PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5’08 KzlProlecc8-PresICIRCIDaIeDue.indd

 

WHY DO HIGH COURT JUDGES JOIN?
JOINING BEHAVIOR AND AUSTRALIA’S SERIATIM TRADITION

By

Rebecca Danielle Wood

A DISSERTATION

Submitted to
Michigan State University
In partial fiJlfillment of the requirements
For the degree of

DOCTOR OF PHILOSOPHY
Political Science

2008

ABSTRACT

WHY DO HIGH COURT JUDGES JOIN?
JOINING BEHAVIOR AND AUSTRALIA’S SERIATIM TRADITION

By
Rebecca Danielle Wood

The American majority opinion tradition is the fulcrum on which theories of
joining behavior on the US. Supreme Court are balanced (e. g., Hammond et a1. 2005;
Maltzman et a1. 2000; Segal and Spaeth 2002). Because majority opinions alone carry
the weight of precedent in the American system, they provide a strong motivation for the
Justices to assent to a common set of reasons. The High Court of Australia, however,
operates under a seriatim opinion tradition, wherein each Justice writes a separate set of
reasons for his or her decision. Even though the Australian system does no explicitly
encourage joint opinions, the High Court produces a substantial—and increasing—
number of j oint opinions each year. The goal of this dissertation is to explain why
Justices on the High Court engage in what appears to be unnecessary joining behavior.
Why do they accept the reasons of their colleagues when there is no institutional reward
for doing so?

This dissertation builds upon the existing literature by developing and testing a
model of j oining behavior that is applicable in courts using a seriatim opinion tradition.
Using data from the High Courts Judicial Database (Haynie et al. 2007), this dissertation
uses a time series fractional cointegration analysis to provide support for the assumption
that the seriatim tradition is functioning properly in the High Court of Australia. In other
words, the analysis finds no evidence of an underlying consensual norm that discourages

individual Justices from writing separate opinions. A series of logistic regression

analyses reveal patterns in joining behavior on the High Court. Justices on the High
Court engage in joining behavior because 1) the Chief Justice successfully facilitates such
behavior, 2) the Justices are able to find coauthors with similar policy views to their own,
3) the case at hand is simple and/or not of great importance, and 4) because the Justices

are accustomed to and comfortable with the process of collegial decision making.

Copyright by Rebecca Danielle Wood 2008

To Marvin and Mary Wood

ACKNOWLEDGEMENTS

Like most dissertations, this one would not have been possible without the
support of faculty, friends and family. My advisor Reginald S. Sheehan maintained a
consistent level of confidence in me that I did not always believe I deserved. He gave me
the freedom to pursue the questions I found most interesting in the manner I found most
compelling. He provided me with the opportunity to work on the High Courts Judicial
Database (Haynie et al. 2007). Not only did this give me the experience of working on a
large collaborative data project, but it also inspired the research in this dissertation.
Reggie also made it possible for me to travel to Australia as part of Michigan State
University’s study abroad program. He has been my sounding board and my advocate.

Had it not been for C. Neal Tate, I would not have pursued my Ph.D. at all. His
Proseminar in Public Law at the University of North Texas introduced me to the
scientific study of judges and courts, and his encouragement convinced me that I might
be able to contribute to this field. He taught me to think about judicial behavior in a
scientific way, and his enthusiasm for the enterprise continues to be an inspiration.

In the relatively short time that I have worked with Thomas H. Hammond, he has
had an enormous impact on this project. His comments on the later versions of this work
challenged me to be ever more precise, and this dissertation is much better for his efforts.
I also greatly appreciate the contributions—and the patience—of Kenneth C. Williams
and Christopher E. Smith.

Some of the research for this dissertation took place during my time as a Visiting

Scholar at the Australian National University Faculty of Law. My thanks to the ANU

vi

 

Faculty of Law for their support, and especially to Fiona Wheeler and Michael Coper
who helped me to secure the position. I also want to thank the Australian scholars and
practitioners who took the time to introduce me to the subtleties of judi cial behavior on
the High Court. Among them are Chief Justice Sir Anthony Mason, Justice Michael
Kirby, Andrew Lynch, John McMillan, Waanda McCarthy, and George Williams.

