed for time as they sort through thousands of certiorari petitions. As one lawyer told McGuire, “Hiring a lawyer at the cert stage who has a reputation at the Supreme Court for playing by the Court’s rules is one of the most important things a client can do in

flemming2.p65

43

01/04/2004, 2:24 PM

44

Litigants, Lawyers, and the Tournament of Appeals

terms of getting attention paid to his cert petition” (McGuire 1993, 184). McGuire quotes another attorney as saying, “I’ve often wondered whether lawyers make a difference ... Well, I’ll put it this way: I can’t imagine a brief signed by Rex Lee or Erwin Griswold or Larry Tribe that would ever make the ‘dead’ [rejected certiorari] list. I just think they would say, ‘Wait a minute. We’ve got to take a look at it’” (McGuire 1993, 180).1 One of the cornerstones of the litigant-centred account is Marc Galanter’s exposition (1974) of “party capability theory,” which ground the conceptual prism for McGuire’s research. Galanter’s analysis of how the “haves” and repeat players are more likely to win their litigation battles than the “have-nots” and one-shotters is one of the most frequently cited articles in the law and social science literature (Grossman et al. 1999). Nevertheless, Galanter’s argument has received mixed empirical confirmation with regard to the US Supreme Court. Research regularly shows that the Office of the Solicitor General, which represents the federal government before the Court, regularly wins on the merits (e.g., Salokar 1992). However, studies of civil liberties cases point to complexities that Galanter’s analysis ignores. For several decades prior to 1970, have-nots or underdogs more often than not won their cases against governments (Ulmer 1985). After 1970, this pattern was reversed (Sheehan 1992). A persuasive explanation for the change appears to be shifts in the Court’s ideological makeup and not changes in the asymmetries of resources between litigants (Sheehan et al. 1992). Studies of the Supreme Court’s certiorari or agenda-setting decisions reveal a similarly mixed pattern; the haves are not invariably winners when they seek judicial review, but neither are they losers. During the 1947-57 terms, the Court routinely favoured the US government when granting certiorari. During this same time, though, the Court was also more sympathetic to labour unions, the putative underdogs, than to corporations, the unions’ upperdog opponents (Provine 1980). A comparison of the liberal Warren Court and the more conservative Burger Court confirms this pattern for the Warren years but then finds that upperdog parties fared better during the Burger era (Ulmer 1981). A tracing of the Court’s certiorari decisions in obscenity cases for the period 1955-87 discovered that most businesses and state and local governments failed to gain access regularly to the Court’s plenary docket. More to the point, success depended on the policy views of the Court; resources or status per se, contrary to Galanter’s argument, were not pivotal to the outcome of a party’s petition for certiorari (McGuire and Caldeira 1993).

flemming2.p65

44

01/04/2004, 2:24 PM

Litigants, Lawyers, and the Tournament of Appeals

This research ignored the attorneys who drafted the petitions, prepared the briefs, and argued the cases before the Supreme Court. Until McGuire’s work, no scholar had teased apart the conflation of party and attorney that formed a central ambiguity in Galanter’s argument (Lempert 1999). McGuire claims that attorney experience before the Court matters more than the status of the attorney’s client in affecting the outcome of a case (McGuire 1995a). His research also challenges the conventional wisdom that the success of the Solicitor General’s Office in the Supreme Court reflects its “special relationship” with the justices. Instead, he shows that it is the litigation experience of the Assistant Solicitors General and not the privileged position of the Office that counts most (McGuire 1998). Finally, his study of the certiorari process reveals that attorneys with previous litigation experience before the Court are more likely to have their petitions accepted than attorneys lacking this experience (McGuire 1993, 180-87). McGuire does not dismiss the status of the parties as a factor; his data show that party status can make a difference. The important point is that lawyers and their clients can each independently affect the progress of litigation. Galanter’s party capability theory has been tested with some regularity in other countries with mixed results, but McGuire’s work on repeat player counsel has not yet been replicated. The relationship between party status and court decisions on the merits is supported by research in the English Court of Appeal (Atkins 1991). In the Philippines, however, party status matters in Supreme Court decisions but not in the way Galanter expected: have-not parties with few resources prevail over the haves (Haynie 1994). In Israel’s High Court, the haves hold a limited advantage but it depends on whether the have-nots opposing them employ lawyers; when they do, the haves do not come out ahead (Dotan 1999). Finally, McCormick’s analysis of who wins in Canada’s Supreme Court revealed that high-status or upperdog parties, mostly governments but especially the federal government, won more often than have-nots or underdog parties during the period 1949-92 (McCormick 1993). A recent comparative analysis of several countries found no consistent patterns linking outcomes with party status across the countries (Haynie et al. 2001). No comparative studies resting on either Galanter’s thesis or McGuire’s arguments have tested the litigant-resources account with regard to the selection of appeals for judicial review. Defining the Players’ Capabilities in the Leave Tournament

Capability refers to the status, experience, and resources the major players – the parties and their attorneys – bring to each round of the tournament.

flemming2.p65

45

01/04/2004, 2:24 PM

45

46

Litigants, Lawyers, and the Tournament of Appeals

This chapter’s chief concern is assessing whether experienced attorneys or repeat players with ample resources seeking access to the plenary docket of the Supreme Court of Canada have an advantage over less experienced lawyers with fewer resources. McGuire’s analysis suggests that this is the case in the United States, but his findings have not been tested in other countries. The question is whether McGuire’s findings also apply to Canada. The repeat player status of attorneys was measured by the number of times they appeared before the Supreme Court. The Supreme Court Reports for 1975-92 were culled to prepare a list of attorneys and how often they argued the merits of cases before the Court. This list was matched with the roster of attorneys representing the applicants and respondents involved in the leaves to appeal during 1993-95. Leave attorneys not on the merits list were classified as one-shot players and given an experience score of 0, while those attorneys on the list were classified as repeat players and given experience scores corresponding to their number of appearances before the Court. As Chapter 2 indicated, experience and resources overlap, and thus it is important to distinguish the two factors and control for the effect of each. Two resource measures introduced in Chapter 2 are used here: the size of the attorney’s firm and whether the attorney was a Queen’s Counsel. Solo practitioners were coded 1 while the values for the other private attorneys equalled the number of attorneys in their firms as listed in the Canadian Law List 1996. Large firms provide lawyers with the resources needed to prepare well-researched and persuasive arguments for leave to appeal or, alternatively, strong briefs against leave. The relationship may not be straightforward, though, as boutique firms specializing in appellate litigation, such as Eberts Symes Street & Corbett in Toronto, may be at no disadvantage when one of its attorneys faces an associate or partner from one of the much larger Seven Sisters. Professional reputation is another resource. Attorneys who were Queen’s Counsel may win the leave tournament if only because several justices were also Queen’s Counsel. If this factor is nothing more than a surrogate for experience before the Court, it will have little independent effect once the latter is taken into account. A dummy variable indicates whether or not the attorney was Queen’s Counsel. Party status or resources may exert independent influence on the leave tournament. Galanter hypothesized that parties with higher status or more resources would be more successful litigants than lower-status parties or those with few or meagre resources. A major obstacle to testing this hypothesis is developing direct indicators of resources or status.

flemming2.p65

46

01/04/2004, 2:24 PM

Litigants, Lawyers, and the Tournament of Appeals

Table 3.1 Definitions and characteristics of variables used in this study Variable Leave granted? Attorney experience Applicant attorney Respondent attorney Attorney Queen’s Counsel? Applicant attorney Respondent attorney Private attorney firm size Applicant attorney Respondent attorney Party resources/status Applicant status Respondent status

Value or range

Mean

N

1, 0 0-22

.189

1,265

1.352 1.125

1,157 1,218

.197 .185

1,265 1,265

58.315 63.063

572 558

2.587 5.470

1,265 1,265

1, 0

0-550

1-9

The solution adopted here follows previous research that ranked the parties by their presumed status or resources. Parties were categorized in the following order from low to high: individual, groups/associations, unions, businesses, Crown corporations, municipal government, provincial government, and federal government. This classification, with slight modification, is very similar to the one McCormick (1993) used to assess whether party status was associated with winning on the merits in Canada’s Supreme Court. Table 3.1 summarizes the definitions of the variables and their characteristics used in this chapter. The measures shown in Table 3.1 are part and parcel of the literature on party capability theory and the repeat player hypothesis. However, the most appropriate indicators should measure the inequalities between attorneys or the parties. The crux of the question, after all, is whether an imbalance in experience or status between attorneys or between the status of one party relative to another affects agenda setting on the Canadian Supreme Court. This chapter therefore focuses on the differences between the measures in Table 3.1 as indicators of inequality or asymmetry between the parties or lawyers. For example, the actual numeric difference in the frequencies of appearances by the two attorneys before Canada’s Supreme Court creates a “lawyer experience advantage” variable. A positive number indicates an advantage for the applicant’s attorney, while a negative number indicates an edge for the respondent’s lawyer. The same tack was taken with regard to the Queen’s Counsel, law firm size, and party status measures to create variables tapping

flemming2.p65

47

01/04/2004, 2:24 PM

47

48

Litigants, Lawyers, and the Tournament of Appeals

asymmetries in status or resources between the attorneys or between the parties.2 The dependent variable is dichotomous (leave granted or not). The multivariate model is logistic regression because of ease of computation and presentation. To show the substantive effect of a variable, the change in probability under different values of the variable is shown (Greene 1997). For dummy independent variables (e.g., Queen’s Counsel or not), the change in probability is from .50 if the variable goes from 0 to 1. For interval variables (such as attorney experience), the change in probability from .50 is for 1 standard deviation. Since the logit curve is steepest at .50 probability, these estimates indicate the maximum impact of these variables. Inequalities: A Simple First Look

Before we proceed to the multivariate analysis, a quick look at how asymmetries between the parties or between lawyers are related to whether or not the justices grant judicial review is in order. Table 3.2 compares the proportions of applications granted leave by the relative imbalance in resources or status between the applicant and respondent. When the status score of the applicant is subtracted from the respondent’s score, the difference between the two measures yields the index of party advantage shown in this table. When an applicant has a status advantage relative to the respondent, the variable has a positive sign; otherwise, the sign is negative. The bivariate relationship is statistically significant with a gamma of 0.15, a

Table 3.2 Applicant success rates by party advantage, leaves to appeal, 1993-95 Index of party advantage –8 –7 to –5 –4 to –3 –2 to –1 0 1 to 2 3 to 4 5 to 7 8

Proportion of wins by applicant (%)

Number of applications

18.0 13.4 15.5 18.9 16.1 24.3 24.3 37.0 40.0

361 201 155 74 261 37 74 27 75

Note: Chi-square = 53.201 (p < .000); gamma = 0.153 (p < 0.01).

flemming2.p65

48

01/04/2004, 2:24 PM

Litigants, Lawyers, and the Tournament of Appeals

moderately strong association between the two variables. Three aspects of Table 3.2 warrant attention. First, only one out of six applications pitted parties of roughly equal status or resources against one another. In other words, the players in the leave tournament are unevenly matched most of the time. Second, the likelihood that leave is granted improves as the status of applicants rises relative to that of respondents, which is consistent with McCormick’s findings on the merits. The third consideration counterbalances this pattern in that nearly 60 percent of the leave applications involve imbalances that favoured the respondents. In most instances, this reflects individual applicants asking leave to appeal lower court losses to governments or businesses. Table 3.3 portrays the proportions of applications granted according to the relative imbalance in Supreme Court experience between attorneys. The pattern in this table is less clear-cut and seems to offer unpromising support for the repeat player hypothesis since the measure of association, gamma, is small and not statistically significant. Attorneys with less experience than their opponents appear to fare as well as those who are more experienced vis-à-vis their opponents. If an applicant’s attorney were substantially more seasoned than the respondent’s attorney, the “win” rate would be expected to be substantially higher than if the inequality were reversed. The pattern in this table does not support this hunch. Moreover, in contrast to the party advantage variable, the vast majority of attorney pairings are clustered in the middle of the indicator’s values. Most of the time, the pairs of attorneys were more or less evenly matched in terms of their experience.

Table 3.3 Applicant success rates by lawyer experience advantage, leaves to appeal, 1993-95 Lawyer experience advantage

Proportion of wins by applicant (%)

Number of applications

42.3 17.6 26.5 18.4 16.1 16.1 26.4 31.1 23.3

26 34 102 114 498 155 129 45 30

≤ –8 –7 to –5 –4 to –2 –1 0 1 2 to 4 5 to 7 ≥8 Notes: Chi-square = 70.940; gamma = 0.009.

flemming2.p65

49

01/04/2004, 2:24 PM

49

50

Litigants, Lawyers, and the Tournament of Appeals

Table 3.4 Applicant success rates by Queen’s Counsel advantage, leaves to appeal, 1993-95 Queen’s Counsel advantage

Proportion of wins by applicants (%)

Number of applications

18.1 18.0 24.6

160 930 175

–1 0 1 Notes: Chi-square = 4.276; gamma = 0.114.

This rough equality between attorneys may blunt whatever advantage the inequalities between the parties they represent may generate in the leave process. Besides experience, though, professional reputations may matter. Table 3.4 presents the data for attorneys who were Queen’s Counsel and those who were not. The attorneys involved in the vast majority of leave applications were evenly matched in that most of the time neither attorney was a QC. When a mismatch occurred, however, applicants who were Queen’s Counsel were more likely to win than when the situation was reversed. There may be some marginal advantage, then, to being a Queen’s Counsel, although the differences are not striking, and although gamma is 0.11, it falls shy of reaching the .05 level of statistical significance. Major law firms, like McCarthy Tetrault with over 500 lawyers, have the staff and intellectual infrastructure to support the firm’s litigators. Large firms publish in-house analyses of court decisions from throughout the country. They conduct seminars on emerging areas of the law. They are actively involved in the legal profession, political parties, the provincial and federal governments, and, of course, the courts. (A former president of the Canadian Bar Association was a partner in McCarthy Tetrault, and a current justice on the Supreme Court worked for the firm before his appointment to the bench.) These advantages might show up in the leave process. Table 3.5 provides mixed support for this expectation. The index of firm advantage is the absolute difference between the firm size of the applicant and the firm size of the respondent. The differences are arrayed in ascending order of the applicant’s advantage so that the index’s relationship to applicant success rates can be shown. The applicants’ success rate, the percentage of applications granted leave for each level of the index, was expected to increase or decrease in a regular fashion with each increment of the index, depending on whether the applicant’s attorney worked in a larger or smaller firm than the respond-

flemming2.p65

50

01/04/2004, 2:24 PM

Litigants, Lawyers, and the Tournament of Appeals

Table 3.5 Applicant success rates by attorney firm advantage, leaves to appeal, 1993-95 Index of firm advantage –549 –250 –70 –25 –9 –1 2 10 26 71 201

to to to to to to to to to to to

Proportion of wins by applicant (%)

Number of applications

19.2 29.0 11.3 9.2 16.7 8.3 23.5 17.6 27.7 17.1 16.2

26 69 71 65 54 48 51 34 47 35 37

–201 –71 –26 –10 –2 1 9 25 70 200 549

Notes: Chi-square = 18.683; gamma = 0.153.

ent’s attorney. If asymmetries in firm resources matter, the success rate of applicants should steadily increase from lower to higher. The pattern in Table 3.5 is neither orderly nor uniform. The pattern generally follows this expectation, but there are exceptions or irregularities. The measure of this relationship has a value of 0.15 but it is not statistically significant, although the chi-square is significant at the .05 level. Inequalities: A More Complicated Second Look

Because bivariate analyses can conceal important relationships, a series of logit models will be used to assess the relative impact of asymmetries between the pairings of attorneys and parties on whether leave was granted or denied. The first model is an additive logit model that ignores for the moment the inequalities between parties and attorneys. This model therefore assumes that the individual levels of attorney experience, whether the attorney was Queen’s Counsel, and the status of the parties separately affected the leave tournament without regard to these same characteristics for their opponents. In this sense, the model sets a baseline for the more complicated models that follow. Table 3.6 presents the results for this model. Attorneys and parties have different goals depending on their side in cases. Applicants and their lawyers hope to have judicial review granted so they can argue that the lower court decision against them should be overturned. Respondents and their attorneys presumably prefer that leave not be granted lest their lower court victories be jeopardized. Applicant

flemming2.p65

51

01/04/2004, 2:24 PM

51

52

Litigants, Lawyers, and the Tournament of Appeals

Table 3.6 Logit model of leave to appeal outcomes, government and private litigants, 1993-95 Independent variable

Expected direction

Applicant lawyer Experience Queen’s Counsel? Respondent lawyer Experience Queen’s Counsel? Appellant party type Respondent party type Constant Percent in modal category Percent correctly predicted Proportional reduction in error Nagelkerke pseudo R2 Model c2 (6 d.f.) N

b

SE

∆p

.08

+ +

.017 .338 *

.027 .186

– – + –

.103 ‡ –.052 .150 ‡ –.008

.027 .201 .032 .028

–2.022 ‡

.233

.09 .18

80.14 80.67 .03 .07 47.584 ‡ 1,133

Notes: The dependent variable is coded 1 if the appeal was accepted (appellant wins) and 0 if it was not. * p ≤.05, † p ≤.01, ‡ p ≤.001, one-tailed; two-tailed test used for constant.

attorneys with prior experience before the Court should succeed more often in persuading the justices to grant leave than those with less or no experience. Similarly, repeat player attorneys for the respondents ought to convince the Court that leave is unwarranted more often than oneshot attorneys. Similar expectations apply to attorneys who were or were not Queen’s Counsel. And, of course, higher-status or resource-rich litigants should get their way more often in the leave process than parties lacking resources or status. The results in Table 3.6 contradict and fail to support some of these expectations. No statistically significant relationship exists between leave decisions and the amount of prior experience the applicants’ attorneys have before the Supreme Court. Yet, for respondents’ attorneys, there is a significant association, which confirms the hunch that there should be a link between experience and leave decisions, while at the same time a positive sign – not the anticipated negative sign – is attached to this link that confounds expectations. When the number of prior Supreme Court appearances by respondent attorneys moves from the mean

flemming2.p65

52

01/04/2004, 2:24 PM

Litigants, Lawyers, and the Tournament of Appeals

by 1 standard deviation, the odds of leave being granted increase 9 percent. How can this finding be explained? The robustness of the finding suggests that it cannot be dismissed as a statistical fluke. Even if that were the case, why does this finding occur only for respondents’ attorneys and not the applicants’ lawyers as well? Surely it is not something that is wanted by respondents and their attorneys except under special circumstances. That is to say, a plausible explanation rooted in the goals of the respondents is hard to come by. One possible explanation may be found in the decisional tendencies of the Supreme Court. Canadian justices are more likely to affirm lower court decisions than overturn them. In contrast to the United States Supreme Court, which affirms about a third of the cases it reviews, the Canadian Supreme Court affirmed 54 percent of the 215 appeals to which it granted leave during 1993-95.3 Of course, the parties and their lawyers filing leave applications hope the justices will do just the opposite and overturn lower court rulings. For the other side of the appeal, attorneys for respondents need to provide reasons why there is no need to review a lower court ruling and why it should be left standing. In something of an ironic twist, the respondents’ arguments may provide justification for taking up the lower court ruling and extending it nationwide rather than letting it stand only as provincial law. While rebutting the applicant’s request for leave, a skillful and experienced lawyer may draw attention to issues that persuade the justices there is good reason to take up the lower court decision to see whether it should become the law of the land. This is not necessarily a defeat for the respondent or the attorney. Indeed, it could be just the reverse, particularly in a country where the high court affirms more often than it overturns the lower courts. In other words, the respondent, instead of resisting judicial review, may urge the Supreme Court to take the appeal and hope that the Supreme Court will extend the provincial ruling across the whole of Canada.4 The other findings in Table 3.6 are somewhat more consistent with initial expectations. Attorneys who are Queen’s Counsel make a difference to the outcomes of the process, but only in the case of applicants. Applicants with QC attorneys are 8 percent more likely to have their applications granted than applicants whose attorneys lack this credential. No statistically significant relationship appears to exist for respondents’ attorneys who are Queen’s Counsel. Finally, the status of the applicants has a statistically significant relationship with leave outcomes, but not the status of the respondents. When applicant status increases by 1 standard deviation, the application is 18 percent more likely to be

flemming2.p65

53

01/04/2004, 2:24 PM

53

54

Litigants, Lawyers, and the Tournament of Appeals

Table 3.7 Litigant advantage model of leave to appeal outcomes, government and private litigants, 1993-95 Independent variable

Expected direction

Lawyer experience advantage Queen’s Counsel advantage Status of party advantage Constant Percent in modal category Percent correctly predicted Proportional reduction in error Nagelkerke pseudo R2 Model χ2 (3 d.f.) N

+ + +

SE

∆p

–.338 * .199 .064 ‡

.023 .150 .015

–.29

–1.245 ‡

.081

b

.12

80.14 80.14 .00 .03 20.897 ‡ 1,133

Notes: The dependent variable is coded 1 if the appeal was accepted (appellant wins) and 0 if it was not. * p ≤.05, † p ≤.01, ‡ p ≤.001, one-tailed; two-tailed test used for constant.

granted. In conclusion, the findings in Table 3.6 are mixed at best, and the model’s overall performance, as indicated by the diagnostics in the table, is not impressive. The central idea behind the presumed advantage of repeat players over one-shotters and the edge that haves hold over have-nots pivots on the inequalities or imbalances between the lawyers and between the parties. Each round in the leave tournament is a confrontation between these two sets of actors. Experience and resources may most shape tournament play when they create privileges for one party relative to the other. Equality, as noted earlier, neutralizes the benefits that experience, credentials, or resources might bestow on a party or its lawyer. To assess the impact of asymmetrical relationships and interactions, the appropriate variables must be included in the model. Table 3.7 reports the results of an additive model that includes only the three inequality variables between parties and lawyers. Once again the results are not encouraging for the litigant-resources account. The account receives a boost from the finding that status inequalities between the parties are statistically significant with a positive sign that matches expectations. Applicants with resource advantages over respondents are more likely to be granted leave than applicants who are disadvantaged vis-à-vis respondents. The results for the variable with the most bearing on the repeat player hypothesis, however,

flemming2.p65

54

01/04/2004, 2:24 PM

Litigants, Lawyers, and the Tournament of Appeals

are disappointing; they also run counter to what the hypothesis predicts. The relationship, although statistically significant at the .05 level, has the wrong sign. The expectation was that applicants with more experienced attorneys than respondents would have an edge. The sign therefore should have been positive. The fact that it is negative can be explained by the larger context of the process, namely, the tendency of the Court to affirm lower court decisions more often than it overturns them. In addition, no statistically significant relationship was found for the Queen’s Counsel variable. The diagnostics included in Table 3.7 show that although the overall equation is statistically significant, its predictive and explanatory powers are weak. Influence is not only relational in nature but also contextual. Indicators of asymmetrical resource advantages may interact with one another to produce more complex relationships. For instance, asymmetrical relationships between parties when combined with similar imbalances between attorneys could interact in ways that exacerbate the effects of these inequalities on leave decisions compared with their impacts taken separately. To test for these effects, the model described in Table 3.8 includes interaction terms for each of the inequality variables. The interaction terms do not alter the statistical significance of the party status advantage variable, nor do they change substantially the parameter estimates for this variable. The model diagnostics are slightly improved over the previous model. However, the interaction terms render the lawyer experience advantage variable statistically non-significant. Two interaction terms pass standard statistically significant thresholds. Both terms combining the lawyer experience advantage variable or the Queen’s Counsel advantage variable with the party advantage variable reach statistical significance. The interaction term between lawyer experience and Queen’s Counsel fails this test. It appears, therefore, that the advantage repeat player attorneys may have over their less experienced or non-QC opponents is conditional on the status advantage their clients hold over the opposing party. An important caveat should be attached to this conclusion. Governments are frequent litigators before Canada’s Supreme Court and there are important differences between governments and other haves (Lempert 1999). When governments, which win most of the time, are included as haves, their inclusion distorts and magnifies the impact of party status and inequalities between parties. More than half of the leave applications involved governments as either applicants or respondents. The construction of the party status variable, following previous research, weighs governments more heavily than other parties in its ranking of

flemming2.p65

55

01/04/2004, 2:24 PM

55

56

Litigants, Lawyers, and the Tournament of Appeals

Table 3.8 Interactive logit model of leave to appeal outcomes, government and private litigants, 1993-95 Independent variable

Expected direction

Lawyer experience advantage Queen’s Counsel advantage Status of party advantage Lawyer experience adv. × Queen’s Counsel adv. Lawyer experience adv. × status of party adv. Queen’s Counsel adv. × status of party adv. Constant Percent in modal category Percent correctly predicted Proportional reduction in error Nagelkerke pseudo R2 Model χ2 (6 d.f.) N

b

SE

∆p

.11

+ + +

–.005 .038 .059 ‡

.028 .161 .016

+

.047

.032

+

.012 †

.005

.08

+

–.075 †

.086

–.05

–1.330 ‡

.086

80.14 80.41 .02 .06 39.619 ‡ 1,133

Notes: The dependent variable is coded 1 if the appeal was accepted (appellant wins) and 0 if it was not. * p ≤.05, † p ≤.01, ‡ p ≤.001, one-tailed; two-tailed test used for constant.

direct parties. What happens if applications involving governments were taken out of the model, leaving only those with private attorneys and private parties to face each other? If government parties and their attorneys hold a privileged status before the Court, a model without governments might offer a purer assessment of the litigant-resources account. It would also bear more directly on the status-seeking model presented in Chapter 2, which focused on private attorneys. Stratification of these lawyers occurs in part because of status-seeking behaviour, but does it affect the outcomes of the leave tournament? Table 3.9 presents the results of this private litigants only model, which adds a variable for disparities in firm resources between the opposing lawyers. With governments and their attorneys out of the model, the party status advantage still manages to reach statistical significance, although at a lower level than in the earlier models. None of the other variables attains statistical significance. The model diagnostics are equally unimpressive. These results prompt the conclusion that much of the impact of the party status variable as a ranking of parties reflects the strength of

flemming2.p65

56

01/04/2004, 2:24 PM

Litigants, Lawyers, and the Tournament of Appeals

Table 3.9 Logit model of leave to appeal outcomes, private litigants only, 1993-95 Independent variable

Expected direction

Lawyer experience advantage Queen’s Counsel advantage Lawyer firm advantage Status of party advantage Constant Percent in modal category Percent correctly predicted Proportional reduction in error Nagelkerke pseudo R2 Model χ2 (4 d.f.) N

+ + + +

SE

∆p

.026 .079 –.001 .083 *

.054 .220 .001 .041

.09

–1.471 ‡

.122

b

82.18 82.18 .00 .02 5.57 522

Notes: The dependent variable is coded 1 if the appeal was accepted (appellant wins) and 0 if it was not. * p ≤.05, † p ≤.01, ‡ p ≤.001, one-tailed; two-tailed test used for constant.

governments as parties, particularly in those instances where individuals confront governments. Conclusion: A Tournament Where Resources Don’t Matter?

Repeat player lawyers who argued cases before Canada’s Supreme Court did not hold any advantages over less experienced lawyers in the tournament for judicial review. Nor did asymmetries in the status or resources between the parties involved in the disputes matter greatly in the Court’s decisions to grant or deny leave. An important exception must be made whenever one of the parties was a government, particularly the federal government. Even with this qualification, though, the results of the various models were far from impressive. And when only private parties and their attorneys are considered, the findings virtually evaporate, leaving the party capability hypothesis high and dry. These results are surprising given McGuire’s finding that repeat player lawyers shape the agenda of the US Supreme Court and the popularity of Galanter’s argument that haves and repeat players come out ahead. They also are surprising in light of other research involving Canada’s Court. McCormick (1993) found for a fifty-year period that individuals as a group before Canada’s Supreme Court were less successful than corporations or governments. More recently, Haynie and her colleagues, following roughly the approach of McCormick and his predecessors, found

flemming2.p65

57

01/04/2004, 2:24 PM

57

58

Litigants, Lawyers, and the Tournament of Appeals

that individuals fared poorly against governments but surprisingly well against businesses (Haynie et al. 2001). In general, governments won more often than individuals or corporations at the merits stage. Both of these studies simply compare the win rates on the merits of the various types of parties; they do not use multivariate methods to control for other factors, nor do they test for the possible independent effects of the lawyers representing the parties. If haves and repeat players (other than governments) do not exert much sway when judicial review is granted, then it is hard to see why things would shift in their favour later on when the Supreme Court decided the merits of their appeals. More recently, Morton and Allen (2001) took a different tack and looked at cases decided by the Supreme Court that involved the premier repeat player and feminist group, the Women’s Legal Education and Action Fund (LEAF), to determine not just LEAF’s win rates but whether it effected changes in Canadian public policy. They found that LEAF generally won on the merits but that even when it lost, LEAF’s challenges often reshaped the policy it had questioned. Morton and Allen’s very different research design makes comparisons between their findings and this study difficult at best. This study, of course, focuses only on the preliminary play of the larger litigation game. Moreover, because the Supreme Court discourages interveners from participating in the leave to appeal stage, neither LEAF nor any other organized interest was involved in the tournament. What kind of explanation can be offered for this chapter’s anomalous results when faced with these arguments and findings to the contrary? Perhaps the most vital thing to keep in mind is that most of the time the tournament’s players are fairly evenly matched. Stark asymmetries between haves and have-nots, between high-status and low-status parties, are not uncommon, but neither are they the norm. The same thing can be said even more strongly with respect to the lawyers representing the parties. The impact, if any, of inequalities tends to be at the margin. Still, marginal effects can be important. McGuire (1993) argued that attorneys with reputations and credibility in the Supreme Court compete successfully for the scarce attention of law clerks and justices in the American certiorari process. The lawyers he interviewed stressed the importance of being known and credible because the volume of cert petitions presses heavily on the limited time available to the Court to process them. These American attorneys felt that their identity as repeat players functioned as an important cue for clerks or justices. Why doesn’t this argument apply to Canada? Doesn’t attorney experience before the Court count? Why don’t repeat players

flemming2.p65

58

01/04/2004, 2:24 PM

Litigants, Lawyers, and the Tournament of Appeals

win the leave to appeal game? Could the institutional arrangements in Canada’s Supreme Court provide an answer to these questions? The US Supreme Court confronts a tidal wave of petitions for writs of certiorari compared with Canada, and grants cert to a mere handful of applicants. Canada’s Supreme Court handles a smaller volume of leave applications and grants leave fairly often. The simple fact that one of six requests for leave to appeal is granted probably reduces the impact of experience and resources in the leave tournament. More complicated explanations for why the value of repeat player status is discounted may have to do with information and the institutional arrangements that the Canadian Supreme Court has developed to review leave applications. The lower volume of requests allows the Court more time to consider each application, and the review process is more intensive. The applications are processed through a decentralized system of three-justice panels. The chief attorney of Legal Services assigns applications to the staff attorneys, who prepare the objective summaries as applications are submitted to the Court. After the staff attorneys finish their summaries, which include recommendations on whether leave should be granted, the chief counsel assigns the applications to the individual leave panels. Before 1996, when the justices’ clerks were removed from the leave process, each clerk prepared another memo, which could also include recommendations, before sending it to the panel justices. These memos were often substantial and lengthy. The third step in this process, of course, involved the panel justices. It may be the case, therefore, that the amount of scrutiny given the applications by the staff attorneys, clerks, and justices mitigated the value of a well-known name, unlike in the United States. It should also be noted that during the study period the decision to grant judicial review depended on the views of a staff attorney, one clerk, and two but more often three justices on the panel. Information about leave applications was abundant but the decentralized process may have made it something of a lottery, depending on the mix of staff attorney, clerk, and panel justices. The leave decision for an experienced attorney or one with an edge over his or her opponent could therefore reflect the luck of the draw, which, combined with the relatively high rate at which leave is granted and the quantity of information, reduces the statistical significance and impact of being a repeat player. It appears, then, that the institutional features of the Canadian agendasetting process – a low volume of requests for judicial review, more liberal access to the Supreme Court’s plenary docket, abundant information, and a decentralized review of leave to appeal applications – diminish

flemming2.p65

59

01/04/2004, 2:24 PM

59

60

Litigants, Lawyers, and the Tournament of Appeals

the advantages of repeat players as they play the leave tournament in Canada. Since the litigant-centred account has failed to generate an explanation of how Canada’s Supreme Court selects appeals for judicial review, it is time to turn to an assessment of the jurisprudential and strategic accounts.

flemming2.p65

60

01/04/2004, 2:24 PM

4 Tournament Rules and the Consequences of Institutional Choices

Introduction: A Jurisprudential Account of the Tournament

In sporting tournaments, the referees enforce the rules of the game. They do not unilaterally change the rules or decide the winners of the tournament except when penalizing players for breaking the rules. The tournament of appeals differs dramatically from its sporting cousins in several ways. For one thing, the rules in the leave tournament are not transparent to the litigants or their lawyers, a situation the justices prefer to maintain. The players must therefore rely on hunches based on what has happened in their cases, check which appeals are granted judicial review, or pick up ideas from other players as to how the justices are interpreting and applying “public importance” as a standard in the leave process. The justices compound the litigants’ problem by refusing, as in the United States, to be bound by the criteria they set for themselves when they choose cases for judicial review. A further complication is the fact that the justices may change the rules of the game by revamping the criteria or the priorities they give to them without telling the players what they have done. As a consequence, the tournament is played as though it were a hockey game with invisible nets, the locations of which are known only to the referees. If the players sense where the nets are and begin to score goals, the referees move the nets to another place on the rink. This may be an overly dramatic depiction of the situation but it points to an interesting paradox: if the rules governing when to grant judicial review matter, it is curious that courts are reluctant to explain them. The reason, of course, is obvious. By deliberately keeping the actual or operative rules ambiguous, the justices preserve their freedom to set their agendas as they see fit. In a sporting tournament, the referees have no interest in who wins or loses other than that the game be played by the rules. High court judges,

flemming2.p65

61

01/04/2004, 2:24 PM

62

Tournament Rules and the Consequences of Institutional Choices

in contrast, are keenly interested in the outcomes of the tournament of appeals and are prepared to alter the rules to achieve their goals. In the United States, the litigant-centred account of agenda setting has largely been told without inclusion of rules, and it has been remarkably successful without them. This is not to say that the story ignores jurisprudential or legal rules pertinent to agenda setting altogether, merely that their role is minimized and limited to the type of issue in dispute or whether lower courts have rendered conflicting interpretations of the law. The preceding chapters cast doubts on whether the litigant-centred account standing alone succeeds as well in Canada. Perhaps, though, the litigant-centred model in the previous chapter was underspecified; it may have excluded important variables suppressing the actual relationship between litigants, their lawyers, and leave decisions. One reason, then, for considering seriously the jurisprudential account is to address this possible weakness. A second reason is that the legal factors in the jurisprudential account have long been part of the conversation about the agenda-setting process – that is, most discussions about certiorari in America or leave to appeal in Canada begin with the rules. For at least these two reasons, an inquiry into the criteria that the Supreme Court of Canada uses when it decides whether to grant leave is appropriate. “Cue theory,” an early attempt by political scientists to explain the American certiorari process, included variables that are part of the jurisprudential account, namely, the issues in the appeals and whether the lower courts disagree about the proper outcomes of cases (Tanenhaus et al. 1963). According to the theory, the justices and clerks, inundated with hundreds or thousands of appeals for judicial review, rely on a few simple cues to expedite their search for qualified cert petitions. As the jurisprudential story progressed, the saliency of particular issues waxed or waned depending on the Court’s priorities, but in general they faded in importance compared with more politically interesting variables. Lower court disagreements or conflicts, in contrast, have been an enduring staple of the account since Congress passed the Judiciary Act of 1925. Felix Frankfurter, future Supreme Court justice and Harvard law professor, remarked after several years of observing the cert process that conflict among the lower courts “remains by far the most frequent ground for granting the writ” (Frankfurter and Landis 1993, 267). Confusion or disagreement about the law in the lower courts continues to be a major consideration for the justices when they grant certiorari. Frankfurter’s conclusion, however, was a guess based on careful observation, and while cue theory relied on information from certiorari deci-

flemming2.p65

62

01/04/2004, 2:24 PM

Tournament Rules and the Consequences of Institutional Choices

sions for its conjectures, the fact remains that until recently the grounds or criteria for granting judicial review were opaque at best. Justices in both countries prefer to remain silent about their decisions, although the American justices on occasion publish dissents when they disagree with particular decisions of their colleagues, and they release now and then a summary explanation for denying certiorari (Stern et al. 1993, 235; Epstein and Knight 1998). Neither practice has caught on in Canada, where the justices are notably mute about particular decisions and until recently avoided even general discussions about what they looked for in the leave process. In 1995 Chief Justice Antonio Lamer explained the Court’s reticence (Crane and Brown 1998, 24): The ability to grant or deny leave represents the sole means by which this Court is able to exert discretionary control over its docket. In order to ensure that this Court enjoys complete flexibility in allocating its scarce judicial resources towards cases of true public importance, as a sound rule of practice, we ... do not produce written reasons for grants and denials of leave.

Two years later, in 1997, the late Justice John Sopinka tried to fill in some of the blanks. His interest in the topic, perhaps because of his academic background, was long-standing. Before his appointment to the Supreme Court in 1988, Justice Sopinka co-authored a book on appellate practice that included several guesses about the Court’s interpretation of the public importance criterion (Sopinka and Gelowitz 1993, 167). He and his colleague felt that applications were more likely to be granted leave if they featured one or more of the following: • • • • •

a novel constitutional issue a significant federal statute of general application a provincial statute similar to legislation in other provinces conflicting decisions in the provincial courts of appeal the need to revisit an important question of law.

They did not define such key words as “novel,” “significant,” and “important,” and thus left them as terms of art for their readers to ponder. The items on their list, nevertheless, are neither unusual nor unexpected for a common law high court in a federal system of government. After several years on the bench, Justice Sopinka drew on his experience to sketch an insider’s perspective of how the Court conducted its business in a speech that, with his permission, was first published in the

flemming2.p65

63

01/04/2004, 2:24 PM

63

64

Tournament Rules and the Consequences of Institutional Choices

1998 edition of the Supreme Court of Canada Practice and in subsequent editions of this manual for the benefit of Supreme Court litigators. Justice Sopinka divided the various rules into general principles and specific guidelines. He underscored the basic principle that misapplications of the law or misinterpretations of Supreme Court judgments were by themselves insufficient grounds for judicial review unless they became an “epidemic,” in which case “we may want to set the record straight” (Crane and Brown 2002, 396). Public importance clearly is a primary consideration, but in a partial contradiction of the first principle, the Court will relax this standard in criminal cases “even in the absence of an ‘earth-shaking’ issue of law” if the applicant appears to have been unfairly convicted (Crane and Brown 2002, 397). Other general principles include whether the issue was properly raised in the lower courts and whether an adequate record exists for review; or if the Court has recently decided an issue, it may choose to let the lower courts mull over its decision for a while before revisiting the question. The justice then listed several guidelines dealing with public importance, some of which were anticipated by his earlier thoughts (Crane and Brown 2002, 397-98): •

•

•

•

•

a constitutional challenge to a statute, common law rule, or government practice conflict between provincial courts of appeal when there should be uniformity between provinces a novel point of law, for example, reversing the burden of proof of causation interpretation of “important” federal laws or a provincial statute that exists in other provinces defining Aboriginal rights.

Aboriginal rights are the most significant addition to this list and reflect the Court’s contemporary policy priorities; the others, aside from wording and elaboration, are similar to Justice Sopinka’s earlier list. Other rules, such as standing, mootness or ripeness, and justiciability, that affect whether the Court can exercise jurisdiction over a case also influence the decision to grant leave, but Canada’s rules on these matters are considerably more flexible than those in the United States (Sossin 1999). A final consideration, Justice Sopinka notes, is that even if the lower court was “plainly right,” the Supreme Court may still grant leave “if a decision binding on the whole country is desired” (Crane and Brown 2002, 398). This clearly adumbrates the finding in the last chapter that

flemming2.p65

64

01/04/2004, 2:24 PM

Tournament Rules and the Consequences of Institutional Choices

successful grants of judicial review in some instances are associated with the respondents’ characteristics. A fully elaborated jurisprudential account of agenda setting would include these various rules, guidelines, or principles. Cue theory as an explanation of agenda setting in the US Supreme Court included only lower court disagreements and two issue areas. Further development of a rules-oriented model stalled and came to a halt until fairly recently. Attention instead focused on testing or assessing cue theory’s components, with contradictory results. Ulmer (1972) concluded that the legal cues were time-bound, while Armstrong and Johnson (1982) disagreed. Provine (1980) took exception to the stress on lower court conflict as a factor and re-emphasized the public policy role of the Court in terms of the kinds of issues raised by the petitioners. Subsequent assessments of cue theory suggest that the justices and clerks sifted the wheat from the chaff since “actual” conflicts were found to matter more in the cert process than attorney claims of conflict (Ulmer 1983). A later study, distinguishing disagreements between the lower courts (the appellate court overturns the lower court) and alleged versus real conflict between appellate courts, revealed that both contributed to putting a petition on the Supreme Court’s “discuss list” for deliberation at conference, but that only actual conflicts affected whether or not cert was granted (Caldeira and Wright 1990b). If petitioners raised civil liberties claims, their requests for review might be placed on the discuss list, although the issue was not enough to prompt a decision to grant cert. It is obvious that lower court disagreements and conflicts play important roles in the jurisprudential story. In both high courts, confusion and misinterpretation in the lower courts attract judicial review, and if they become pandemic, the justices feel the need to straighten things out. Other elements of the story are rarely included, however. Unfortunately, Justice Sopinka gave his speech on the leave process a year after data collection for this project was completed. Since the fullest exposition of a jurisprudential account is part of Perry’s in-depth and detailed study (1991) of how the American Supreme Court “decides to decide,” based on his interviews with five justices and sixty-four clerks, this project used his discussion to compile a list of legal factors expected to influence the decision to grant judicial review in Canada. Identifying the Tournament’s Rules of Play

Although Perry identified various specific indices of “certworthiness” through these interviews, he offered primarily anecdotal evidence for

flemming2.p65

65

01/04/2004, 2:24 PM

65

66

Tournament Rules and the Consequences of Institutional Choices

their effects; this study is the first instance where a multivariate statistical model has tested the impact of these variables on the selection of cases for judicial review. Some of the factors that Perry claims influence the process are not easily translated into operational variables, however, because they are not on the record or in the files and briefs but instead are part of the choice repertories of the justices. An example is “percolation,” a judicial preference to wait until an issue has been heard more than once in the lower courts. Another is the judicial desire to set aside a case in favour of another that is “in the pipeline.” A final example is the “egregiousness” of the lower court decision, which, like obscenity, may be identifiable by justices only when they see it. The factums or briefs filed by the applicants’ attorneys provided information regarding the following indicators adapted from Perry’s study and revised for this project: •

• • • • •

• • • • • •

conflicting decisions – among provincial appellate courts – between a provincial court and Federal Court of Appeals – between a lower court decision and a Supreme Court ruling federal interests affected by the lower court decision provincial interests affected by the lower court decision scope of interests potentially affected by the case limited generality of case affecting only the direct parties parochialism of case focusing only on the procedural rules or specialized statutes novel issue revisit an earlier Supreme Court ruling urged in factum jurisdictional error by the lower court or administrative tribunal fact-specific quality of application’s argument error correction emphasis without link to larger legal issues frivolous appeal.