Harold J. Spaeth spent a great deal of time coaching me during the early phases of
this project. His comments on my early research gave me the confidence to submit and
present my first solo conference paper, and I am very grateful for that. Patrick McLeod
provided me with much-needed tools to make sense of my mountains of data, and Harold
Clarke taught me how best to use those tools. Frank Feigert taught me what I deeded to
know in order to be a positive force in the lives of my students, as did Donald Songer (if
from a distance). Susan Johnson and RoHan Williams saw me through many of the
rough spots, be they data-related or otherwise. Kris Dunn provided invaluable support,
especially in the last-minute formatting debacle. I owe a debt to my fellow graduate
students, particularly those involved with the more recent incarnations of the department
softball team.

My family provided unwavering support and understanding during this process.
Amanda Vannieuwenhoven provided much late-night assistance sorting through files of
coding sheets. Aaron Wood provided the logistical support for my world travels. My
mother Mary Wood read countless drafts at a time when volunteers for the task were hard
to come by. She, along with my father Marvin Wood, dropped everything to help me on
innumerable occasions. They have been the enthusiastic team in my corner since the

beginning, and I cannot thank them enough for it.

vii

TABLE OF CONTENTS

List of Tables ..................................................................................................................... xi
List of Figures ................................................................................................................... xii
List of Figures ................................................................................................................... xii
Introduction: Joining Behavior and the Australian High Court .......................................... 1
Chapter One: Practice and Procedure in the High Court .................................................... 2
The Institutional Context .............................................................................................. 2
Staffing the High Court ................................................................................................ 5

The Legal Profession ............................................................................................. 6

The Selection Process ............................................................................................ 9

The Resulting Judges ........................................................................................... 10

High Court Jurisdiction ............................................................................................... 14
Original Jurisdiction ............................................................................................ 14
Appellate Jurisdiction .......................................................................................... 18

High Court Jurisdiction and the Privy Council ........................................................... 19
Daily Operating Procedures ........................................................................................ 24
Setting the Agenda .............................................................................................. 24

The Panel System ................................................................................................ 27

Hearing Cases on the Merits ................................................................................ 30

The Decision and Reasons ................................................................................... 36

Practice and Procedure in Comparative Perspective .................................................. 42
Chapter Two: A Brief History of the High Court of Australia ......................................... 44
The Early History of the High Court .......................................................................... 44
Establishing an Identity (1903-1919) .................................................................. 45
Growing Pains (1919-1950) ................................................................................ 51

Time to Regroup (1950-1964) ............................................................................. 58

The Modern High Court ............................................................................................. 60
The Barwick Court (1964-1981) ......................................................................... 60

The Gibbs Court (1981-1987) ............................................................................. 67

The Mason Court (1987-1995) ............................................................................ 71

The Brennan Court (1995—1998) ......................................................................... 77

The Gleeson Court (1998-present) ...................................................................... 83

The High Court Centenary and Beyond ..................................................................... 85
Chapter Three: Existing Theories of Joining Behavior .................................................... 89
The American Literature ..................................................................................... 89

viii

Judging Across the World ................................................................................... 95

Australian Jurimetrics and Beyond ..................................................................... 98
Conceptualizing Joining Behavior ............................................................................ 106
Decision Patterns ............................................................................................... 107
Joining Behavior in the US. Supreme Court .................................................... 108
Joining Behavior in the Context of the High Court ........................................... 120
Chapter Five: Consensual Norms on the High Court ..................................................... 122
Defining Types of Opinion-Writing Behavior ......................................................... 123
Terminology ...................................................................................................... 123

A lexicon for describing High Court opinion-writing ....................................... 125
Consensual Norms in the Literature ......................................................................... 127
Consensual Norms in the Australian Context ........................................................... 131
The Norm of Concurring Opinions ................................................................... 132
Dissent and the Consensual Norm ..................................................................... 138

An Underlying Norm of Consensus in Australia? ............................................. 144
Chapter Six: The Production of Joint Opinions ............................................................... 157
Joining Behavior and Case-Level Characteristics .................................................... 157
Characteristics of the Decision Making Environment ....................................... 158
Case Characteristics ........................................................................................... 168