Two additional indicators were constructed from the case-level data collected from the Bulletin with regard to the following: • •

issue involved the Charter of Rights and Freedoms dissent in the lower appellate court.

Many of these indicators are clearly similar to those identified by Justice Sopinka. It appears, then, that the American and Canadian Supreme

flemming2.p65

66

01/04/2004, 2:24 PM

Tournament Rules and the Consequences of Institutional Choices

Courts have similar considerations in mind when their justices pick cases for judicial review. Each of the indicators is a dummy variable where 1 equals the presence of the characteristic or argument and 0 indicates its absence. The previous chapter presented models that tested the litigant-resources hypothesis. Those models did not include jurisprudential variables. It will be a useful check on the results presented in this chapter if the indicators from the models are included as controls for the jurisprudential account. The inclusion of the litigant and attorney indicators will also address concerns that the models in Chapter 3 were underspecified; that is, the results may have reflected the limited number of variables included in the models. To recap briefly, the first measure is the ranking of parties by their status or resources. The second measure focuses on whether the attorneys for the applicants and respondents were repeat players before the Court. The third variable is whether the attorneys were Queen’s Counsel. The three indicators of inequality or asymmetrical relationships between the players based on these measures are also included in this analysis. The construction of other independent variables was straightforward, for example, the jurisdictional origins of the applications (e.g., Federal Court of Canada) and whether there was a dissent in the lower appellate court, all of which are identified by dummy variables. The jurisdictional origins of leave applications are important for this analysis because of Canada’s regional tensions. For the first time, the analyses will include one of the distinctive features of the Canadian leave tournament, namely, the use of three-justice panels to decide whether leave should be granted. Each of the panels in operation during the three-year study period is identified by a dummy variable. Table 4.1 briefly describes the variables used in this analysis. The two hypotheses will be tested in a series of steps. The analysis begins with the jurisprudential hypothesis, proceeds to the litigantresources hypothesis, and then adds the origins of the leave applications and the identities of the panels responsible for processing the applications. This phase of the analysis takes the Court as a whole as its unit of analysis. However, the presumptive agenda setters are the individual panels, not the Court as a whole. The final step, therefore, is an analysis of the individual leave panels to determine whether the criteria for granting leave varied among the panels. Because the dependent variable, grant or not grant, is a dummy variable, logistic regression is used for all analyses.

flemming2.p65

67

01/04/2004, 2:24 PM

67

68

Tournament Rules and the Consequences of Institutional Choices

Table 4.1 Definitions and characteristics of major variables used in this study Value or range Dependent variable Leave granted Jurisprudential variables Charter question Novel issue Conflict in lower courts Conflict between provincial and federal courts Conflict with Supreme Court Dissent in lower court Provincial interests Federal interests Broad socio-economic interests Revisit Supreme Court ruling Jurisdictional error Error correction Fact-specific Parochial Frivolous Limited generality Litigant-resource variables Lawyer experience advantage Queen’s Counsel advantage Party status advantage

Mean

N

1, 0

.190

1,266

1, 0 1, 0 1, 0 1, 0 1, 0 1, 0 1, 0 1, 0 1, 0 1, 0 1, 0 1, 0 1, 0 1, 0 1, 0 1, 0

.138 .104 .077 .006 .021 .121 .036 .126 .048 .032 .013 .297 .530 .070 .088 .290

1,266 1,266 1,266 1,266 1,266 1,255 1,266 1,266 1,266 1,266 1,266 1,266 1,266 1,266 1,266 1,266

–22 to 22 –1 to 1 –8 to 8

.228 .012 –2.890

1,134 1,266 1,266

Note: All dummy variables are coded 1 = yes, 0 = no. Dummy variables for provinces and leave panels are not shown.

Round 1: The Court as the Arena of Play

The results for the first round of the analysis are shown in Table 4.2. The table shows a set of statistical models that vary in terms of the variables included in them, moving from the simple to more complex. The negative constant for the first equation indicates the Court’s predisposition to deny leave. In general, the Court granted leave roughly 1 out of 5 times for the applications coded for this study. The jurisprudential hypothesis receives support from several indicators that either counter the predisposition to deny leave or reinforce it. A claim of conflicting lower court decisions, as the American literature amply suggests, is more likely to be heeded by Canada’s Court than an argument lacking this trump card. Lawyers who point to conflicting decisions among provincial appellate courts draw the favourable attention of the justices. As the court of final appeal in a centralized national

flemming2.p65

68

01/04/2004, 2:24 PM

Tournament Rules and the Consequences of Institutional Choices

court structure, the resolution of conflicting lower court decisions is the warp and woof of the Supreme Court’s responsibilities. These claims, it should be noted, are “alleged” and not “real” conflict. Therefore, the coefficient for this measure probably underestimates the actual effect of Table 4.2 Alternative logit models of agenda setting in Canada Jurisprudential variables

Expected

1

2

3

4

b b b b Issues Charter question + .0840 .1829 .1583 .4166 Novel issue + 1.3116 ‡ 1.2561 ‡ 1.2259 ‡ 1.5843 ‡ Conflict Lower courts + 1.1345 ‡ 1.0678 ‡ 1.0791 ‡ 1.1193 † Federal Court + .3638 .3543 .3318 .5133 Supreme Court + .2824 .4702 .4264 .8291 .6790 † .6921 † .8741 † Dissent + .9210 ‡ Public Importance .9815 † .9847 † 1.3584 † Provincial interests + .9210 † Federal interests + .8926 ‡ .9850 ‡ .9959 ‡ .8429 † Economic, social effects + .0364 .0004 –.0160 .2139 .9024 † .8036 * 1.4598 † Revisit SCC decision + 1.1876 ‡ Jurisdiction error + .8197 .8200 .8268 1.2765 * Error correction – –.5481 –.4351 –.4514 –.2862 Fact-specific – –.9613 ‡ –1.0074 ‡ –1.0181 ‡ –.8967 † Parochial – –.1669 –.0838 –.0974 –.2150 Frivolous – –2.3131 † –2.4114 † –2.4432 † –1.9054 * Limited generality – –.2698 –.3320 –.3578 –.0872 Litigant-resources variables Experience advantage + –.0605 * –.0617 * –.0443 QC advantage + .2819 .3060 .1532 .0604 † .0726 † Status advantage + .0574 † Jurisdiction variables British Columbia + .3123 .1158 Ontario + .0528 .1202 Quebec + .1538 .3191 Federal Court + .0917 .5084 Panel variables (non-significant not shown) Panel 6 0 –1.0921 * Constant

–1.6373 ‡ –1.4519 ‡ –1.5259 ‡ –2.0888 ‡

Number of cases Nagelkerke pseudo R2 Model χ2

1,255 1,124 1,121 988 .431 .437 .439 .429 390.250‡ 362.428‡ 363.682‡ 284.324‡ (16 d.f.) (19 d.f.) (23 d.f.) (32 d.f.)

* p < .05, † p < .01, ‡ p < .001, one-tailed; two-tailed test used for constant.

flemming2.p65

69

01/04/2004, 2:24 PM

69

70

Tournament Rules and the Consequences of Institutional Choices

conflicting lower court decisions on leave decisions. A dissenting vote in the lower appellate court also catches the eye of the Court; dissents are significantly and positively related to leave decisions. The two variables dealing with other kinds of alleged conflict fail to reach customary standards of statistical significance. The substantive implications of this finding should be viewed in light of the very few times attorneys pointed out “vertical conflicts.” Just seven applications alleged that there were provincial conflicts with the Federal Court of Canada, and only twenty-six applications claimed a conflict with a Supreme Court ruling. Public importance is another facet of the jurisprudential hypothesis, and the only explicit criterion in Canada’s rules governing leaves to appeal. Six of the significant variables reflect this criterion; three of them are positively related to leave decisions, three negatively. Applications urging the Supreme Court to revisit previous decisions are positively related to leave to appeal decisions. Arguments that show how lower court decisions affect the interests of either the provincial governments or the federal government also improved the chances that applications would be granted leave. In contrast, lawyers who stressed the need to correct procedural or interpretative errors but neglected to link them to larger questions of public importance did not convince the Court that leave was warranted. Arguments that were fact-specific for the same reason reduced the chances of leave. Finally, applications failed when they were frivolous and lacked any clear or obvious legal reason why the Supreme Court should hear the case. Examples of frivolous applications can most readily be found in requests filed by applicants without the aid of legal counsel, or pro se applications. In one criminal appeal, Legal Services’ factual summary stated that the pro se applicant claimed that the spirit of God caused him to violate the traffic laws and that the Court of Appeal erred when it denied him his fundamental right of religious freedom under the Charter. The applicant was convicted of not having his car’s headlights on and for speeding. In his handwritten factum, the applicant cited the Bible to establish a higher authority for his position and argued that “it takes courage when led by the spirit of God to perform an action when it leads you to do something that secondary law prohibits especially when it is enforced by certain government bodies on earth who ignore God’s divine laws and superior principles.” Another example involved a man who had been convicted of disrupting the services of the Jehovah’s Witnesses, of which he had been a member, and who as a condition of probation was prohibited from attending the church. Several years earlier, the man stated in his factum,

flemming2.p65

70

01/04/2004, 2:24 PM

Tournament Rules and the Consequences of Institutional Choices

the Witnesses had “disfellowshipped” him in an in camera proceeding, which he refused to recognize, after he claimed “JEHOVAH of the Watchtower is SATAN the DEVIL” [sic]. Convinced of this “astonishing revelation,” he set about preaching it to his fellow religionists in “Martin Luther-fashion” by wearing placards outside the church that included such messages as “Jehovah is BE-EL’SE-BUB the rule of the DEMONS, Matthew 12:24” [sic]. The man sought judicial review on the grounds that he had been denied his rights under the Charter’s protections of religious freedom. As Justice Sopinka remarked in his speech about the Supreme Court’s procedures, “The applications generally fall into three categories: (1) frivolous, which are obviously to be dismissed; (2) obvious grants; and (3) all others which are not obvious grant or dismiss” (Crane and Brown 2002, 398). As Table 4.2 shows, the first model is statistically significant, with a pseudo R2 of .431. More elaborate versions of this model only slightly improve on this result, and, indeed, the model’s performance deteriorates when the leave panels are included. The second model, with indicators of litigant resources, adds the most to the pseudo R2, although one of the three variables is not statistically significant. If an attorney was Queen’s Counsel and the opposing lawyer was not, the advantage, if there was one, failed to reach statistical significance, although the hypothesized sign was in the right direction. The other two variables dealing with litigant resources play significant roles when the Court is deciding what cases to decide. The sign for the experience advantage variable, though, is contrary to expectations, as it was in the model in Chapter 3. A more complex model that includes jurisprudential variables fails to alter these previous results. A judicial explanation for this finding can be found in Justice Sopinka’s comment that even if an appellate court’s decision is “plainly right,” the Court may still grant leave so that it can make the decision binding on the whole country. Being granted leave clearly is not a defeat for the respondent whose interests transcend the provincial foundations of the lower court decision. One final note: the coefficient for the experience advantage variable is one-fifth the size it attains when the model contains only the three litigant resource variables. This suggests that attorney experience matters most in terms of the arguments that repeat players and one-shotters present to the Court. Experience, then, is related to legal skills and is not necessarily a surrogate for reputation or visibility in the profession. The results for the status advantage variable indicate that applicants with higher status relative to the status of the respondents are more

flemming2.p65

71

01/04/2004, 2:24 PM

71

72

Tournament Rules and the Consequences of Institutional Choices

likely to have their cases heard by the Court. This finding is consistent with the previous analysis as well. The size of this variable’s coefficient remains the same in a model based on only the three litigant resource variables. Party status, therefore, aside from the kind of argument made in the factum and unlike attorney experience, appears to exert an independent effect on the Court’s leave decisions. It is well to recall, however, that governments are ranked higher than private parties in the construction of this variable. This suggests that governments, especially federal agencies, are favoured parties in the leave process. Even though regional interests deeply colour Canadian politics, the third model suggests that leave applications from British Columbia, Ontario, or Quebec are not significantly related to leave decisions. These provinces generate the largest volume of applications and for this reason were included in the equation. The negative sign for Ontario is contrary to expectations. Ontario is the largest, richest province in Canada and Toronto is its financial capital as well as the home of the “Bay Street lawyers” who work for the Seven Sisters and represent Canada’s major businesses and corporations. It was anticipated that the Court would be favourably disposed to applications from this province; the negative sign suggests otherwise. One possible explanation might be found in the high esteem in which Ontario’s appellate judges are held. There may be fewer legitimate grounds for appeal or insufficient justification for leave to appeal in the opinions by these judges, which dampens the need for judicial review by the Supreme Court. A positive and significant relationship between leave decisions and applications challenging decisions in the Federal Court of Canada was expected. The Federal Court has both a trial division and an appellate division; all appeals from the appellate division go to the Supreme Court. The Federal Court’s jurisdiction centres on administrative law and the Court exercises judicial review of federal administrative agencies. As this is a court with a national jurisdiction involving the federal government, it is surprising that the variable indicating whether an application involved a Federal Court case is not statistically significant. It is possible that the federal interests variable absorbed the impact of this variable. The fourth iteration adds the leave panels to the logit model; only the results for Panel 6 were statistically significant, and, as Table 4.2 indicates, the panel was negatively disposed toward granting reviews. The indicators for two other panels also had negative signs, while the rest had positive signs; the panels evidently differ in their disposition toward granting leave. Given the absence of statistical significance for these

flemming2.p65

72

01/04/2004, 2:24 PM

Tournament Rules and the Consequences of Institutional Choices

results, however, they should be treated cautiously. Overall, it appears instead that after other factors are taken into account, the leave panels contribute little to an understanding of how Canada’s Supreme Court selects cases for judicial review. Round Two: The Panels as Venues for Tournament Play

The conclusions from the first round of analysis may be premature, perhaps misleading. The models in this round and as displayed in Table 4.2 took the Court as a whole as its unit of analysis. At this level of analysis, after controlling for jurisprudential factors and the resources of litigants, the discrete impacts of the panels on the likelihood of leave being granted (with one exception) appear to be minimal. This apparent absence of differences in leave rates across panels is consistent with how applications are assigned to the panels. After being vetted by the staff attorneys in the Legal Services section (as it was then called), leave applications were assigned more or less randomly to the panels by the section’s supervisor. Applications from Quebec were one exception to this rule, as they were forwarded to a panel composed of the three Quebec justices during the period of this study. While the panels do not differ in how often they grant leave, is it possible that they differ in their selection criteria? For some idea of the changes that occur in the panels, Table 4.3 shows the panel composition and the dates when the panels were created for 1990-96. When Justice Lamer became chief justice in July 1990, he arranged the first set of panels shown in this table. The chief justice then shuffled them again after Justice Iacobucci’s appointment in January 1991. Justice Stevenson’s illness and subsequent resignation in June 1992 meant that some justices alternately served on more than one panel during Stevenson’s illness and until Justice Major’s appointment to replace him in November 1992. After January 1993, the panels remained fairly stable until Chief Justice Lamer reconstituted them at the start of the Court’s term in November. During this time, however, the chief justice created several special panels to review particular leave applications. One clear informal rule affecting Chief Justice Lamer’s selection of justices to serve on the panels was that one of the panels should include at least two of the three Quebec justices. Indeed, in November 1994, the chief justice joined the other two Quebec justices to create a fourth panel to review applications from Quebec involving the province’s civil law code. Since decisions on applications are governed by majority rule

flemming2.p65

73

01/04/2004, 2:24 PM

73

74

Tournament Rules and the Consequences of Institutional Choices

Table 4.3 Leave to appeal panels, 1990-96 Date panel formed July 1990 January 1991 February 1992 January 1993 November 1993 November 1994 November 1995 November 1996

Panel

Panel

Panel

Lamer, LaForest, L’Heureux-Dubé Lamer, McLachlin, Sopinka Lamer, McLachlin, Iacobucci Lamer, McLachlin, Major Lamer, Iacobucci, Cory Lamer, Iacobucci, Gonthier Lamer, Iacobucci, Gonthier Lamer, Cory, McLachlin

Wilson, Sopinka, McLachlin LaForest, Cory, Stevenson LaForest, Sopinka, Stevenson LaForest, Cory, Iacobucci LaForest, Major, Sopinka LaForest, Major, Cory LaForest, Major, Cory LaForest, Gonthier, Major

Gonthier, Cory, Stevenson Gonthier, L’HeureuxDubé, Iacobucci Gonthier, L’HeureuxDubé, Cory Gonthier, L’HeureuxDubé, Sopinka Gonthier, L’HeureuxDubé, McLachlin L’Heureux-Dubé, Sopinka, McLachlin L’Heureux-Dubé, Sopinka, McLachlin L’Heureux-Dubé, Sopinka, Iacobucci

Notes: The names of the justices from Quebec are italicized. In 1994, a fourth panel was established consisting of the three Quebec justices to review civil code appeals from Quebec. Source: Various editions of Supreme Court of Canada Bulletin of Proceedings.

and applications from Quebec are assigned to panels with either two or three Francophone justices, the joint effect of these rules was to limit Anglophone control over appeals arising from Quebec. To determine whether selection criteria varied from panel to panel or were the same for all of them, a single logit model was constructed of interaction terms between each panel and the jurisprudential, litigantresources, and jurisdiction indicators. For example, Panel 1 was entered as a term with the variables novel issue, and then lower court conflict, and so on until there were interaction terms between the panel and every one of the independent variables. This process was repeated for each of the other panels. The advantage of this approach is that it overcomes the problem of estimating ten separate models, each with an insufficient number of cases to estimate the parameter values of the twenty-three independent variables. The single-model approach with interaction terms preserves the advantages of having a large number of observations while at the same time permitting tests for panel-by-panel differences. The price of this approach is a complex model with over 200 interaction terms. As a matter of presentation and convenience, Table 4.4 shows the results for only those variables that were statistically significant at .05 in one-tailed tests.

flemming2.p65

74

01/04/2004, 2:24 PM

Tournament Rules and the Consequences of Institutional Choices

Table 4.4 Panel-by-panel model of leave to appeal decisions, 1993-95 Panel

Independent variable

1

Novel issue Conflict – lower courts Dissent – lower court Federal interests Error correction Fact-specific Frivolous Limited generality Experience advantage Queen’s Counsel advantage Status of party advantage Federal Court

2

3

4

5

6

×

×

× ×

×

×

7

8

9 10 ×

× ×

× ×

×

×

×

×

× ×

× ×

×

×

Notes: Parameters estimated in one model. Nagelkerke R2 = .607. Log likelihood = 433.589. × entered if p = .05 in one-tailed test.

The model has a Nagelkerke pseudo R2 of .607. (The Nagelkerke pseudo R2 is an estimated measure of association between variables that varies from 0 to 1, with higher values indicating stronger associations.) Although there was no convincing evidence for the hypothesis that leave to appeal rates vary from panel to panel, the results of this model indicate that the criteria used by the panels differ considerably. The factors associated with whether or not leave is granted differ from one panel to the next. Leave decisions for Panel 4, for instance, are significantly related to six of the factors used in this analysis, while none of the variables is significantly related to the decisions made by Panels 1 and 5. The institutional norm of processing leave applications through a set of panels raises hurdles for understanding the process. In general, the aggregated model suggests that leave rates do not differ substantially across panels after other factors are taken into consideration. The disaggregated model, however, suggests that the panels’ deliberations involve different criteria. Understanding the process from the perspective of the panels becomes a complicated task. The complications increase with the annual changes in the composition of the panels, as the changes may prompt entirely new combinations of factors in each of the panels. The result over time is that no single set of criteria establishes itself in the panels’ deliberations. In other words, the rules of the tournament vary by panel and over time with changes in the composition of the panels.

flemming2.p65

75

01/04/2004, 2:24 PM

75

76

Tournament Rules and the Consequences of Institutional Choices

Conclusion: Uncertainty as the First Rule of Play?

The Supreme Court of Canada does not move the nets around the hockey rink, hoping to fool players who hope to score goals by convincing the justices that they should grant leave to appeal in their cases, but its decentralized process of review does add uncertainty to the litigants’ quest for judicial review. The jurisprudential account tells a compelling story in which legal rules matter because of how the Court has organized itself. Moreover, the justices discourage organized interests from participating in the leave tournament; the effect of this practice in concert with the relative accessibility of the Court’s agenda depoliticizes the tournament and increases the importance of the formal or jurisprudential rules to the process. The Supreme Court of Canada selects appeals for judicial review through a process shaped by institutions and norms that differ in many respects from those in the United States. Compared with the American agenda-setting process, the selection of appeals for review in Canada is decentralized through the use of leave panels. Both courts nevertheless have something in common: their rules, procedures, and norms reflect their own particular interests, preferences, and needs. The three-justice panels in Canada’s leave process are devices adopted by the Court; they are not the product of Parliament, which had little or no hand in crafting the process. This means that the choice criteria in the Canadian agenda-setting process, namely, the prominence of jurisprudential considerations in the selection of cases for judicial review, whether by design or default, are an endogenous characteristic determined by the Court itself. The status of parties and whether or not their lawyers are repeat players influence the odds of leave being granted, but they take a backseat to the jurisprudential factors derived from the Court’s interpretations of the public importance criterion embedded in the Supreme Court Act of 1975. Placing leave decisions exclusively in the hands of several leave panels, however, appears to lead to different standards for leave among the panels. In sum, the tournament rules are the structurally induced consequences of how Canada’s Supreme Court has designed its leave to appeal process and the selection of cases for judicial review.

flemming2.p65

76

01/04/2004, 2:24 PM

5 The Court Game: Strategy and Consensus among the Tournament’s Judges

The tournament’s premier repeat players, of course, are the justices themselves. They are the only players directly involved in nearly 200 rounds of tournament play as members of the leave panels, and even more when they convene in conference. In short order, they become thoroughly familiar with the process and develop clear ideas as to why certain applications should be granted leave while others should not. They are the only players who know first-hand why their colleagues may choose to grant leave in some circumstances but not in others. Finally, they are the only tournament players who get to know the other repeat players, the lawyers and parties who regularly file leave applications. And, of course, only the justices know the rules of the game and have the power to change them to suit their interests. All of this makes it intriguing to ask whether there is an “agenda game” in Canada’s Supreme Court that is part of the tournament of appeals. The impetus for this question comes from evidence that an agenda game does exist in the United States. The Strategic Account of Agenda Setting in the United States

Is it possible that the selection of cases in Canada is linked to how the cases will be decided on their merits? Do justices on the leave panels anticipate the outcomes of cases that the Court might review, and do these anticipations influence their decisions to grant leave to appeal? Game theory distinguishes between “sincere” and “sophisticated” behaviour (Morrow 1994; Ordeshook 1992). Justices act sincerely when they vote for an alternative that they most prefer at the agenda-setting stage without regard for what may happen during judicial review. They act strategically or engage in sophisticated voting when they anticipate the outcome on the merits if the appeal is heard and then vote accordingly, which may mean voting against their preferences if they believe

flemming2.p65

77

01/04/2004, 2:24 PM

78

The Court Game

they will not attain their goals if the case is reviewed. In the agenda game on the US Supreme Court, when justices think they have the votes to win on the merits, they vote to grant cert; if they think they will lose, they vote to deny certiorari. Thus, strategic concerns can trump jurisprudential considerations or personal policy preferences when the justices decide what cases to decide (Perry 1991). Schubert (1959, 1962) developed a strategic account to explain the behaviour of American justices during the late 1940s and early 1950s in liability cases between railroads and their injured employees. He reasoned that rational judges would select cases that satisfied their policy preferences as long as they perceived that they had reasonable chances of success on the merits. As Schubert put it, the goal of justices favouring the railroad workers was “to induce the lower courts to decide a maximum number of cases favorably to the claims of injured workers” (Schubert 1962, 298). Justices accordingly voted to decline certiorari if they felt that the ideological balance on the Supreme Court was adverse to this goal and likely to jeopardize workers’ claims in the lower courts. Alternatively, a Court that was friendlier to workers’ appeals offered opportunities to expand workers’ protections in the lower courts, and thus the rational justice voted to grant certiorari. Schubert proceeded to specify the justices’ optimal strategies according to their goals and the size of the bloc or the voting strength of their preferred policies on the Court. For justices who were outnumbered and unlikely to succeed on the merits, a rational course of action was to vote against their preferences in order to block the Court’s review of the appeal. Schubert’s strategic account differs from studies that also assume that American justices are policy-directed but argue that the justices vote sincerely according to their policy views in order to correct what they perceive as “errors” in the lower courts (Brenner and Krol 1989; Provine 1980; Songer 1979; Ulmer 1972). When voting sincerely, justices cast their votes based on their views of the outcomes in the lower courts but they do not take into account what might happen at the merits stage. It is common for the US Supreme Court to overturn lower court decisions; it does so in roughly two-thirds of the cases it hears. What this means is that the justices routinely grant certiorari because lower court decisions differ significantly from the preferred doctrinal positions of at least four of the justices and perhaps a majority of the full court (Baum 1977). This bias toward error correction creates the backdrop for the certiorari game in the United States and affects how the justices play it. Strategic behaviour on the bench takes two forms: defensive denials and aggressive grants (Perry 1991). Defensive denials occur when a justice

flemming2.p65

78

01/04/2004, 2:24 PM

Strategy and Consensus among the Tournament’s Judges

fears that a lower court decision or existing precedent that the justice favours will be overturned or weakened if the Court grants judicial review. Under these circumstances, the justice votes not to grant certiorari as long as there are not enough votes on the merits to deny the appeal or to affirm the precedent. Caldeira and co-workers (1999) provide an example of this kind of strategic recommendation made by a clerk to Justice Thurgood Marshall that was found among Marshall’s papers: CA1 [First Circuit Court of Appeals] has ruled contrary to CA2 [Second Circuit Court of Appeals] on a number of claims. The lawyers are good, and the full range of potential arguments have been made. In the normal case, this would be a pretty clear grant. Here, though, I would deny. The rumor mill has it that SOC [Sandra Day O’Connor] has changed her vote in Hodgson and that the statute will be upheld in its entirety. If we can’t win in Hodgson, it’s hard to imagine the abortion case we could win. This case has the First Amendment dimension, which might help us with AAS [Antonin Scalia] or AMK [Anthony M. Kennedy], but I doubt it. Because every abortion case on which cert is granted creates a new opportunity to overrule Roe, I would deny on defensive grounds. Tactical judgments aside, the case is a grant.

Note that while there was inter-circuit conflict, which is part and parcel of Rule 10 and strongly correlated with granting certiorari, and despite Justice Marshall’s strong support for abortion rights, the clerk suggests a defensive denial to avoid the risk of the case being used to overrule Roe v. Wade. Aggressive grants are votes by justices to place cases with favourable or preferred outcomes on the docket with the expectation that the Court will affirm the decisions. What makes this vote strategic is that the justices take into account the odds of success. The factors affecting this calculus are the justice’s ideological preference in the case, the ideological predisposition of the Court if it votes on the merits, and the interaction between these two considerations (Caldeira et al. 1999, 562). The decision is not made solely on the justice’s ideological views (if it were, it would be a sincere vote) but in consideration of the ideological balance on the bench. Aggressive grants are risky decisions for American justices because the Court overrules lower court decisions more often than it affirms them (Benesh et al. 2002). The gamble for justices who vote to grant certiorari with the hope of affirming the lower court decision is therefore greater. Some researchers feel that aggressive grants

flemming2.p65

79

01/04/2004, 2:24 PM

79

80

The Court Game

are nonetheless more common than defensive denials (Brenner 1979; Boucher and Segal 1995), but their research did not include rejected writs of certiorari. When Caldeira and co-workers (1999) corrected this omission by studying all paid or civil applications for one year, they found that defensive denials were more frequent.1 The balance between these forms of sophisticated behaviour varies with the ideological balances of the lower courts and the Supreme Court. The Puzzle of Consensus in the Canadian Leave Tournament

The American justices make their decisions within a specific context that facilitates sophisticated behaviour. All nine justices participate in conference to pick appeals that the Supreme Court will review. They all sit to hear and decide the merits of appeals. They have intimate knowledge of each other’s policy preferences and voting proclivities. They can guess with considerable confidence whether appeals will be denied or affirmed. Compared with this information-rich situation, the institutional context of agenda setting in Canada inhibits obvious forms of strategic behaviour. Divisions among the justices, whether ideological or simply differences in views, clearly are another necessary precondition for strategic behaviour on a high court. If justices always agree on the merits of appeals, they need not worry about whether they will be on the winning or losing side, although unanimously decided appeals may be the results of negotiation and jockeying between the justices (Maltzman et al. 2000; Murphy 1973). Division within the US Supreme Court became a wellestablished norm beginning in the late 1930s or early 1940s (Epstein et al. 2001; Walker et al. 1988). Unanimous decisions now occur only about a third of the time, and dissents are common. The pattern of votes and formation of majorities and minorities generally reflect the justices’ liberalism or conservatism. The prevalence of dissensus, division, and ideological voting encourages strategic behaviour on the American Supreme Court. Various methods reveal the ideological structure underlying division on the American Court. The first is similar in spirit and methods to roll call analyses in legislatures (e.g., Poole and Rosenthal 2000) and focuses on the votes of the justices. The so-called attitudinal model, the reigning explanation of behaviour on the American Court, rests on sixty years of research into voting patterns in non-unanimous cases (Segal and Spaeth 2002). The attitudinal model rests most securely on scaling methods that reveal a distinctive and fundamental left/right, liberal/conservative dimension to voting on the US Court, especially with regard to civil

flemming2.p65

80

01/04/2004, 2:24 PM

Strategy and Consensus among the Tournament’s Judges

rights and liberties cases (cf. Ducat and Dudley 1987; Dudley and Ducat 1986; Hagle and Spaeth 1992). Grofman and Brazill (2002), employing multidimensional scaling, recently estimated the policy preferences of the justices based on their votes whether to affirm lower court decisions. They found that a single dimension of conflict explained some 80 percent of the variance in voting for fifteen natural courts between 1953 and 1991. Voting tendencies differ in Canada’s Supreme Court since its justices vote unanimously on the merits of appeals more often than they publicly divide and dissent. Between 1991 and 1995, for instance, the justices voted unanimously in 72 percent of the cases they heard on the merits (Supreme Court of Canada 2001). This level of agreement despite ups and downs has persisted for some time (McCormick 2000). This normative bias toward consensus, whatever its source, obviously limits opportunities for dissents. Relatively infrequent dissents could be random events, depending on the specifics of the circumstances and nature of the cases. It should come as no surprise, however, to learn that dissents are not random in Canada; instead, they emerge from differences between the justices, and these differences occurred before the advent of Charter litigation and the expansion of the Court’s policy role. Fouts (1969) analyzed the Court’s non-unanimous civil liberties and economic regulation decisions for 1950-60 and found that he could scale the two sets of decisions. This meant that Canada’s justices voted in ways consistently different from one another in these cases. Moreover, when Fouts combined the two scales, he was able to identify blocs of justices with regard to their overall liberalism or support for civil liberties and the federal government’s role in regulating the economy. Peck (1967, 1969) studied the votes cast by the justices from 1958 through 1966-67. He developed several scales dealing variously with criminal rights, civil liberties and economic regulation, support for provincial authority versus federal authority, “underdog” scales tapping whether the justices favoured plaintiffs in negligence litigation, unions, or taxpayers pitted against the government. In both studies, Peck concluded that blocs of justices voted according to whether they favoured private parties over the government, favoured the government over private parties, or were neutral between the parties. Tate and Sittiwong (1989) looked at the individual votes of the justices on the bench from 1949 to 1985. They found that the justices’ votes on the merits varied with their backgrounds, such as their regional origins, religious affiliation, political party, and prior judicial or political experience.

flemming2.p65

81

01/04/2004, 2:24 PM

81

82

The Court Game

These studies were precursors of voting patterns after the Charter and the Court’s increased discretion in setting its own agenda. Heard (1991) found what he termed “profound” differences among the justices who decided Charter cases between 1983 and 1989. Justices varied in their receptiveness to Charter claims and voted with other justices who shared their views. Equally important, he discovered that the success of Charter claims varied with the balance of views on the merits corams or panels reviewing the appeals; some panels ruled in favour of parties raising these claims while others did not, depending on which justices were assigned to the corams. A review of the first ten years of votes on Charter questions between 1982 and 1992 came to similar conclusions and confirmed the pattern of blocs and divisions (Morton et al. 1994). A third study focuses on the early Lamer Court and provides pertinent insights for this discussion. Ostberg and co-workers (2002), using factor analyses of justices’ votes in non-unanimous Charter cases between 1991 and 1995, determined that three dimensions structured the justices’ voting tendencies. The first factor dealt with the tension between communitarian and libertarian values. Figure 5.1 arrays the justices according to their factor scores on this dimension. Justice L’Heureux-Dubé anchors one end of the communitarianlibertarian scale while Justices Sopinka, Major, and Lamer hold down the opposite end. This dimension is interpreted as dealing with the relative weight justices give to the concerns of society and to groups who claim harm from discriminatory treatment. When the justices divide on the merits, their votes suggest that they each weigh differently the issues involved in Charter cases. Disagreements on whether to grant judicial review in the United States suggests that similar disagreements ought to surface when Canada’s Figure 5.1 Ideological differences between justices in Charter cases: Libertarian versus communitarian dimension LIBERTARIAN

Sopinka Major Lamer (.737)

COMMUNITARIAN

Iacobucci LaForest Cory

(.675) (.569)

(.151)

(.088)

(–.016)

McLachlin Gonthier (–.114)

(–.236)

L’HeureuxDubé (–.810)

Note: The varimax rotated factor scores for the justices on the libertarian-communitarian dimension are shown in parentheses. The figure is not drawn to scale. Source: Adapted from Ostberg et al. 2002.

flemming2.p65

82

01/04/2004, 2:24 PM

Strategy and Consensus among the Tournament’s Judges

justices decide whether to grant leave to appeal. If the justices vote sincerely when setting the Court’s agenda, the American example suggests that the justices on the leave panels should divide often. Provine (1980) documented the differences in individual voting habits of the justices in the Vinson and Warren Courts in the US as well as the alignments of justices that emerged as they decided to decide. For the later Burger Court, Epstein and Knight (1998, 58-59) found that 50 percent of the time at least one justice dissented when the conference denied certiorari; when the conference granted certiorari, there was also almost always at least one dissent. Dissents may be motivated by sincere or sophisticated concerns; they are also bargaining or negotiating efforts by justices. Justices William Brennan and Thurgood Marshall routinely published brief dissenting opinions whenever the Court refused certiorari in capital punishment cases. Through their dissents, the justices signalled their views to the litigating community and to the bench. A dramatic example is Justice Goldberg’s dissent in 1963, which was joined by Justices Brennan and Douglas, that urged the Supreme Court to address questions involving the death penalty after the Supreme Court declined to hear a case raising these issues. This dissent led to a series of cases challenging the constitutionality of the death penalty (Epstein and Kobylka 1992, 43). On other occasions, justices threaten to publish dissents to persuade their peers to change their cert decisions. In a case the Court at first declined to hear that challenged a state’s law outlawing sodomy, Justice White threatened to publish a dissent that pointed to conflicts and inconsistencies among the circuits as justification for granting certiorari. His ploy succeeded, and within a week he got the three votes he needed for cert (Epstein and Knight 1998, 61). These are not isolated instances of successful dissents. During the 1983 term, fifty-five individual dissents to initial conference decisions denying certiorari succeeded in attracting the four votes required to grant cert (Epstein and Knight 1998, 63). Dissent, then, is neither uncommon nor a waste of time in the American agenda-setting game. Things are different in Canada. Dissenting votes are rare, and published dissents hardly ever appear in the Bulletin of Proceedings. Of over 1,200 applications requesting leave to appeal submitted to the Supreme Court during 1993-95, only 30 were coded by this study as having a dissenting vote. Put another way, only 30 dissenting votes were cast out of a total of over 3,600 individual votes. In effect, leave decisions are virtually always unanimous. If the justices dissent roughly a third of the time on the merits of cases and there are stable patterns of voting disagreement and opposing alignments of justices, how is it that the agenda-setting

flemming2.p65

83

01/04/2004, 2:24 PM

83

84

The Court Game

process does not telegraph these disagreements ahead of time? How can this striking contrast with the United States Supreme Court be explained? The Agenda Game in Pure Form

To solve this puzzle, a rudimentary rational choice model and basic game theory notions are used. The game is presented in “pure form” in order to reveal its fundamental logic without empirical qualifications or complications; the latter will be introduced later to test the model. A rational actor with strategic inclinations looks ahead to the final outcome when the game has more than one sequence to it. Figure 5.2 presents the decision tree for a potentially dissenting justice and the sequence of events affecting this decision. If a panel votes 2:1 on a leave application, it could produce several scenarios of play in the agenda game. The first scenario involves a dissenting member of the panel hoping for a defensive denial by persuading four justices at conference to join him or her in overriding the panel’s decision to grant leave. A related scenario reverses the situation, and the dissenting justice prefers an aggressive grant. The last scenario involves Figure 5.2 Leave to appeal decision tree Leave Panel

Justice dissents

Conference

Leave granted

Justice does not dissent

Leave denied

Judicial Review Coram 5 A

R

7 A

9

R

A

R

Note: A = affirm lower court decision; R = reverse lower court decision.

flemming2.p65

84

01/04/2004, 2:24 PM

Strategy and Consensus among the Tournament’s Judges

a justice who is not on a panel but who disagrees with the panel’s decision and wishes to “appeal” this decision at conference for either reason. (Sincere dissents that reflect an intent to correct policy errors in the lower courts are not considered, since by definition the dissenter would ignore the consequences of this choice and not act strategically.) Justices rarely dissent, according to the votes published in the Court’s Bulletin of Proceedings. Why? Strategic considerations work backward from the expected final outcome. The outcome in this instance is whether judicial review of an appeal will affirm or reverse the lower court decision. Thus, the justice needs to know the probability of this outcome and whether this outcome is the preferred outcome. While the justice prefers one outcome to another, estimating its likelihood is made difficult by the fact that Canada’s Supreme Court does not always sit en banc when it hears cases. The fact that the Court typically reviews appeals in corams of varying sizes presents the dissenter with two problems. The first deals with whether or not the Court will sit en banc to hear the appeal if it is granted. The second involves the composition of the coram. As Benesh and her colleagues concluded with regard to American justices, “ease in prediction does matter” and strategic voting in the United States depended on anticipating the voting margins and the makeup of majority coalitions deciding the merits of cases granted certiorari (Benesh et al. 2002, 232). The situation for justices in Canada is even more problematic and uncertain. The first problem facing the dissenting justice is getting an idea of when particular appeals will be heard by corams of five, seven, or all nine justices. Coram sizes are not equally probable. Between 1986 and 1997, 29 percent of all cases were heard by panels of five justices, nearly 48 percent by panels of seven justices, and 23 percent by the full Court (Hausegger and Haynie 2003; cf. Greene et al. 1998, 115). The modal panel size was seven justices, but a dissenting justice still faces at least a 50-50 chance of being wrong without further information. Of course, the dissenting justice is concerned not with overall probabilities but with the odds of specific cases being heard by different-sized panels. If the dissenter guesses that the case will be heard by a panel of five or seven justices, the next question is which five or seven justices? There are 36 different combinations of seven justices on a nine-justice court; a panel of five justices has 126 combinations. Once the dissenter determines the probable size of the panel and its likely composition, he or she must then estimate the chances that the panel will affirm or reverse the lower court decision. A dissenting justice thus faces a three-pronged problem in uncertainty when trying to anticipate the outcome of the appeal if it is granted

flemming2.p65

85

01/04/2004, 2:24 PM

85

86

The Court Game

leave. With success so difficult to estimate, the stakes or rewards would have to be substantial for a rational justice to choose to dissent. The problem for the potential dissenter is exacerbated at conference, the second stage in the agenda game. Conference makes its decisions according to majority rule, in contrast to the Rule of Four in the United States. This means that the Canadian dissenter must gain the support of four other justices to override the panel decision. However, the two justices making up the panel majority need only three votes at conference to retain the status quo. The dissenter faces two difficulties. The first is identifying the non-panel justices willing to overturn the panel decision. The second problem involves the transaction costs of seeking the extra vote at conference needed to form a majority. A way of considering the dissenter’s problems at conference is to view them in stylized terms as a unidimensional game with nine players, each of whom has policy or value preferences that they want to maximize. Figure 5.3 shows a hypothetical conference with the justices located at different points along the continuum. Justices are presumed to vote to support or overturn a panel’s leave recommendation according to whether they feel that the appeal, if heard, will maximize their value preferences. These justices, it should be noted, confront the same uncertainties as the dissenting justice in anticipating the outcome if the appeal is reviewed. Because dissent necessarily implies dissimilar preferences or interests, the dissenting justice (PJd) is located at the opposite end of the continuum from the other panel justices (PJ1 and PJ2 ). As shown in Figure 5.3, the dissenter may seek either an aggressive grant or a defensive denial; it is simply a matter of where the case is located on the continuum. A rational justice will not dissent absent a non-trivial chance of persuading four non-panel justices to vote at conference to change the panel decision. It may also be presumed that PJ1 and PJ2 would not have voted to grant leave if they had thought that their recommendation would be overturned at conference. More specifically, if the two panel justices had correctly guessed the preferences of the median justice ( Jm ), Figure 5.3 Hypothetical agenda game with one panel dissenter

1

2

PJd

Aggressive Grant

flemming2.p65

86

3

Jm

PJ1 5

PJ2 6

Defensive Denial

01/04/2004, 2:24 PM

Strategy and Consensus among the Tournament’s Judges

their preferences with regard to the leave application would not be reversed. In order to win, the dissenting justice must have more accurate information about the preferences of Jm than the panel majority. PJd also pays higher transaction costs to play this round in the agenda game because of the need to attract one more justice than the panel majority. If the probability of a justice joining PJd varies directly with the distances between the policy value of the leave application and the individual preferences of the four non-panel justices needed for a majority, then the chances of success of PJd are likely to be slimmer than for the panel majority. While Figure 5.3 portrays this round of play in stylized form, its point should be clear. The odds are stacked against the dissenter. Unless the panel majority seriously misjudges the preferences of the other justices, there is little chance that the dissenter will win. Similar problems face a non-panel justice who disagrees with a panel decision and wants the conference to change it. The non-panel dissenter may disagree with the decision but unless it is a preference outlier (the policy location of the case in Figure 5.3 is far from the preferences of the median justice or the majority), the potential dissenter will remain silent. An added difficulty is that the non-panel dissenter needs to find four non-panel justices to overturn a panel’s choice. If the panel is unanimous, its members need only two more votes to sustain their decision. The non-panel justice, then, faces an even steeper uphill battle when challenging a panel recommendation than does a potential dissenter on the panel. Matters may improve somewhat if the two justices ally themselves, but the odds of success still remain discouragingly slim. Strategic behaviour in which the justice votes according to the likely outcomes of cases granted leave to appeal is hindered or discouraged by uncertainty and transaction costs on Canada’s Supreme Court. The upshot is that justices at the time they decide to grant leave to appeal make their choices in an institutional context that encourages consensus if not unanimity. Two additional considerations can be introduced at this point. Consensus may be furthered by the expectation that the chief justice may not assign a justice who voted to deny leave to the merits coram to hear the appeal. Of course, this means that the dissenter has in effect forfeited the chance to take part in making the decision or authoring the opinion. This latter opportunity is an important consideration because, unlike in the United States, the chief justice or most senior justice in the majority does not assign the writing of the opinion. Instead, a member of the majority usually volunteers to draft the judgment. If no volunteer steps forward, the chief justice will assign the opinion when it is

flemming2.p65

87

01/04/2004, 2:24 PM

87

88

The Court Game

unanimous; if the bench is split, the chief justice will pick writers for the majority and dissenting opinions. The consequences for dissenting in a leave decision, therefore, are not insignificant and stem in the first instance from the chief justice’s prerogative to appoint justices to appeals corams.2 If a justice feels that an appeal should not be heard, the chief justice could conclude that the justice should not be on the coram, since the dissenter claimed that the appeal lacked public importance. This strategic logic gains additional currency when it is recalled that the Court affirms more often than it reverses lower court decisions. Instead of seeking a defensive denial, a rational justice who for policy reasons wants to limit the reach of the opinion will vote to grant leave in the hope of being on the coram. Unanimity or constrained dissents may reflect these strategic considerations. In other words, consensus emerges in part from justices’ interests in being involved in the merits decisions. The flipside of the game is when a justice seeks judicial review when the panel recommends that leave be denied. The problem this justice faces arises from the fact that the Court declines roughly 85 percent of all leave applications, which means that the dissenter has to overcome the Court’s preference for denying leave. There may also be a conference norm against overruling panel decisions. All of the justices presumably have policy interests in seeing that their leave decisions are not regularly overturned at conference. Moreover, there are savings in time and effort at conference if the justices are not forced to review numerous appeals of panel decisions by dissenting justices. Thus, it would not be surprising if second-guessing panels at conference were frowned on, making things even harder for potential dissenters. In sum, the Canadian agenda game may well discourage defensive denials while fostering aggressive grants; as a result, dissents are infrequent at the panel stage. Empirical Consideration of the Pure Agenda Game

The pure agenda game highlights the uncertainties and costs facing justices who might respond strategically to panel decisions. There are empirical qualifications to this discussion that warrant attention, but they do not fundamentally alter the discussion’s conclusions. The place to begin is to briefly review the process once more as described by Justice Sopinka (Crane and Brown 2002) and by Greene and his colleagues (Greene et al. 1998) based on interviews with eight of the justices in 1994. The panel justices receive objective summaries and memoranda from the Court’s staff attorneys outlining the issues raised by applications; the summaries also include the attorneys’ recommendations on whether

flemming2.p65

88

01/04/2004, 2:24 PM

Strategy and Consensus among the Tournament’s Judges

Figure 5.4 Hypothetical probability distribution of leave outcomes Number of applications

Obvious denials

Low

???