The Disposition of the Case .............................................................................. 179
Joining Behavior and Judge-Level Characteristics ................................................... 182
Approach to the Working Environment ............................................................ 182
Social Background Characteristics .................................................................... 185
Chapter Seven: Data and Measures ................................................................................ 189
Joining Behavior on the High Court ......................................................................... 189
Operational Definitions of Joining Behavior .................................................... 193
Predicting the Single Opinion for the Court ...................................................... 195
Predicting Other Joint Opinions ........................................................................ 197
Operational Hypotheses ............................................................................................ 198
Characteristics of the Decision Making Environment ....................................... 198
Case Characteristics ........................................................................................... 205
Outcome of the Case ......................................................................................... 211
Judge Characteristics ......................................................................................... 213
Approach to the Working Environment ............................................................ 213
Social Background Characteristics .................................................................... 215
Controlling for Panel and Coalition Size ........................................................... 216
Chapter Eight: Analyzing Joining Behavior on the High Court ..................................... 222
Methodology ............................................................................................................. 222
Models of Single-Opinion Dispositions ............................................................ 225
Case-Level Models of Joining Behavior ........................................................... 227
Judge-Level Models of Joining Behavior .......................................................... 234
Joining Behavior on the Gleeson Court ............................................................. 241

ix

Chapter Nine: Conclusion and Future Research ............................................................. 256

Evaluating the Findings ............................................................................................ 257
Appendix A: Judges of the High Court of Australia by Seat .......................................... 261
Appendix B: Coding Classifications ............................................................................... 262
Bibliography .................................................................................................................... 266

LIST OF TABLES

Table 1 - Justice Attributes on the Barwick Court .......................................................... 64
Table 2 - Justice Attributes on the Gibbs Court .............................................................. 69
Table 3 - Justice Attributes on the Mason Court ............................................................ 74
Table 4 - Justice Attributes on the Brennan Court .......................................................... 81
Table 5 - Justice Attributes on the Gleeson Court .......................................................... 84

Table 6 - Unit Root Tests on Log-Transformed Separate Opinions per

OWO ................................................................................................................. 150
Table 7 - Unit Root Tests of Log-Transformed Separate Opinions per 100

Opinions ............................................................................................................ 151
Table 8 -— Unit Root Tests and Robinson's Gaussian Semiparametric

Estimates of d (OWO Data) .............................................................................. 154
Table 9 - Cointegrating Regression of Dissents on Concurrences ............................... 155
Table 10 — Descriptive Statistics for Model VariablesT ................................................ 201
Table 11 - Correlation Coefficients between Control and Independent

Variables ........................................................................................................... 218
Table 12 - Collinearity Diagnosticsl ............................................................................. 220
Table 13 - Case-Level Logistic Regression Predicting Single-Opinion

Dispositions ....................................................................................................... 224
Table 14 — Case-Level Logistic Regression Predicting Joining Behavior ................... 228
Table 15 — Judge-Level Logistic Regression Predicting Joining Behavior .................. 233
Table 16 — Judge-Level Logistic Regression Predicting Joining Behavior

on the Gleeson Court ........................................................................................ 242
Table 17 - Individual Models of Joining Behavior on the Gleeson Court .................... 245

Table 18 - Reduced Individual Justice Models of Joining Behavior on the
Gleeson Court ................................................................................................... 246

Table 19 - Performance of Hypotheses in Seven Models of Joining
Behavior ............................................................................................................ 252

xi

LIST OF FIGURES

Figure 1 - Original Jurisdiction Cases as a Percent of Reported Cases:

1965-2002 ........................................................................................................... 17
Figure 2 - Number of Privy Council Cases in Commonwealth Law

Reports by Year .................................................................................................. 23
Figure 3 - Average Panel Size by Year ........................................................................... 31
Figure 4 - Total Number of Cases by Year ..................................................................... 32
Figure 5 - Percentage of Cases with at least One Joint Opinion ..................................... 40
Figure 6 - Concurring and Joint Opinions in Majority per 100 Reported,

1965-2002 ......................................................................................................... 135
Figure 7 - Percentage of Cases with Concurrences and Joint Opinions in