Obvious grants

High Perceived public importance

to grant or deny leave. In addition, the panel justices have copies of the factums for their review. As Justice Sopinka noted, applications tend to fall into three categories: obvious dismissals, obvious grants, and those in between. Although this project had no data on the staff attorneys’ recommendations, some unknown portion of the consensus at the panel level undoubtedly reflects the recommendations. This does not mean that the recommendations are determinative, but it would be surprising if they did not influence the justices, if only by winnowing out those that clearly ought to be denied or granted while identifying those applications where the decision was not obvious. Figure 5.4 depicts a hypothetical distribution of leave applications using Justice Sopinka’s categorization. Leave applications range from obvious denials with zero probability of being accepted to those that are obvious grants, with a 100 percent chance of being accepted. The distribution is downward, sloping from left to right because most requests for judicial review in Canada (or the United States) are not heard; it can be presumed that the vast majority of applications fall short of the standards of public importance compared with the relative few that meet those standards. Located somewhere between these obvious denials and grants are applications about which honest differences over granting judicial review occur. The staff attorneys’ recommendations help define the width of this middle zone by identifying the obvious applications and offering suggestions for the

flemming2.p65

89

01/04/2004, 2:24 PM

89

90

The Court Game

more questionable ones. Recall that the Court grants leave to about 15 percent of the applications it receives. For a typical year, this translates into about eighty to ninety appeals, some of which are obvious grants while others are less obvious selections. The point of Figure 5.4 is that the two obvious categories probably reflect a “strong consensus” among the justices as to whether the applications warrant leave or not, while the middle category reflects a “weak consensus.” The staff attorneys’ recommendations may reinforce these categories and their underlying level of consensus. The panel justices vote by memorandum. Applications recommended for leave are placed on the “B” list and circulated among the justices. Justices who feel that particular applications should not be granted leave must notify the Court within a week of notice so conference can review the decision. Notice of applications denied leave is also circulated among the justices. If a justice disagrees with a panel’s intention to deny leave, the justice must ask that the application be placed on the “D” list; evidently thirty to forty applications are typically placed on the “D” list each year (Greene et al. 1998). When the justices meet in conference, they review only the applications placed on these two lists; all others are dismissed. These procedures narrow the range of applications that will be considered by conference. In effect, the conference pays attention to the applications located toward the right side of the distribution in Figure 5.4, and possibly to only those where there is a dissent. If applications are reviewed on a “fire alarm” basis when there is a dissent, then a substantial number are not formally reviewed by conference. The procedure ignores most of the decisions made at the panel stage and does not challenge the strong consensus at that stage. Greene and co-workers (1998, 111) present excerpts from two interviews that deal with the give and take between justices or the reciprocity that fosters at least weak consensus in setting the Court’s agenda: Greene: Is there a tradition that if one judge thinks that leave should be granted, then it is granted? Judge A: I think there is a generosity in that regard. If one judge really believed a matter to be important, I think the other two members on the panel would accommodate. What is the harm in that? I think the cooperative nature of the court during my time here is that if one wants it heard, I think he or she could probably prevail and have it ... ...

flemming2.p65

90

01/04/2004, 2:24 PM

Strategy and Consensus among the Tournament’s Judges

Greene: Would it ever happen that a judge would want a case heard and the panel would turn it down? Judge B: Yes. A judge may want a particular case heard, and so it goes on what’s called a “D List,” and then at the meeting of the entire court you can put forward your position as to why it should be heard ... Sometimes there’s a majority that agrees with you, and the case goes on the [B] list ...

These comments suggest at least three things. First, if a norm of reciprocity infuses relationships between the justices, it may explain why panel votes tend to be consensual in marginal cases that are neither obvious grants nor obvious denials. If a justice feels particularly strongly about a case, the other justices evidently accommodate the justice, which produces unanimous votes. At the same time, the uncertainty and costs of dissenting reinforce the attractiveness of this norm. Each justice presumably knows the risks and costs of trying to move a case onto or off the docket. Reciprocity emerges and is reinforced by the situation in which the justices make their choices. By accommodating each other’s wishes, the justices play a game of tit-for-tat. Repeated play in the agenda tournament based on this understanding fosters reciprocity (Axelrod 1984). Unanimous leave votes, therefore, emerge from institutional arrangements and understandings that create either weak or strong consensus. Strong consensus reflects in part the views of the justices about whether an application meets or fails to meet the public importance criterion. As Justice Sopinka pointed out, many applications are clear-cut denials or grants. Strong consensus is framed and supported by the recommendations of the staff attorneys who draft the objective summaries. Weak consensus also leads to unanimity, but reflects the strategic context created by the Court’s procedures and institutions. Given the hurdles blocking successful dissents, once a panel justice learns how the other panel justices are voting, there are compelling reasons to vote the same way. Unanimity in this instance arises from an institutionally induced weak consensus; it is weak because the votes cover over underlying differences between the justices. Judge B in the foregoing quotation refers to the willingness of the conference to listen to a justice’s argument as to why an application should be granted or denied leave after the panel decided the other way. Judge B’s comment illustrates the uncertainty of the process but also suggests that the conference will override the panel.3 The problem remains

flemming2.p65

91

01/04/2004, 2:24 PM

91

92

The Court Game

that the frequency of dissents and the responses of the justices when they meet in conference are not known. If thirty to forty applications from the “D” list are routinely granted by conference in a typical year, they would account for a sizable portion of the Court’s docket. This perspective puts the number of dissents in a different light. Rather than comparing the frequency of dissents with the total number of applications or votes on them, which tends to makes dissents a trivial phenomenon, perhaps the more appropriate comparison is with the typical size of the Court’s docket. From this angle, if the conference routinely overrode panel decisions, the agenda would change dramatically, at least in size if not in the mix of issues. Presumably, most appeals of panel denials are turned down to avoid crowding the Court’s agenda. Recall as well that the Court has little or no control over the number of appeals as of right that come to it every year, which probably reinforces the inclination of conference to support panels when they deny leave. Finally, reference questions, although few in number, consume considerable time and energy, adding a further constraint on the size of the docket. Reciprocity between panel justices may reveal itself in another way. Negotiations over the issues raised in leave applications could prompt the justices to agree to hear some cases they otherwise would have declined. Issues may be suppressed when the justices cannot agree that all or some should be heard if leave is granted. As one example, the leave panel of LaForest, Cory, and Major granted leave to a criminal case, “but only on the issues regarding the search, namely whether the power of search incident to arrests extends to vehicles and whether the delay between the arrest and the search was such that the search could no longer be considered an incident to the arrest.” Other issues raised in the application were ignored. In a tort dispute, this same panel granted leave “on the issue regarding non-delegable duty only.” There is no evidence that these were negotiated grants; they do illustrate, however, how suppressing issues in leave applications could resolve differences between justices and lead to unanimous votes on leave panels. Another possible qualification is that the agenda game in its pure form may overstate the uncertainty of the Canadian context. Data for this project indicate that members of the leave panels are almost always assigned to the merits corams. Out of 156 cases granted judicial review for which complete information is available, 119 or 70 percent of the corams included all three members of the leave panel.4 Thus, while selecting cases for judicial review, the justices are reasonably sure that they will hear the appeals. Moreover, five-justice corams are used most

flemming2.p65

92

01/04/2004, 2:24 PM

Strategy and Consensus among the Tournament’s Judges

of the time to hear criminal cases appealed as of right. As a consequence, guessing whether the Court will sit as a coram of seven or nine is easier for the justices. Setting aside the five-justice corams, the Court sat en banc for one-third of the cases and as seven-justice corams for the remaining two-thirds between 1986 and 1997. There is still a chance that the justice will make a mistake, but it is smaller than the one portrayed earlier. Moreover, the uncertainty of who will sit on the corams is reduced because the panel is very often part of the coram. The number of combinations of justices, excluding the panel members, for seven-justice corams is reduced to fifteen from thirty-six, which halves the uncertainty for panel justices trying to figure out who else will sit on the coram. The justices may also have ideas about how the chief justice decides the size of corams. According to one justice, “The chief justice [Lamer] does not strike panels of nine unless he feels that there is an issue of general national importance where the judges might have differing views” (Greene et al. 1998, 111). Chief Justice Lamer expanded a bit on how he struck panels during his interview with Greene et al. (1998, 115-17). He began by affirming the previous justice’s comment and then mentioned other considerations that affect the size and composition of the merits coram: If there is a possibility that the outcome of a case might be different with fewer than nine judges, I’ll do my best to strike a panel of nine judges. How do I know if there will be a division? First, my executive legal officer helps me to flag these cases. Also, I know my colleagues and I have a fairly good idea about what they are thinking on particular issues. I might ask what the other judges think about a particular issue ... I wouldn’t like to see a minority in the court impose its views on the court ... If I see that a case is strictly common law, I am inclined to take two of the Quebec judges off. I would not like to see the swing vote to be in the hands of a civilian [Quebec civil code] judge. Similarly, if a case is strictly Civil Code with no issue of general importance, I would of course include the Quebec judges, but I would not use all nine judges ... As I mentioned earlier, I try to avoid setting up panels of seven where there might be a four to three split ...

It is worth noting that since the panel justices almost always vote the same way on leave applications, the chief justice may assume that by including them on corams, the odds of division on the merits are reduced. In a sense, the institutional arrangements encourage a form of

flemming2.p65

93

01/04/2004, 2:24 PM

93

94

The Court Game

“bluffing” if justices vote to hear cases to ensure their places on the corams. Hausegger and Haynie (2003) studied the size and composition of panels for 1991-97. She found that the chances of the Court sitting as a panel versus en banc improved when there was a right of appeal (usually routine criminal cases), when the case did not involve a Charter claim or an intervener, and if the appeal came from Quebec. These factors can be seen as reducing the “general national importance” or public importance of the appeals, as the justice suggested earlier. Hausegger and Haynie also determined that assignments to merits panels greatly depended on how often the puisne justices agreed with the chief justice; whether it was a civil law appeal (if so, the Quebec justices were automatically assigned to the panel); and to lesser degrees with the seniority of justices, their expertise, and whether the non-Quebec justices came from the province where the appeal originated, in which case the chief justice matched the justice with the province (see also Hausegger and Haynie 2003). The Canadian justices do not work in the dark. The uncertainty underscored by the agenda game in pure form is mitigated by practice and experience. Nevertheless, compared with the situation in the United States, agenda setting in Canada occurs in a more uncertain institutional context that encourages consensus or at least discourages dissent. There is, however, a third alternative explanation for why panel votes are almost always unanimous. The chief justice could assign like-minded justices to the same panel. The empirical question becomes, did the chief justice create panels that were largely homogeneous in terms of voting and value preferences? Table 5.1 presents the information to answer this question. The voting preferences of the individual justices for the nine sets of panels during the three years in this study are drawn from McCormick (1998), who studied the merits votes for all reported decisions with oral arguments for the 1990-97 terms of the Lamer Court. The frequencies with which pairs of justices voted together on the merits were then ranked for each justice; these ranks are shown in Table 5.1. The scores should be read down and not across. For example, in the first 1995 panel, Justice Iacobucci was Chief Justice Lamer’s second most frequent partner while Justice Gonthier was the chief justice’s sixth most frequent. The rankings on each panel can be summed to get a measure of voting preferences and thus an overall indication of homogeneity on the panel. The minimum score is 9, indicating that the three justices most often voted together on the merits, while the maximum is 45 for those justices who voted least often together. As Table 5.1 indicates, the

flemming2.p65

94

01/04/2004, 2:24 PM

Strategy and Consensus among the Tournament’s Judges

Table 5.1 Leave to appeal panels and “preferred voting partners” on the merits 1995 Leave panels LAM

IAC GON

Lam x 4 Iac 2 x Gon 6 5 Vote score = 25

5 3 x

LAF

MAJ COR

LaF x 6 Maj 8 x Cor 3 3 Vote score = 29

6 3 x

L’H

SOP MCL

L’H x 8 Sop 6 x McL 3 7 Vote score = 38

6 8 x

1994 Leave panels IAC

COR

Lam x 4 Iac 2 x Cor 3 1 Vote score = 13

LAM

2 1 x

LAF

MAJ SOP

LaF x 6 Maj 8 x Sop 5 1 Vote score = 26

5 1 x

L’H GON MCL

L’H x 4 Gon 1 x McL 3 6 Vote score = 21

6 1 x

L’H GON

SOP

L’H x 4 Gon 1 x Sop 6 7 Vote score = 32

8 6 x

1993 Leave panels LAM MCL

MAJ

Lam x 7 McL 7 x Maj 4 3 Vote score = 32

4 7 x

LAF

COR IAC

LaF x 6 Cor 3 x Iac 2 1 Vote score = 19

6 1 x

Notes: 1 The “preferred voting partner score” indicates the degree to which the three justices on a leave panel vote together when deciding the merits of appeals. The score is the sum of the rankings. The minimum score is 9 when each of the three justices votes with the others on the merits more often than with any of the other justices. The maximum score is 45 and indicates that the three justices are least likely to vote with one another on the merits. 2 Italicized names indicate a change in the panel membership compared with 1993 panels. Lam = Lamer; Cor = Cory; Gon = Gonthier; Iac = Iacobucci; LaF = LaForest; L’H = L’HeureuxDubé; Maj = Major; McL = McLachlin; Sop = Sopinka. Source: Rankings of pair-wise voting tendencies are adapted from McCormick 1998.

panels are not especially homogeneous in terms of the justices’ voting tendencies. The leave panels were not filled with judges who voted together on the merits. In a nutshell, the absence of dissenting votes as recorded in the Bulletin of Proceedings does not appear to be due to homogeneous preferences on the panels. An extension and variation of this third alternative explanation is whether the panel members vote alike whenever they are assigned to the merits corams. We know that at least a weak consensus reigns at the panel stage. This consensus may extend only as far as the choice between granting and denying leave. Do the justices reported in the Bulletin of Proceedings vote alike if they are assigned to decide the merits of the cases to which they granted leave to appeal? More specifically, do panel justices vote together on merits corams even when the corams

flemming2.p65

95

01/04/2004, 2:24 PM

95

96

The Court Game

result in divided judgments? A negative answer would suggest that panel consensus is weak and confined to the question of granting leave. A positive answer suggests a stronger form of consensus, one that reflects the concerns of the panel justices over the substantive issues of the appeal, concerns that carry over to influence their votes on the merits. Out of 37 merits decisions made by corams without all three justices from a leave panel, 12 of the cases, or 32 percent, were divided judgments. In contrast, of the 34 non-unanimous decisions made by corams with all three leave justices, the panel justices divided in 14 of the judgments, or about 41 percent of the time. The percentages and numbers are not large, but the differences point to the weak consensus among the justices on the panels at the time they grant judicial review, a consensus that reflects the institutional character of the court game and a consensus that falters when the panel justices cast votes on the merits of the appeal and the direction of Canadian law. Conclusion: Uncertainty and Consensus in the Agenda Game

Consensus in the Canadian agenda game, this chapter argues, emerges from the institutions and procedures used by the Canadian Supreme Court to process its cases. The mere existence of uncertainty, however, is not the issue. Rather it is the variance associated with the guesses that justices necessarily make if they wish to be strategic actors that is substantially greater in Canada than in the United States. In simple terms, if the probability of overturning a panel decision is .5 and the chances of a favourable merits decision also is .5, then the overall probability of success is .25, which may be sufficiently high, given the “payoff” or rewards of success, for the justice to dissent. However, these odds are based on observations and experience, which incorporate error, and hence there is a variance or uncertainty about the accuracy of the probabilities. If the variances of the probabilities are high, uncertainty is exacerbated and the probabilities become highly fallible guides for deciding to dissent. In Canada, the variances of the probabilities regarding conference decisions and merits decisions are higher than in the United States. American justices necessarily make guesses about what will happen if certiorari is granted to an appeal, but the variance of their guesses is smaller because the context of choice is much richer in information. This chapter has developed an explanation for why Canada’s justices almost always come to a unanimous decision when they grant or deny judicial review. In other words, it sought to understand the nature of the process. This perspective slights another aspect of the agenda game. Is it possible that the Court’s game has a positive bias? That is, does the

flemming2.p65

96

01/04/2004, 2:24 PM

Strategy and Consensus among the Tournament’s Judges

game encourage justices to grant leave instead of discouraging them from doing so? The argument in this chapter was that justices decline to dissent in order to assure themselves a seat on the merits corams and a role in developing the judgments. The uncertainties and transaction costs of dissenting that are inherent in the institutional arrangements of the game provide further reasons for voting to hear cases that might otherwise be denied leave. Moreover, reciprocity among justices on the panels, perhaps the suppression of issues to gain agreements, plus the willingness of the conference to accede to at least some of the requests that an application be removed from the “D” list, also contribute to this positive bias. This does not mean that the entire game is tilted toward more rather than less judicial review. Evidence from the earlier chapters clearly shows that the Court limits access to its dockets if it can. Nevertheless, at the margins, when cases are in the grey area and are neither obvious denials nor obvious grants, the Court’s agenda game fosters decisions by the justices that place on the Court’s docket some cases that under different institutional arrangements would have been denied judicial review.

flemming2.p65

97

01/04/2004, 2:24 PM

97

6 Conclusion: Canadian Variations on American Themes

Do institutions matter? When the judges sitting on courts of final appeal no longer must review every appeal of lower court rulings, they are freed from spending time on questions that often barely rise above the parochial concerns of the unhappy, losing litigants. With discretion to decide what to decide, however, being free to choose also means being free to choose how to choose. The judges must design procedures, draft rules, and develop norms as they sift through stacks of requests for judicial review, picking out ones that they feel warrant attention while dismissing many others. Only the judges, of course, will know the reasons for their choices. What difference, then, do the rules, norms, and routines developed by high courts make in terms of the character of the process and its outcomes? A Summary of the Findings

American interest in the Supreme Court’s agenda-setting authority began mostly with Felix Frankfurter’s annual Harvard Law Review assessments of how the justices implemented the Judiciary Act of 1925, or “Judges’ Bill,” that broadened the Supreme Court’s latitude to grant certiorari (e.g., Frankfurter and Landis 1993). The future Supreme Court justice’s book The Business of the Supreme Court, published in 1928, remains to this day an excellent history and analysis of the evolution of the Court’s jurisdiction. During the 1950s, law professors and social scientists took up Frankfurter’s example and resumed the exploration of how the Court docketed its business. Since then, a steady stream of studies over the years has produced a deep and sophisticated literature in the United States that has no parallel in other countries. Surveys of Canada’s leave to appeal process, for example, first appeared in 1982 in Supreme Court Law Review, an annual overview of the Court’s opinions, and the Review has continued to publish these surveys. They provide

flemming2.p65

98

01/04/2004, 2:24 PM

Canadian Variations on American Themes

useful descriptive information about the leave applications such as the issues raised, the court and province of origin, and the proportions of applications granted leave, but have limited analytical value. This book is the first systematic investigation into how a non-American court of final appeal selects appeals for judicial review. It is also the first study to draw on the extensive American literature with the purpose of replicating its major approaches or accounts to see whether they provide insight into the dynamics of another high court’s decisions, with the added bonus of checking whether or not the way the US Supreme Court grants judicial review is unique. The evidence from Canada reveals that the litigant-centred account has limited utility in solving the puzzle of how Canada’s Supreme Court grants judicial review. A stratified community of lawyers with repeat players at its centre exists, but these players do not dominate the leave to appeal process; they are no more successful than attorneys with limited or no experience before the Supreme Court. Moreover, attorneys awarded the title of Queen’s Counsel, even though several justices also were QC, had no edge over their plebeian adversaries. Nor were lawyers from the Seven Sisters or other large, prestigious firms at an advantage. As might be expected, parties were also stratified, but asymmetric matchups of unequal parties mattered only when governments were the other party. A third aspect of the American litigant account is the presence or absence of amicus curiae briefs filed by organized interests. Organized interests are visibly and actively involved in the American agenda-setting process; they were nowhere to be found when Canada granted judicial review.1 The justices, while encouraging interveners to come forward at the merits stage, discourage them from stepping into the leave to appeal process.2 An important component of the litigant-resources account in the United States is thus missing in Canada, a component that gives the account an overtly political dynamic as issues before the US Supreme Court and group interests interact over time. The other American accounts show more promise, although they must be adapted to the Canadian situation. The jurisprudential account offers a persuasive story of how Canada’s justices set their agenda. This study is the first of its kind to build on and to test Perry’s illuminating discussion (1991) of this account in the United States, an account that is more elaborate than earlier efforts aimed at measuring the impact of legal factors, such as lower court conflicts, on whether or not certiorari is granted. The Canadian version of this account, though, takes a couple of unexpected turns. The jurisprudential account becomes the preferred choice virtually by default, since the Canadian agenda-setting process

flemming2.p65

99

01/04/2004, 2:24 PM

99

100 Conclusion

shrinks and diminishes the role of parties and their attorneys in addition to shutting out the participation of interest groups. The Canadian jurisprudential account would be a simple story if it were not for the decentralized way in which the Supreme Court allocates responsibility for deciding when and to whom to grant judicial review. Both the use of three-justice panels and the periodic rotation or reassignment of the justices creates a shifting patchwork of jurisprudential practices across the panels, each of which stresses some rules more than others as they go about deciding the business of Canada’s Supreme Court, which in turn makes the evolution of a stable set of agenda-setting norms problematic. The strategic account also takes a different turn in Canada that distinguishes it from the American version. The American story pivots around the ability of Supreme Court justices to predict how each of them will vote on the merits of the appeals they are considering for judicial review. If their voting intentions are shaped by sophisticated or strategic considerations, the American en banc tradition supplies the justices with the information they need to forecast or anticipate the outcomes of cases so that they can vote accordingly on whether to grant review. The strategic account points out that agenda-setting and judicial review processes are intertwined; they are reciprocal influences on each other as votes and anticipations of votes at one stage affect the other, and vice versa. The absence of an en banc tradition and the use of panels at both stages in Canada complicate this relationship and add a large dollop of uncertainty to it. A justice who disagrees with a leave to appeal decision and perhaps has strategic motives has the tough task of figuring out what will happen at conference, whether the needed votes can be collected, and then guessing how the votes will fall if the appeal is heard. The stakes for making a mistake, at least in the abstract, are substantial if dissenting reduces the chances of sitting on the merits coram and authoring an opinion. The consequence of these institutional realities is that Canadian justices, despite the ideological differences that divide them, vote unanimously almost all the time on the leave panels as they decide to decide. It may also be the case that the institutionally induced unanimity fostered by this process incorporates a bias toward granting judicial review to leave to appeal applications that fall between obvious grants and obvious denials. The Supreme Court conducts the tournament of appeals according to its own rules, with two qualifications. The first is a reminder that Parliament stipulated that leave be granted when an appeal has “public importance.” The operative meaning of this criterion, of course, was left in

flemming2.p65

100

01/04/2004, 2:24 PM

Canadian Variations on American Themes

the hands of the justices. The second qualification deals with the Court’s reliance on three-justice panels in the leave process. In 1956, Parliament amended the Supreme Court Act to eliminate the option that a single judge in chambers could grant or deny leave to appeal after a justice came under heavy critical fire for denying leave in a high-profile murder case (Crane and Brown 1996, 6). This amendment prompted the use of panels, although the number of justices on the panels was the Court’s decision. The justices could try other arrangements, including adopting the American en banc norm instead of continuing the panel tradition. One reason for not making this change is that as a matter of consistency it could require a move to en banc courts at the merits stage as well. This move would also eliminate the Court’s flexibility when assigning justices to corams that vary in size with the importance of the appeals. Not all appeals on the plenary docket warrant the attention of all nine justices. Corams or panels are common among high courts around the world. It may be the American norm of sitting en banc whenever judicial review is granted that is the exception. There are two other reasons for not changing Canada’s version of the panel system. The panels divide the labour of processing the leave requests among the justices and reduce the total amount of time the bench as a whole spends on them. A switch to en banc review could be considered an inefficient option compared with current practice. A more important reason is that en banc proceedings would mean that the three Quebec justices would be in the minority when civil code applications from their province are reviewed. This would create a delicate political problem for the Court. Quebec applications would require the votes of two non-Quebec justices to form a majority, and the Quebec justices could be outvoted on civil code applications. If these applications were handled separately by a panel of the three Quebec justices while all other applications were reviewed en banc, it would mean that civil code applications would need a minimum of two votes for leave to appeal while applications from the rest of Canada would require five votes, an equally unattractive, if not untenable, option. For this reason alone, the norm of three-justice panel reviews is unlikely to be challenged.3 The justices’ institutional choices have deeper ramifications than first meets the eye, particularly when compared with agenda setting in the United States. The process in Canada is distinctively jurisprudential because of how the justices play the tournament of appeals. The Court has also depoliticized the Canadian appeals tournament in the sense of discouraging the involvement of organized interests in the tournament,

flemming2.p65

101

01/04/2004, 2:24 PM

101

102 Conclusion

although it may also reduce the amount of information the justices have available to them regarding the public importance of leave applications. If interveners were more commonly involved in the leave process, they might provide different perspectives on the policy and legal questions raised by appeals. As it is, the issues framed by the applicant’s lawyer and the respondent’s rebuttal are interpreted and evaluated by a Legal Services staff attorney and three panel justices. Potential interveners in the tournament must wait until after leave is granted before filing a motion asking for permission to intervene that is heard by a single justice. The Canadian Supreme Court prefers to control who appears before it, and the Court selectively grants intervention according to its assessment of whether an intervener can provide a “fresh perspective” on those questions posed by the appeal for which the Court granted judicial review.4 Moreover, interveners are granted opportunities to make oral arguments at the Court’s discretion when it hears the case. This presumably is one reason why the Court discourages interveners in the leave tournament. There is the option of allowing amici curiae, or friends of the court, to file factums regarding leave, although the Court has restricted this option mostly to the appointment of counsel at public expense to represent litigants without counsel or to present arguments on particular issues raised by interveners but not addressed by the parties in cases (Crane and Brown 2002, 373-76). Whatever the reason, it is striking that the organized interests and groups that in the early 1980s mounted a public campaign to convince the Supreme Court to allow them freer access as interveners at the merits stage either have not demanded similar access to the agenda-setting process or have failed to persuade the Court to relax its rules (Brodie 2002). Canada Compared with Several Other High Courts

An account of agenda setting that rests on a replication in one high court of hypotheses drawn from another high court may seem like fragile ground for generalizations. It may help if a few short, admittedly non-random thumbnail sketches of how other high courts grant judicial review are introduced to suggest that the Canadian process is not unique, although the American style of granting judicial review may be. •

flemming2.p65

Australia: The seven-judge High Court sets its own agenda through panels consisting of two justices who conduct oral hearings to determine whether special leave to appeal should be granted. (Special leave is analogous to Canada’s leave to appeal; until recently, Canada allowed

102

01/04/2004, 2:24 PM

Canadian Variations on American Themes

•

•

•

•

flemming2.p65

for oral hearings, although they had fallen into disuse by the time of this study.) In 1996-97, the panels heard 268 requests for special leave and granted leave to 23 percent of them (High Court of Australia 1998). Appeals are heard by corams of three, five, or seven justices (Blackshield et al. 2001). Germany: The constitutional court has no option but to review statutes if one-third of the members of the federal legislature request it. In this respect, Canada and its reference procedure is similar to Germany. Special admission is required, however, for all other requests or constitutional complaints. Panels consisting of three judges consider these requests and turn down roughly 90 percent of them; the volume of requests is substantially larger than in Canada, though (Alexy and Dreier 1997, 18-21). Finland: In 1996 the Supreme Court had twenty-one judges and was divided into two sections that rotated the responsibility for dealing with leaves to appeal and hearing appeals every two months. An appeal can be heard only at the discretion of the Supreme Court, which follows three rules: (1) the appeal has precedential value or consequences for other cases or it will assure uniformity of the law; (2) the request highlights procedural errors in the lower courts; or (3) the case has special importance for the party. Decisions to grant or deny leave are not explained. The legal importance of the appeals rather than their political, social, or economic consequences appear to matter most; as a result, the proportion of procedurally based appeals tends to be high. The volume of requests during 1993-95 was about five times greater than in Canada; the acceptance rate ranged between 7 and 9 percent (Aarnio 1997, 67-71). Norway: The Norwegian Supreme Court has nine judges. It conducts hearings through five- and three-judge panels, the latter dealing with less important substantive or procedural matters, not unlike the Canadian arrangement. The court has discretion to deny leave but the decision must be unanimous. Three-judge panels review requests according to their “general significance” and the panels do not explain their decisions (Eng 1997, 189-92). Spain: Access to Spain’s appellate courts is unimpeded. The Supreme Court, the court of final appeal in non-constitutional matters, with over ninety judges, hands down roughly 25,000 decisions a year. However, sections or panels of three judges in the separate Constitutional Court may reject any individual complaint if the judges unanimously agree that a precedent exists that renders the claim ill-founded (Miguel and LaPorta 1997, 263).

103

01/04/2004, 2:24 PM

103

104 Conclusion

•

Sweden: The Supreme Court with sixteen judges decides between 130 and 160 cases a year. Panels of three judges or perhaps a single judge determine whether to grant certiorari. The rules stipulate that certiorari can be granted only when there is need for a general ruling that will set a precedent or “special reasons” exist, as when serious mistakes were made in the lower courts (Bergholtz and Peczenik 1997, 293-94).

These sketches suggest that Canada’s institutional arrangements with respect to granting judicial review are similar in basic outline to those of other high courts. It may also mean that the jurisprudential account may satisfactorily explain how these courts select appeals for review if adjustments are made for the common law elements currently embedded in the jurisprudential account. Speculations on the Stability of Canada’s Tournament

How stable are the features of the Canadian tournament of appeals? Mounting requests for judicial review might lead to changes in the tournament. One feature of the tournament is the relatively low volume of play and the relatively high success the players have in reaching the Court’s docket. Although the number of leave applications has grown over time, the volume has not yet reached a point where anything more than tinkering at the margins of the process is needed.5 What is significant about the volume of applications is that it is an exogenous, external force over which the Supreme Court has little control. In the United States, the growing number of certiorari petitions has led inevitably to shrinking proportions of petitions being granted. In 1926, the year after the Judges’ Bill, the US Supreme Court granted review to 20 percent of the 586 paid certiorari petitions it received. As the petitions grew more numerous, the grant rate slowly declined over the years until 1968, when it dropped below 10 percent for the first time as the Court considered 1,255 petitions. By 2001, the grant rate slumped to 4 percent of 2,210 petitions. Cert was granted to 14 percent of a handful of unpaid petitions in 1935 but dropped to 2-4 percent within ten years. By the time the volume reached nearly 7,000 petitions in 2001, the Court granted cert to a mere 0.1 percent of the petitions, or one out of a thousand (Epstein et al. 2003, 68-71). In comparison, in 1970, five years before Parliament amended the Supreme Court Act to restrict the right of appeal and expand the discretion of the Canadian Supreme Court to grant leave, the Court received 158 applications for leave to appeal; slightly over 25 percent were granted leave (Flemming

flemming2.p65

104

01/04/2004, 2:24 PM

Canadian Variations on American Themes

2000, 47). In 1991, the number of requests for leave reached 480, of which 17 percent were granted, and by 2001, when the volume reached 668, the acceptance rate had dropped to 11 percent (Supreme Court of Canada 2001). While by American standards Canada’s Court has hardly been deluged by requests for judicial review, access to Canada’s plenary docket has narrowed and grown more restrictive as requests for leave have risen. The significance of this change is that with time it may also affect the structure or hierarchy of Canada’s Supreme Court bar. It was suggested earlier that the relatively high access that applicants have to the Court’s docket meant that the bar was more diffuse than its American counterpart; Canadian attorneys are more likely than their peers in the United States to have argued the merits of their cases before the Court. No inner core of repeat players appears to dominate the bar, nor do these players win the tournament more often than players on the periphery of this professional subcommunity. In 1964, the American lawyer and a repeat player in his own right Eugene Gressman, co-author of Supreme Court Practice, then in its third edition, wrote (Gressman 1964, 762): For the most part, there is no organized Supreme Court Bar with experts attuned to the problems of the [United States Supreme] Court and conversant with the sophisticated considerations relevant to the selection of cases for review. Practitioners before the Court are highly individualized, some of them invoking its processes but once in a lifetime.

In 1964, the US Supreme Court granted cert to 11 percent of the paid petitions and 2 percent of the unpaid petitions; the volumes were 1,041 and 1,170, respectively (Epstein et al. 2003, 69). Within twenty years, according to McGuire, a cohesive, exclusive, and experienced Supreme Court Bar had emerged. The question is whether a similar process will occur in Canada where increasing competition in the tournament for access to the plenary docket will establish an elite of repeat players with significant influence over the leave to appeal process. It is hazardous to speculate on what political events might occur that would force the justices to change how they play the tournament. In the introduction to this book, the Court’s decision regarding the rights of same-sex couples in Egan v. Canada was used as an example of how the Court has become a lightning rod of conservative criticism. Other cases and decisions could easily be added to compile a lengthy list of controversies prompted primarily by conservative politicians, commentators, and scholars. While ordinary Canadians sometimes agree with

flemming2.p65

105

01/04/2004, 2:24 PM

105

106 Conclusion

these criticisms in specific instances and cases, they generally give the Court high marks and continue to support it. As two observers of public opinion trends in Canada put it, “there is no evidence that a groundswell of opposition to judicial authority is in the offing” (Fletcher and Howe 2001, 290). Nor is there any sign that the Supreme Court’s decisions to grant judicial review or the way it makes them has drawn the ire of the Court’s skeptics. As noted earlier, a contretemps nearly fifty years ago, at a time when the Supreme Court’s stature was far from what it is now, prompted Parliament to intrude into the Court’s internal procedures. Conservative criticism of the Court’s decisions, both led and magnified by the National Post and the Canadian Alliance political party, could reach a point where the Court might reconsider how it grants judicial review. However, a change in priorities is more likely than a revamp of how they are implemented, and far easier to effect as well. If the Canadian Alliance alone or in a coalition were to form a government, the Supreme Court could simply avoid certain issues by denying leave to appeal. The same option, of course, is open to the Court should the Liberals grow disillusioned and begin to question the Court’s sway over public policies. “In the land of certiorari ... law provides precious little constraint on judicial action,” an American law professor lamented recently after going over the history of the Judges’ Bill in the United States (Hartnett 2000, 1720). He complained that aside from lower court conflicts, the judges have refused to enunciate and elaborate the principles informing their decisions when they grant judicial review. This book shows that basic common law factors and policy considerations clustering around the notion of “public importance” can explain how Canada’s Supreme Court sets its agenda even though it has been as reticent about the matter as its American neighbour. The particular institutional choices that create the tournament of appeals in Canada heighten the importance of jurisprudential considerations while increasing the uncertainty of their application even as they mute the politicization that pervades certiorari in the United States. There is, of course, a more fundamental form of politics that is inescapable from the decision to grant judicial review. Canada’s Supreme Court can refuse to review cases and does. The administrative justification for this authority should not obscure the power that comes from the exercise of this discretion. Agenda-setting authority in a tournament of appeals constructed by the justices in which they are the key players augments and bolsters their impact on public policies and on the day-to-day concerns of government officials, Parliament, and Canadian citizens.

flemming2.p65

106

01/04/2004, 2:24 PM

Notes

Chapter 1: Judicial Review and Agenda Setting 1 Newfoundland Association of Public Employees v. R. (2002 NLCA 72). 2 Erin Anderssen, “Ottawa to Enshrine Same-Sex Rights,” Globe and Mail, 20 January 1999. 3 Robert Fife, “Right Wants Power Back from Ottawa,” National Post, 22 February 1999. 4 A discussion of the procedures affecting the writ of certiorari in the United States can be found in Stern et al. (1993). Except for differences in legal details, the writ of certiorari for present purposes is directly analogous to the requests for leave to appeal in Canada. 5 An omnibus bill passed by Parliament in 1996 included a provision placing some limits on “appeals as of right” by giving the Court the latitude to dismiss criminal appeals based on lower court dissents on the law after an oral hearing. 6 The Supreme Court Act eliminated monetary thresholds on the Court’s jurisdiction and required leave to appeal in civil cases. In criminal cases, the right to appeal was retained in criminal cases involving indictable offences when either a court of appeal overturned an acquittal or a court of appeal divided on an issue of law. Crane and Brown (2002) provide a detailed legal discussion of the process. Russell (1987, 344-49) gives the political background. 7 Rules of the Supreme Court of the United States (adopted 27 January 2003, effective 1 May 2003), . 8 One indication of the difficulties in deciphering what “public importance” means to the Canadian legal community is that during interviews with attorneys and clerks, “national importance” was often used in place of “public importance.” The two phrases are not necessarily the same, and some attorneys would correct themselves. 9 In addition to federal references, reference questions from the provinces also come to the Supreme Court, but as appeals as of right (Hogg 1996, 211). 10 John Geddes, “Civil Cases to Get Top Priority in SCC, Says New Chief Justice,” Law Times, 9 July 1990. Chapter 2: Tournaments and Stratification of Canada’s Supreme Court Bar 1 All information pertaining to the firms and lawyers discussed in this introduction was drawn from their websites as of November 2002.

flemming2.p65

107

01/04/2004, 2:24 PM

108 Notes to pages 20-88

2 The lead author of this book, Robert L. Stern, was Acting Solicitor General and a member of the committee formed by the Supreme Court when it revised its rules in 1995. The second author, Eugene Gressman, now deceased, also was a Mayer, Brown partner. 3 Tony Mauro, “Practitioners’ Views of the High Court,” Legal Times, 14 August 2002. 4 It is worth noting that the US practice book was first published nearly forty years before the Canadian book made its appearance, an indirect indicator that the emergence of sufficient demand and the need among lawyers for a systematic presentation of the rules of Canada’s Supreme Court and explanation of its procedures lagged behind the United States. 5 I wish to thank Kevin McGuire for giving me permission to use his survey questionnaire for this study. 6 It should be noted that the provincial and urban locations of the survey respondents are representative of the leave attorneys. 7 Tony Mauro, “Supreme Advocacy: Building a Better Advocate,” The American Lawyer, October 2002. 8 This incident is reminiscent of the “orgy of patronage appointments” to Canada’s courts made by Prime Minister Pierre Elliott Trudeau in 1984 before he retired from government (Simpson 2002). 9 It bears remembering that CPAC televises oral arguments in Canada’s Supreme Court, and even though the number of viewers is typically small, some cases draw considerable attention from the public. Chapter 3: Litigants, Lawyers, and the Tournament of Appeals 1 Rex Lee and Erwin Griswold are former Solicitors General. Lawrence Tribe is a noted constitutional scholar and litigator before the Supreme Court. 2 It should be noted that previous studies compared the aggregated win rates of the parties, which are not direct measures of their resources or status but the outcomes of these factors. 3 This proportion was calculated using the judgments published at , after eliminating the cases heard as appeals as of right for the three-year period. 4 I am grateful to Gregory A. Caldeira for making this point as obvious to me as it was to him. It is also possible, however, that this effect arises only when government attorneys are included in the analysis. A later model in this chapter focusing only on private attorneys fails to show a significant relationship between leave decisions and the disparities in attorney experience. Chapter 5: The Court Game 1 Whether justices are more likely to behave strategically or sincerely when voting to grant certiorari remains a contested issue, as indicated by the conclusion of Segal and Spaeth (2002) that justices’ policy preferences regarding affirming or reversing the lower court overwhelm their apprehensions about how their colleagues will vote on the merits. 2 To anticipate the empirical side of this argument, which follows in the next section, the price of forfeiting the chance to author the opinion is not insignificant. Of the 119 cases heard by corams that included members of the leave

flemming2.p65

108

01/04/2004, 2:24 PM

Notes to pages 91-104

panel, 72 of the opinions, or 60 percent, were authored by a justice who voted on the panel to grant judicial review. 3 Nevertheless, the norm governing whether the conference will override the panel remains ambiguous. In a 1993 speech, Chief Justice Lamer stressed that the panels make the final decision even if the conference discusses whether to add or delete a case from one of the two lists (Crane and Brown 1996, 32). In his 1997 speech on the Court’s procedures, Justice Sopinka makes the same claim. 4 This understates the odds of panel members being on merits corams because at least two members were assigned to 26 of the other 37 corams. Chief Justice Lamer, who assigned justices to the corams, most often excluded himself (9), then LaForest (6), L’Heureux-Dubé (4), Sopinka (3), and Cory, Gonthier, Iacobucci, and Major (1 each). Only McLachlin sat on every coram for which she was also a panel justice. Chapter 6: Conclusion 1 Canada’s justices could anticipate the involvement of interveners if they were involved in the appeals heard by the lower courts. Using a list of leave applications from Ontario, published judgments of the Ontario Court of Appeals were checked for interveners; very few instances of intervener participation could be found. The justices would have to rely on the nature of the claim and history to predict when interveners might appear if they hope to use this cue in their decisions to grant judicial review. 2 According to the Court’s rules prior to their revision in 2002, “No person shall intervene on an application for leave ... unless ordered by a judge prior to the hearing.” The new rules state: “Any individual interested in an application for leave to appeal ... may make a motion for intervention to a judge.” This revision may signal a change in the Court’s views about interveners in the leave process. 3 Canada’s justices could adopt the “Rule of Four” used by the US Supreme Court, and process all leave applications en banc using this voting rule, which allows a minority of four justices to grant leave. The problem would still remain that Quebec justices, if they voted as a bloc, would require a fourth justice to hear civil code cases, which is not the case under the Court’s current procedures. 4 Governments can intervene as a matter of right when the Court reviews appeals involving constitutional issues. Other parties in either constitutional or nonconstitutional cases must file motions to intervene with the Court, which generally favours the requests of public interest organizations over those made by private interests (Crane and Brown 2002, 298-307). As a consequence, the mix of interveners in Canada differs from the profile of amici curiae in the United States, where business and trade associations are common (Flemming 2000, 50). 5 During 1993-95, for instance, the justices’ clerks were removed from the leave process, presumably as an efficiency measure, since their involvement after the staff attorneys’ review added another step to the process and presumably slowed it down.

flemming2.p65

109

01/04/2004, 2:24 PM

109

References

Aarnio, Aulis. 1997. “Precedent in Finland.” In Interpreting Precedents: A Comparative Study, edited by D. Neil MacCormick and Robert S. Summers, 65-101. Aldershot, UK: Ashgate Dartmouth. Alexy, Robert, and Ralf Dreier. 1997. “Precedent in the Federal Republic of Germany.” In Interpreting Precedents: A Comparative Study, edited by D. Neil MacCormick and Robert S. Summers, 17-64. Aldershot, UK: Ashgate Dartmouth. Antoniadis, Judy. 1996. The Canadian Law List. Aurora, ON: Canada Law Book, Inc. Armstrong, Virginia, and Charles A. Johnson. 1982. “Certiorari Decision Making by the Warren and Burger Courts: Is Cue Theory Time Bound?” Polity 15: 141-50. Arthurs, Harry W., Richard Weisman, and Frederick H. Zemans. 1987. “The Canadian Legal Profession.” American Bar Foundation Research Journal 1986: 447532. –. 1988. “Canadian Lawyers: A Peculiar Professionalism.” In Lawyers in Society: The Common Law World, edited by Richard L. Abel and Philip S.C. Lewis, 12385. Berkeley: University of California Press. Atkins, Burton M. 1991. “Party Capability Theory as an Explanation for Intervention Behavior in the English Court of Appeal.” American Journal of Political Science 35: 881-903. Axelrod, Robert. 1984. The Evolution of Cooperation. New York: Basic Books. Barclay, Scott. 1999. “Appealing (but Not Necessarily Winning) to Improve Your Social Status.” Law and Policy 21 (October): 427-42. Baum, Lawrence. 1977. “Policy Goals in Judicial Gatekeeping: A Proximity Model of Discretionary Jurisdiction.” American Journal of Political Science 21: 13-35. Benesh, Sara C., Saul Brenner, and Harold J. Spaeth. 2002. “Aggressive Grants by Affirm-Minded Justices.” American Politics Research 30: 219-34. Bergholtz, Gunnar, and Aleksander Peczenik. 1997. “Precedent in Spain.” In Interpreting Precedents: A Comparative Study, edited by D. Neil MacCormick and Robert S. Summers, 293-314. Aldershot, UK: Ashgate Dartmouth. Blackshield, Tony, Michael Coper, and George Williams. 2001. The Oxford Companion to the High Court of Australia. New York: Oxford University Press. Bogart, W.A. 1994. Courts and Country: The Limits of Litigation and the Social and Political Life of Canada. Toronto: Oxford University Press.

flemming2.p65

110

01/04/2004, 2:24 PM

References

Boucher, Robert L., and Jeffrey A. Segal. 1995. “Supreme Court Justices as Strategic Decision Makers: Aggressive Grants and Defensive Denials on the Vinson Court.” Journal of Politics 57: 824-37. Brenner, Saul. 1979. “The New Certiorari Game.” Journal of Politics 41: 649-55. Brenner, Saul, and John F. Krol. 1989. “Strategies in Certiorari Voting on the United States Supreme Court.” Journal of Politics 51: 828-40. Brodie, Ian. 2002. Friends of the Court: The Privileging of Interest Group Litigants in Canada. Albany: State University Press of New York. Bushnell, Ian. 1982. “Leave to Appeal Applications to the Supreme Court of Canada.” Supreme Court Law Review 3: 479-558. –. 1992. The Captive Court: A Study of the Supreme Court of Canada. Montreal and Kingston: McGill-Queen’s University Press. Caldeira, Gregory A., and John R. Wright. 1988. “Organized Interests and Agenda Setting in the US Supreme Court.” American Political Science Review 82: 1109-28. –. 1990a. “Amici Curiae before the Supreme Court: Who Participates, When, and How Much?” Journal of Politics 52: 782-806. –. 1990b. “The Discuss List: Agenda Building in the Supreme Court.” Law and Society Review 24: 807-36. Caldeira, Gregory A., John R. Wright, and Christopher J.W. Zorn. 1999. “Sophisticated Voting and Gatekeeping in the Supreme Court.” Journal of Law, Economics, and Organization 15: 549-72. Carty, Kenneth, William Cross, and Lisa Young. 2000. Rebuilding Canadian Party Politics. Vancouver: University of British Columbia Press. Crane, Brian A., and Henry S. Brown. 1996. Supreme Court of Canada Practice, 1996. Scarborough, ON: Carswell. –. 2002. Supreme Court of Canada Practice, 2002. Scarborough, ON: Carswell. Dotan, Yoav. 1999. “Do the ‘Haves’ Still Come Out Ahead? Resource Inequalities in Ideological Courts: The Case of the Israeli High Court of Justice.” Law and Society Review 33: 1059-80. Ducat, Craig R., and Robert Dudley. 1987. “Dimensions Underlying Economic Policymaking in the Early and Later Burger Courts.” Journal of Politics 49: 521-39. Dudley, Robert, and Craig R. Ducat. 1986. “The Burger Court and Economic Liberalism.” Western Political Quarterly 38: 236-49. Eng, Svein. 1997. “Precedent in Norway.” In Interpreting Precedents: A Comparative Study, edited by D. Neil MacCormick and Robert S. Summers, 189-218. Aldershot, UK: Ashgate Dartmouth. Epp, Charles R. 1998. The Rights Revolution: Lawyers, Activists, and Supreme Courts in Comparative Perspective. Chicago: University of Chicago Press. Epstein, Lee. 1991. “Courts and Interest Groups.” In The American Courts: A Critical Assessment, edited by John B. Gates and Charles A. Johnson, 335-71. Washington, DC: CQ Press. Epstein, Lee, and Jack Knight. 1998. The Choices Justices Make. Washington, DC: CQ Press. Epstein, Lee, and Joseph F. Kobylka. 1992. The Supreme Court and Legal Change: Abortion and the Death Penalty. Chapel Hill: University of North Carolina Press. Epstein, Lee, Jeffrey A. Segal, and Harold J. Spaeth. 2001. “The Norm of Consensus on the US Supreme Court.” American Journal of Political Science 45 (April): 362-77.

flemming2.p65

111

01/04/2004, 2:24 PM

111

112 References

Epstein, Lee, Jeffrey A. Segal, Harold J. Spaeth, and Thomas G. Walker. 2003. The Supreme Court Compendium: Data, Decisions, and Developments. Washington, DC: CQ Press. Flemming, Roy B. 2000. “Processing Appeals for Judicial Review: The Institutions of Agenda Setting in the Supreme Courts of Canada and the United States.” In Political Dispute and Judicial Review: Assessing the Work of the Supreme Court of Canada, edited by Hugh Mellon and Martin Westmacott, 40-62. Scarborough, ON: Nelson Thomson Learning. Flemming, Roy B., Peter F. Nardulli, and James Eisenstein. 1992. The Craft of Justice: Politics and Work in Criminal Court Communities. Philadelphia: University of Pennsylvania Press. Fletcher, Joseph F., and Paul Howe. 2001. “Public Opinion and Canada’s Courts.” In Judicial Power and Canadian Democracy, edited by Paul Howe and Peter H. Russell, 255-96. Montreal and Kingston: McGill-Queen’s University Press. Fouts, Donald E. 1969. “Policy-Making in the Supreme Court of Canada, 19501960.” In Comparative Judicial Behavior, edited by Glendon Schubert and David J. Danelski, 257-92. New York: Oxford University Press. Frankfurter, Felix, and James M. Landis. 1993 [1927]. The Business of the Supreme Court: A Study in the Federal Judicial System. Holmes Beach, FL: Gaunt. Galanter, Marc. 1974. “Why the ‘Haves’ Come Out Ahead: Speculations on the Limits of Legal Change.” Law and Society Review 9: 95. Gertner, Eric. 1988. “The Supreme Court’s Expanding Jurisdiction and Other Procedural Tales.” Supreme Court Law Review 10: 417-28. Greene, Ian, Carl Baar, Peter McCormick, George Szablowski, and Martin Thomas. 1998. Final Appeal: Decision-Making in Canadian Courts of Appeal. Toronto: James Lorimer and Company. Greene, William H. 1997. Econometric Analysis. Saddle River, NJ: Prentice Hall. Gressman, Eugene. 1964. “Much Ado about Certiorari.” Georgetown Law Journal 52: 742-66. Grofman, Bernard, and Timothy J. Brazill. 2002. “Identifying the Median Justice on the Supreme Court through Multidimensional Scaling: Analysis of ‘Natural Courts’ 1953-1991.” Public Choice 112: 55-79. Grossman, Joel B., Herbert M. Kritzer, and Stuart Macauley. 1999. “Do the ‘Haves’ Still Come Out Ahead?” Law and Society Review 33: 803. Hagle, Timothy, and Harold J. Spaeth. 1992. “The Emergence of a New Ideology: The Business Decisions of the Burger Court.” Journal of Politics 54: 120-34. Hartnett, Edward A. 2000. “Questioning Certiorari: Some Reflections Seventy Years After the Judges’ Bill.” Columbia Law Review 100: 1643-1738. Hausegger, Lori, and Stacie Haynie. 2003. “Judicial Decisionmaking and the Use of Panels in the Canadian Supreme Court and the South African Appellate Division.” Law and Society Review 37: 635-57. Haynie, Stacie L. 1994. “Resource Inequalities and Litigation Outcomes in the Philippine Supreme Court.” Journal of Politics 56: 752-72. Haynie, Stacie, C. Neal Tate, Reginald S. Sheehan, and Donald R. Songer. 2001. “Winners and Losers: A Comparative Analysis of Appellate Courts and Litigation Outcomes.” Unpublished paper presented at the Annual Meeting of the American Political Science Association, San Francisco, CA, 30 August-2 September.

flemming2.p65

112

01/04/2004, 2:24 PM

References

Heard, Andrew D. 1991. “The Charter in the Supreme Court of Canada: The Importance of Which Judges Hear an Appeal.” Canadian Journal of Political Science 24: 289-307. Heinz, John P., and Edward O. Laumann. 1982. Chicago Lawyers: The Social Structure of the Bar. New York: Russell Sage Foundation. –. 1997. “The Constituencies of Elite Urban Lawyers.” Law and Society Review 31: 441-72. Heinz, John P., and Peter M. Manikas. 1992. “Networks among Elites in a Local Criminal Justice System.” Law and Society Review 26: 831-61. Heinz, John P., Robert L. Nelson, Edward O. Laumann, and Ethan Michelson. 1998. “The Changing Character of Lawyers’ Work: Chicago in 1975 and 1995.” Law and Society Review 32: 751-75. High Court of Australia. 1998. Annual Report 1997-1998. Canberra: High Court of Australia. Hogg, Peter W. 1996. Constitutional Law in Canada. Scarborough, ON: Carswell. Johnston, Richard, Andre Blais, Elisabeth Gidengil, and Neil Nevitte. 1996. The Challenge of Direct Democracy: The 1992 Canadian Referendum. Montreal and Kingston: McGill-Queen’s University Press. Lavine, Sharon. 1992. “Advocating Values: Public Interest Intervention in Charter Litigation.” National Journal of Constitutional Law 2: 27-62. Lempert, Richard. 1999. “A Classic at 25: Reflections on Galanter’s ‘Haves’ Article and Work It Has Inspired.” Law and Society Review 33: 1099. Maltzman, Forrest, James F. Spriggs, and Paul J. Wahlbeck. 2000. Crafting Law on the Supreme Court: The Collegial Game. New York: Cambridge University Press. McCormick, Peter. 1993. “Party Capability Theory and Appellate Success in the Supreme Court of Canada, 1949-1992.” Canadian Journal of Political Science 26: 523-40. –. 1998. “Birds of a Feather: Alliances and Influences on the Lamer Court 19901997.” Osgoode Hall Law Journal 36: 339-68. –. 2000. Supreme at Last: The Evolution of the Supreme Court of Canada. Toronto: James Lorimer and Company. McGuire, Kevin T. 1993. The Supreme Court Bar: Legal Elites in the Washington Community. Charlottesville: University Press of Virginia. –. 1994. “Amici Curiae and Strategies for Gaining Access to the Supreme Court.” Political Research Quarterly 47: 821-37. –. 1995a. “Capital Investments in the US Supreme Court: Winning with Washington Representation.” In Contemplating Courts, edited by Lee Epstein. Washington, DC: CQ Press. –. 1995b. “Repeat Players in the Supreme Court: The Role of Experienced Lawyers in Litigation Success.” Journal of Politics 57: 187-96. –. 1998. “Explaining Executive Success in the US Supreme Court.” Political Research Quarterly 51: 505-26. McGuire, Kevin T., and Gregory A. Caldeira. 1993. “Lawyers, Organized Interests, and the Law of Obscenity: Agenda Setting in the Supreme Court.” American Political Science Review 87: 717-26. Miguel, Alfonso, and Francisco J. LaPorta. 1997. “Precedent in Spain.” In Interpreting Precedents: A Comparative Study, edited by D. Neil MacCormick and Robert S. Summers, 259-92. Aldershot, UK: Ashgate Dartmouth.

flemming2.p65

113

01/04/2004, 2:24 PM

113

114 References

Monahan, Patrick J. 1991. Meech Lake: The Inside Story. Toronto: University of Toronto Press. Morrow, James D. 1994. Game Theory for Political Scientists. Princeton, NJ: Princeton University Press. Morton, F.L., and Avril Allen. 2001. “Feminists and the Courts: Measuring Success in Interest Group Litigation in Canada.” Canadian Journal of Political Science 34(1): 55-84. Morton, F.L., Peter H. Russell, and Troy Riddell. 1994. “The Canadian Charter of Rights and Freedoms: A Descriptive Analysis of the First Decade, 1982-1992.” National Journal of Constitutional Law 5: 1-60. Murphy, Walter F. 1973. Elements of Judicial Strategy. Chicago: University of Chicago Press. Ordeshook, Peter C. 1992. A Political Theory Primer. New York: Routledge. Ostberg, C.L., Matthew E. Wetstein, and Craig R. Ducat. 2002. “Attitudinal Dimensions of Supreme Court Decision Making in Canada: The Lamer Court, 1991-1995.” Political Research Quarterly 55: 235-56. Peck, Sidney R. 1967. “The Supreme Court of Canada, 1958-1966: A Search for Policy through Scalogram Analysis.” Canadian Bar Review 45: 666-75. –. 1969. “A Scalogram Analysis of the Supreme Court of Canada, 1958-1967.” In Comparative Judicial Behavior, edited by Glendon Schubert and David J. Danelski, 293-334. New York: Oxford University Press. Perry, H.W. Jr. 1991. Deciding to Decide: Agenda Setting in the United States Supreme Court. Cambridge, MA: Harvard University Press. –. 1992. “Certiorari.” In The Oxford University Companion to the Supreme Court of the United States, edited by Kermit L. Hall. New York: Oxford University Press. Poole, Keith T., and Howard Rosenthal. 2000. Congress: A Political-Economic Analysis of Roll Call Voting. New York: Oxford University Press. Provine, Doris Marie. 1980. Case Selection in the United States Supreme Court. Chicago: University of Chicago Press. Russell, Peter H. 1987. The Judiciary in Canada: The Third Branch of Government. Toronto: McGraw-Hill Ryerson. –. 1993. Constitutional Odyssey: Can Canadians Become a Sovereign People? Toronto: University of Toronto Press. Salokar, Rebecca Mae. 1992. The Solicitor General: The Politics of Law. Philadelphia: Temple University Press. Schubert, Glendon A. 1959. Quantitative Analysis of Judicial Behavior. Glencoe, IL: Free Press. –. 1962. “Policy without Law: An Extension of the Certiorari Game.” Stanford Law Review 14: 284-327. Segal, Jeffrey A., and Harold J. Spaeth. 2002. The Supreme Court and the Attitudinal Model Revisited. New York: Cambridge University Press. Sheehan, Reginald S. 1992. “Governmental Litigants, Underdogs, and Civil Liberties: A Reassessment of a Trend in Supreme Court Decision-Making.” Western Political Quarterly 45: 27-39. Sheehan, Reginald S., William Mishler, and Donald R. Songer. 1992. “Ideology, Status, and the Differential Success of Direct Parties Before the Supreme Court.” American Political Science Review 86: 464-71. Simpson, Jeffrey. 1993. Faultlines: Struggling for a Canadian Vision. New York: HarperCollins.

flemming2.p65

114

01/04/2004, 2:24 PM

References

–. 2002. The Friendly Dictatorship. Toronto: McClelland and Stewart. Snell, James G., and Frederick Vaughan. 1985. The Supreme Court of Canada: History of the Institution. Toronto: Osgoode Society. Songer, Donald R. 1979. “Concern for Policy Outputs as a Cue for Supreme Court Decisions on Certiorari.” Journal of Politics 41: 1185-94. Songer, Donald R., Charles M. Cameron, and Jeffrey A. Segal. 1995. “An Empirical Test of the Rational-Actor Theory of Litigation.” Journal of Politics 57: 1119-29. Sopinka, John, and Mark A. Gelowitz. 1993. The Conduct of an Appeal. Toronto: Butterworths. Sossin, Lorne. 1996. “The Sounds of Silence: Law Clerks, Policy-Making, and the Supreme Court of Canada.” University of British Columbia Law Review 30: 279308. –. 1999. Boundaries of Judicial Review: The Law of Justiciability in Canada. Scarborough, ON: Carswell. Stager, David A.A., and Harry W. Arthurs. 1990. Lawyers in Canada. Toronto: University of Toronto Press. Stern, Robert, Eugene Gressman, Stephen M. Shapiro, and Kenneth S. Geller. 1993. Supreme Court Practice, 7th ed. Washington, DC: Bureau of National Affairs. Supreme Court of Canada. 2001. Bulletin of Proceedings: Special Edition, Statistics – 1991-2001. Ottawa: Supreme Court of Canada. Tanenhaus, Joseph, Marvin Schick, Matthew Muraskin, and Daniel Rosen. 1963. “The Supreme Court’s Certiorari Jurisdiction: Cue Theory.” In Judicial DecisionMaking, edited by Glendon Schubert. New York: Free Press. Tate, C. Neal, and Panu Sittiwong. 1989. “Decision Making in the Canadian Supreme Court: Extending the Personal Attributes Model across Nations.” Journal of Politics 51: 900-16. Ulmer, S. Sidney. 1972. “The Decision to Grant Certiorari as an Indicator to the Decision ‘on the Merits.’” Polity 4: 429. –. 1978. “Selecting Cases for Supreme Court Review: An Underdog Model.” American Political Science Review 72: 902-10. –. 1981. “Selecting Cases for Supreme Court Review: Litigant Status in the Warren and Burger Courts.” In Courts, Law, and Judicial Process, edited by S. Sidney Ulmer, 284-97. New York: Free Press. –. 1983. “Conflict with Supreme Court Precedents and the Granting of Plenary Review.” Journal of Politics 45: 474-78. –. 1984. “The Supreme Court’s Certiorari Decisions: Conflict as a Predictive Variable.” American Political Science Review 78: 901-11. –. 1985. “Governmental Litigants, Underdogs, and Civil Liberties in the Supreme Court: 1903-1968 Terms.” Journal of Politics 47: 899-909. Walker, Thomas G., Lee Epstein, and William J. Dixon. 1988. “On the Mysterious Demise of Consensual Norms in the United States Supreme Court.” Journal of Politics 33: 825-41. Welch, Jillian. 1985. “No Room at the Top: Interest Group Interveners and Charter Litigation in the Supreme Court of Canada.” University of Toronto Faculty of Law Review 1985: 204-31. Young, Robert A. 1998. The Secession of Quebec and the Future of Canada. Montreal and Kingston: McGill-Queen’s University Press.

flemming2.p65

115

01/04/2004, 2:24 PM

115

Index

Aboriginal rights, 64 agenda-setting accounts jurisprudential, 4, 61-76, 99-100 litigant-centred, 3, 43-60, 99 strategic choice, 4, 77-97, 100-2 agenda-setting conventions of other high courts, 102-4 “aggressive grant,” as strategic behaviour in Supreme Court of Canada, 84-85, 86, 88 in US Supreme Court, 4, 78, 79-80 Alberta percentage of Canadian lawyers, 26 response to survey of leave to appeal attorneys, 27, 28 amicus curiae briefs in Supreme Court of Canada, 13, 23 in US Supreme Court, 13, 23 See also interest groups (Canada); interest groups (US) “appeals as of right,” in Canadian criminal cases, 7, 8-9, 93 appellate litigation, as legal specialty (Canada), 21, 31, 32, 33, 36-37, 46 attitudinal model, and US Supreme Court voting behaviour, 80-81 attorneys (Canada) as government employees, 25-26 proportion of, by province, 26-27 “Seven Sisters” legal firms, 21, 72, 99 status-seeking model, 14, 33-40 See also Supreme Court attorneys (Canada)

flemming2.p65

116

attorneys (US) percentage of lawyers practising in New York, 28 status-seeking model, and leave to appeal applications, 35 stratification of, 21 See also Supreme Court Bar (US) Australia, High Court, 102-3 “Bay Street lawyers,” 21, 72 Bloc Québécois, 10 “boutique firms.” See attorneys (Canada); attorneys (US); names of specific legal firms Brennan, William (Justice, US Supreme Court), 83 British Columbia proportion of Canadian lawyers, 26 response to survey of leave to appeal attorneys, 27, 28 Bulletin of Proceedings (of Supreme Court of Canada) published votes of leave panels, 15 as source of data for study of agenda setting, 17-18 Burger, Warren (Chief Justice, US Supreme Court), 19, 44, 83 Canada, Supreme Court. See Supreme Court (Canada) Canadian Abortion Rights Action League (CARAL), 21 Canadian Alliance, 1, 106

01/04/2004, 2:24 PM

Index

Canadian Charter of Rights and Freedoms and agenda-setting guidelines of Supreme Court of Canada, 66, 69 as attorney specialization, and participation in leave process, 40 interpretation by Supreme Court of Canada, 2 and pay equity ruling by Supreme Court of Canada, 1 and religious freedom cases, 70-71 voting patterns among justices of the Supreme Court of Canada, 82 See also Constitution (Canada) Canadian Civil Liberties Union, 21 Canadian Law List, xii, 26, 46 CARAL. See Canadian Abortion Rights Action League (CARAL) “cert.” See certiorari process (US) certiorari process (US) “cert pool,” writing of short assessments of applications, 15 and “cue theory,” 62, 65 decisions by US Supreme Court, 4 definition, 3 favouring of US government in granting certiorari, 44 and Judiciary Act, 1925, 98 leave to appeal grant rate, 104 See also leave to appeal process (Canada) Charlottetown Accord, 11 Chief Justice (Supreme Court of Canada) assignment of leave panels, 93-94 assignment of writing opinions, 87-88 See also justices, of Supreme Court of Canada Chrétien, Jean, 10 civil liberties cases and agenda of US Supreme Court, 65 non-unanimous votes in Supreme Court of Canada, 81 wins against governments, in US, 44 Clerk of the Court (US), 14-15 commercial law cases, superceded by Charter cases, 12, 13

flemming2.p65

117

consensus. See voting behaviour (Supreme Court of Canada) Constitution (Canada) appeals, as percentage of docket of Supreme Court of Canada, 12-13 challenges, and agenda setting of Supreme Court of Canada, 64 crises, focused on Quebec’s demand for sovereignty, 11 and reference questions put to Supreme Court, 7, 9-10 See also Canadian Charter of Rights and Freedoms; Charlottetown Accord; Meech Lake Accord corams (Supreme Court of Canada) assignment of justices to, 92-93 composition of, 83 and consensus, 94, 97 definition, 82 merits decisions by, 95-96 size variance, 85, 101 voting behaviour, 93-96 See also leave panels (Supreme Court of Canada) Corbett, David, 22 Cory, Peter (Justice, Supreme Court of Canada), 74, 82, 92, 95 criminal cases (Canada) and agenda setting of Supreme Court of Canada, 64 “appeals as of right,” 7, 8-9, 93 percentage of docket of Supreme Court of Canada, 12-13 “cue theory,” and US certiorari process, 62, 65 “defensive denial,” as strategic behaviour in Supreme Court of Canada, 84, 86, 88 in US Supreme Court, 4, 78-79, 80 Department of Justice (Canada), 39 dissent as bargaining over agenda in US Supreme Court, 83 in Supreme Court of Canada decisions (see voting behaviour [Supreme Court of Canada])

01/04/2004, 2:24 PM

117

118 Index

Douglas, William O. (Justice, US Supreme Court), 83 Eberts, Mary, 21 Eberts Symes Street & Corbett (Canadian legal firm), 21, 46 economic regulation, and nonunanimous votes in Supreme Court of Canada, 81 Egan v. Canada (1995), 1, 2, 105 English Court of Appeal, 45 Federal Court of Canada, 72 Finland, Supreme Court, 103 Frankfurter, Felix (Justice, US Supreme Court), 62-63, 98 “friend of the court.” See amicus curiae briefs Galanter, Marc, 44-45 See also “party capability theory” game theory agenda game in US Supreme Court, 77-80 agenda setting by Supreme Court of Canada, 84-97 “sincere” behaviour, 77, 78 “sophisticated” behaviour, 77 See also “aggressive grant,” as strategic behaviour; “defensive denial,” as strategic behaviour; voting behaviour (Supreme Court of Canada) Germany, Constitutional Court, 103 Goldberg, Arthur J. (Justice, US Supreme Court), 83 Goldstein and Howe (US legal firm), 19 Gonthier, Charles (Justice, Supreme Court of Canada), 74, 82, 94, 95 Goodmans (Canadian legal firm), 21 governments legal counsel, 25-26 as litigant, 55-57, 72 Gowling Lafleur Henderson (Canadian legal firm), 22 Gressman, Eugene, 105 Griswold, Erwin, 44

flemming2.p65

118

Harris, Mike, 21 Iacobucci, Frank (Justice, Supreme Court of Canada), 73, 82, 94, 95 interest groups (Canada) as absent in Supreme Court of Canada agenda setting, 13, 99, 101-2 criteria for intervener status, 102 financial support from federal government, 3 as interveners only after leave to appeal granted, 3, 13, 23, 99 less active than in US legal system, 13, 23-24 interest groups (US) and agenda setting of US Supreme Court, 3, 13, 99 amicus curiae briefs, before US Supreme Court, 13, 23 Israel, High Court, 45 “Judges’ Bill” (US, 1925). See Judiciary Act (US, 1925) judgments, drafting (Supreme Court of Canada), 87-88 judicial activism, by Supreme Court of Canada, 1-2 Judiciary Act (US, 1925), 5, 62, 98, 104, 106 jurisprudential account, of agenda setting, 61-76 ambiguity of rules, and freedom to set agenda by justices, 61-62 “cue theory,” and US certiorari process, 62 as decentralized, in Supreme Court of Canada, 99-100 and granting judicial review in US Supreme Court, 4 guidelines to appeal case selection by Supreme Court of Canada, 64-68 review of lower court decisions, 68-71 and Rule 10, US Supreme Court, 6 and silence of justices regarding their decisions, 63

01/04/2004, 2:24 PM

Index

variables, definitions and characteristics, 68-73 See also leave panels (Supreme Court of Canada); public importance justices, of Supreme Court. See names of individual justices; voting behaviour (Supreme Court of Canada) Kennedy, Anthony M. (Justice, US Supreme Court), 79 LaForest, Gérard (Justice, Supreme Court of Canada), 74, 82, 92, 95 Lamer, Antonio (Chief Justice, Supreme Court of Canada) on appeals in commercial law cases, 12 arrangement of leave panels, 73, 93 coram partners, 94, 95 and granting of judicial review, 12-13 on justices’ silence regarding their decisions, 63 and “natural courts,” 10 political context of court, 10-11 voting pattern in Charter cases, 82 Lang Michener (Canadian legal firm), 22 Law Branch, Supreme Court of Canada, 15 lawyers. See attorneys (Canada); attorneys (US); Supreme Court attorneys (Canada); Supreme Court Bar (US) LEAF. See Women’s Legal Education and Action Fund (LEAF) leave panels (Supreme Court of Canada) as agenda setters, 67 compared with agenda-setting process in US Supreme Court, 15, 59, 76 as decentralized process decision making, 76 decisions model, by panel, 75 as disadvantage for repeat players, 59

flemming2.p65

119

establishment by 1956 amendment of Supreme Court Act, 101 and jurisprudential account of agenda setting, 99-100 personnel changes, 73-74 as reduction of work load of court, 101 size, composition, and voting patterns, 92-96 and strategic account, of agenda setting, 81, 83-97, 100-2 voting by, 15-16, 78, 83-97 See also corams (Supreme Court of Canada); leave to appeal process (Canada) leave to appeal process (Canada) “aggressive grant,” 84-85, 86, 88 application volume, 7-8, 12 applications granted, 12, 30 “B” list, 16, 90, 91 “C” list, 15-16 consensus voting, and accommodation of dissenting opinions, 90-92, 97 and coram composition, 85, 92 “D” list, 16, 90, 91, 92 decision tree, 84 “defensive denial,” 84, 86, 88 game theory applied to process, 84-90 process steps, 14-16 staff attorneys’ recommendations, 89, 91 “tournament” metaphor for process, 16-17 See also “certiorari” process (US); leave panels (Supreme Court of Canada) Lee, Rex, 44 L’Heureux-Dubé, Claire (Justice, Supreme Court of Canada), 74, 82, 95 Liberal Party of Canada, 10, 11 litigant-centred account, of agenda setting, 43-60 of agenda setting by US Supreme Court, 3, 4, 43-45, 62 limited in analysis of Supreme Court of Canada decisions, 99

01/04/2004, 2:24 PM

119

120 Index

See also Litigants (Canada); Litigants (US); Supreme Court attorneys (Canada); Supreme Court Bar (US) litigants (Canada) applicants with resource advantages, success rate, 53-57 corporations, compared with individuals in leave cases, 57 governments, success rate in leave process, 55-57 leave to appeal success rate, by party advantage, 48-49 party status classification, 46-47, 72 status, and appeal grants, 69, 71-72 litigants (US) and granting of judicial review by US Supreme Court, 4 party capability theory (Galanter), 44, 45 status, and access to US Supreme Court appeal process, 3 Litigation Committee, Department of Justice (Canada), 35-36 lower court decisions (Canada) affirmation percentage of lower court decisions, by Supreme Court, 53, 55 conflicts, and review by Supreme Court of Canada, 64, 68-70 correction of misinterpretations by Supreme Court, 65 “egregiousness” of, 66 made binding on entire country, by decision of Supreme Court, 64-65 Ontario appellate judges, held in high esteem, 72 lower court decisions (US) bias towards error correction of, by Supreme Court, 78 conflicts, and agenda-setting guidelines of Supreme Court, 66 percentage affirmed, by Supreme Court, 53 referral of appeal cases to Supreme Court, 5-6 Major, John C. (Justice, Supreme Court of Canada), 73, 82, 92, 95

flemming2.p65

120

Marshall, Thurgood (Justice, US Supreme Court), 79, 83 Mauro, Tony, 21 Mayer, Brown, Rowe and Maw (US legal firm), 21 McCarthy Tetrault (Canadian legal firm), 21, 31, 50 McGuire, Kevin emphasis on attorney status in outcome of cases, 45, 46 study of US “Supreme Court Bar,” 21-22, 25 survey questionnaire, adoption of, The Supreme Court Bar, xi McLachlin, Beverley (Justice, Supreme Court of Canada), 74, 82, 95 Meech Lake Accord, 11 merits stage, of appeal (Supreme Court of Canada) decisions made by corams, 96 interest groups, as interveners, 3, 13, 99 as more accessible for attorneys than in US, 30 Miller Thomson (Canadian legal firm), 21 Montreal lawyers participating in leave to appeal process, 28, 29 percentage of nation’s lawyers, 28 response to survey of leave to appeal attorneys, 27 Mulroney, Brian, 10, 11 Nagelkerke pseudo R2, definition, 75 “natural courts,” in Supreme Court of Canada, 10 New Brunswick Court of Appeal, 9 New Democratic Party, 10 Newfoundland and Labrador Court of Appeals, 1, 2 Norway Supreme Court, 103 obscenity cases (US), 44 O’Connor, Sandra Day (Justice, US Supreme Court), 79 Office of the Solicitor General (US), 26, 35, 39, 44

01/04/2004, 2:24 PM

Index

Old Age Security Act, and same-sex rights, 1, 105 Ontario proportion of Canadian lawyers, 26 response to survey of leave to appeal attorneys, 27 Ottawa lawyers participating in leave to appeal process, 28, 29 legal agents, as liaison with Supreme Court of Canada, 22 as location for law practice and Supreme Court, 22-23 response to survey of leave to appeal attorneys, 27 panels, of Supreme Court of Canada. See corams (Supreme Court of Canada); leave panels (Supreme Court of Canada) Philippines, Supreme Court, 45 political context, of Supreme Court of Canada and agenda setting, 106 emergence of regionalized, “fourth” party system, 11 during Lamer court, 10-11 See also public importance, of issues (Supreme Court of Canada) Process Clerk, Registrar’s Office, Supreme Court of Canada, 14-15 Progressive Conservative Party, 10, 11 public importance, of issues (Supreme Court of Canada) compared with US Supreme Court agenda setting, 106 declared by Supreme Court Act, 1975, 2, 5 interpretation by justices, 100-1 as jurisprudential variable, in logit model of agenda setting, 69, 70 Lamer on silence of Court regarding, 63 Sopinka on Court’s interpretation of, 63, 64 Quebec percentage of Canadian lawyers, 26

flemming2.p65

121

Quebec applications for leave to appeal given to panel of Quebec justices, 73-74 Quebec civil code cases, handled by Quebec justices, 15, 93, 94, 101 Quebec Secession Reference of 1998, 9-10, 11 referendum, 1995, 11 response to survey of leave to appeal attorneys, 27 sovereignty demand, and Canadian constitutional crises, 11 Quebec Secession Reference (Supreme Court of Canada, 1998), 9-10, 11 Queen’s Counsel attorneys (Canada) advantage when representing applicant before Supreme Court, 50, 53, 99 appointment as reflection of political ties, 33 and merits experience before Supreme Court, 33 and professional status model of leave activity, 38 as repeat players before Supreme Court, 46 reference questions, to Supreme Court of Canada and advisory role of Court, 7, 9-10 as part of executive function, 9 as percentage of reviews, 9-10 See also Canadian Charter of Rights and Freedoms Reform Party of Canada, 1, 10 religious freedom cases, 70-71 repeat players. See Supreme Court attorneys (Canada); Supreme Court Bar (US) review panels. See leave panels (Supreme Court of Canada) same-sex rights, and Old Age Security Act, 1, 105 Scalia, Antonin (Justice, US Supreme Court), 29, 79 “Seven Sisters” law firms, 21, 72, 99 Sidley Austin Brown & Wood (US legal firm), 19-20, 35

01/04/2004, 2:24 PM

121

122 Index

Sopinka, John (Justice, Supreme Court of Canada), 95 agenda-setting guidelines of Supreme Court of Canada, 66 perspective on how Court conducts its business, 63-64 speculation on Supreme Court’s interpretation of “public importance” criterion, 63 on strategic voting, 88-89, 91 voting pattern in Charter cases, 82 Spain, Supreme Court, 103 status of attorney (see Queen’s Counsel attorneys [Canada]; Supreme Court attorneys [Canada]) of litigant (see litigants [Canada]) Stevenson, William (Justice, Supreme Court of Canada), 73 strategic account, of agenda setting “aggressive grants,” 4, 78, 79-80, 84-85, 86, 88 “defensive denials,” 4, 78-79, 80, 84, 86, 88 and leave panels of Supreme Court of Canada, 81, 83-97, 100-2 See also leave panels (Supreme Court of Canada); voting behaviour (Supreme Court of Canada) Supreme Court Act (Canada) 1956 amendment, 101 1975 amendment, 5, 9, 76, 104-5 Supreme Court attorneys (Canada) as advisors to other attorneys, 33-34 appellate litigation, as legal specialty, 21, 31, 32, 33, 36-37, 46 applicant lawyers’ experience advantage and leave outcomes, 49-50 as counsel to interveners in Supreme Court cases, 40 firm size, unrelated to appeal outcomes, 50-51 inequalities between, and Supreme Court outcomes, 47-57 lack of formal Supreme Court Bar, 25 larger proportion with merits experience than in US, 41

flemming2.p65

122

leave to appeal, by province and city, 27-29 litigation experience before Supreme Court, 29-33, 43, 49-50, 70, 71 location, not significant in shaping leave activity, 40, 41 logit model of leave to appeal outcomes, 51-57 Ottawa agents, for Supreme Court liaison, 22 professional status model of leave activity, 38-40 questionnaire, as source of data for study of Supreme Court of Canada agenda setting, 18 repeat players, before Supreme Court, 17, 30-31, 57 resources, and outcome of Supreme Court cases, 36, 46-47 respondent lawyers’ court experience and leave outcome, 52-53 stratification, based on Supreme Court experience, 30-33 See also Queen’s Counsel attorneys (Canada) Supreme Court Bar (US) agenda-setting role for US Supreme Court, 41, 43 chosen for experience in Supreme Court by litigants, 43 concentration in Washington, DC, 22, 37 as elite professional community, 20, 40-41 emergence as result of volume of Supreme Court applications, 105 membership criteria, 24-25 repeat player advantage, 29, 58-59 specialists in US Supreme Court hearings, 19-21 study of, by McGuire, 20 Supreme Court (Canada), 69 agenda, pathways to, 6-10, 12-13, 14-16 agenda-setting guidelines, 5, 63-65 application volume, 7-8, 12, 104-5 applications granted, 12, 30 and Canadian Charter of Rights and

01/04/2004, 2:24 PM

Index

Freedoms as indicator of “public importance,” 66, 69 and cases of “public importance,” 2, 5, 17, 63-65, 69, 70, 100-1, 106 clerking, and professional status model of attorney leave activity, 38-39 compared with US Supreme Court, 3-5, 12, 14-16, 30, 59, 94, 106 criminal cases, “appeals as of right,” 7, 8-9, 93 hearings of oral arguments, 13-14 judgments, drafting, 87-88 judicial activism, 1-2 jurisprudential account, of agenda setting, 4, 61-76, 99-100 litigant-centred account, of agenda setting, 3, 4, 43-60, 62, 99 reference questions, 7, 9-10 rulings, and public policy, 2 signed opinions, volume of, 30 strategic account, of agenda setting, 4, 78-80, 84-86, 88 and Supreme Court Act, 5, 9, 76, 104-5 See also Chief Justice (Supreme Court of Canada); interest groups (Canada); jurisprudential account, of agenda setting; leave panels (Supreme Court of Canada); leave to appeal process (Canada); litigantcentred account, of agenda setting; litigants (Canada); lower court decisions (Canada); strategic account, of agenda setting; Supreme Court attorneys (Canada); voting behaviour (Supreme Court of Canada) Supreme Court (US) agenda-setting guidelines, 3-4, 66 application volume, 30, 104, 105 applications granted, compared with Supreme Court of Canada, 12, 53 compared with Supreme Court of Canada, 3-5, 12, 14-16, 30, 59, 94 en banc sitting tradition, 15, 100 and “Judges’ Bill” (Judiciary Act, 1925), 5, 62, 98, 104, 106

flemming2.p65

123

and Rule 10, governing writ of certiorari decisions, 5-6 signed opinions, volume of, 30 See also certiorari process (US); interest groups (US); jurisprudential account, of agenda setting; litigant-centred account, of agenda setting; lower court decisions (US); strategic account, of agenda setting; Supreme Court Bar (US) Supreme Court Law Review (Canada) leave to appeal activity report, xi survey of applications for leave to appeal process, 3, 98 Supreme Court of Canada Manual: Practice and Advocacy, 22 Supreme Court of Canada Practice, xi, 22, 25, 63-64 Supreme Court Practice (US), 21, 105 Supreme Court Reports (Canada) as source of data for study of agenda setting, 18 as source of data for repeat player attorneys, 46 Sweden, Supreme Court, 104 Symes, Beth, 21 Toronto lawyers participating in leave to appeal process, 28-29 percentage of nation’s lawyers, 28 response to survey of leave to appeal attorneys, 27 Torys (Canadian legal firm), 21 Tribe, Larry, 44 Trudeau, Pierre Elliott, 10 Tupper, Sir Charles, 33 Turner, John, 21 unanimity, in voting (Supreme Court of Canada). See voting behaviour (Supreme Court of Canada) US Supreme Court. See Supreme Court (US) Vancouver lawyers participating in leave to appeal process, 28, 29

01/04/2004, 2:24 PM

123

124 Index

response to survey of leave to appeal attorneys, 27 Vinson, Frederick M. (Justice, US Supreme Court), 83 voting behaviour (Supreme Court of Canada) at conference, 86-87 dissent votes, 83-84, 85, 100 individual justices, in Charter cases, 82 non-unanimous votes, analysis of, 81 review panels, vote by memorandum, 15, 90 strategic voting, 84-88 unanimous voting, 81, 83, 96 See also corams (Supreme Court of Canada); leave panels (Supreme Court of Canada)

flemming2.p65

124

voting behaviour (US Supreme Court) attitudinal model, as explanation of voting behaviour, 79 conference votes, 15, 16 dissent voting patterns of justices, 79, 82-83 strategic voting, 4, 78, 79-80, 85 Warren, Earl (Chief Justice, US Supreme Court), 44, 83 White, Byron R. (Justice, US Supreme Court), 83 Wilson, Bertha (Justice, Supreme Court of Canada), 74 women, and pay equity, 1 Women’s Legal Education and Action Fund (LEAF), 21 merits success rate, 58 shaping of public policy as intervener in Supreme Court cases, 58

01/04/2004, 2:24 PM

Printed and bound in Canada by Friesens Set in Stone by Artegraphica Design Co. Ltd. Copy editor: Frank Chow Proofreader: Diana Breti Indexer: Annette Lorek

flemming2.p65

125

01/04/2004, 2:24 PM