Majority, 1965-2002 ......................................................................................... 136
Figure 8 - Concurring and Dissenting Opinions per Opinion-Writing

Opportunity ....................................................................................................... 149
Figure 9 - Opinion to Vote Ratio on the High Court over Time ................................... 190
Figure 10 - Percentage of Unanimous Cases with Single-Opinion

Dispositions ....................................................................................................... 1 97
Figure 11- Percentage of Cases with Any Joint Opinion and Smaller

Coalitions Only ................................................................................................. 198
Figure 12 - Liberalism Scores by Party of Appointing Prime Minister ........................ 205
Figure 13 - Predicted Probabilities of a Single-Opinion Disposition by

Panel Size .......................................................................................................... 226
Figure 14 - Predicted Probabilities of Joint Opinions by Panel Size ............................ 229
Figure 15 - Predicted Probabilities of a Joint Opinion by Panel Liberalism ................ 230
Figure 16 - Predicted Probabilities of a Joint Opinion by Panel Diversity ................... 231
Figure 17 - Predicted Probabilities of a Joint Opinion in Criminal Cases by

Panel Liberalism ............................................................................................... 232
Figure 18 - Predicted Probabilities of a Decision to Join by Panel Size ....................... 235

xii

Figure 19 - Predicted Probabilities of a Decision to Join by Ideological
Distance ...................... ' ....................................................................................... 237

Figure 20 - Predicted Probabilities of Joining Behavior by Nonparty

Participation Rate .............................................................................................. 238
Figure 21 - Predicted Probabilities of Decision to Join across Approach to

Working Environment ...................................................................................... 239
Figure 22 - Predicted Probabilities of Decision to Join on Gleeson Court

by Coalition Size ............................................................................................... 243
Figure 23 - Scree Plot for Gleeson Court Conditional Distance Matrix ....................... 248

Figure 24 - Derived Configuration of Joining Behavior Conditional on
Votes ................................................................................................................. 249

Figure 25 - Liberalism Scores of Gleeson Court Justices by Party of
Appointing Prime Minister ............................................................................... 251

Figure 26 - Average Number of Decisions to Join over Previous Five
Cases, by Justice ............................................................................................... 254

xiii

 

INTRODUCTION: JOINING BEHAVIOR AND THE AUSTRALIAN HIGH COURT

It is because of this basic need for mutually confirmatory studies that quantitative
analysis must finally operate at an international level.

— Anthony Blackshield (1972, 64)

The research presented in this dissertation adapts existing judicial behavior theories
to courts with varied institutional arrangements. Specifically, theories of joining
behavior that have been developed almost exclusively to explain majority opinion
coalitions on the US. Supreme Court will be generalized to explain joining behavior on
the High Court of Australia—a court without a majority opinion tradition.

The High Court is producing an increasing number of joint opinions each year.
The goal of this research is to explain why Justices on the High Court engage in what
seems to be unnecessary joining behavior. Why do they accept the reasons of their
colleagues when there is no institutional reward for doing so? To address this question,
this dissertation is organized into the following chapters. Chapter One provides an
introduction to practice and procedure on the High Court. Chapter Two presents a brief
history of the High Court, with a special focus on the issues and personalities that have
shaped the modern Court. Chapter Three focuses on the foundations for a theory of
joining behavior. Chapter Four presents an econometric analysis of consensual norms
on the High Court. Chapter Five presents eleven conceptual hypotheses derived from the
existing theories, but adapted for the unique characteristics of the High Court. Chapter
Six provides operational hypotheses and describes the data and measures used to test
them. Chapter Seven conducts a series of tests of these operational hypotheses and
presents the results. Finally, Chapter Eight offers some concluding remarks about joining

behavior on the High Court.

CHAPTER ONE: PRACTICE AND PROCEDURE IN THE HIGH COURT

As in the case of the United States, we believe that the decisions of this court will
breathe a living spirit into the dry bones of a parchment constitution, and that your
names will live in history with those of the illustrious exponents of American
constitutional law.

— Govemor-General Tennyson to the Court in 1903 (Bennett 1980, 23)

The Govemor-General’s remarks at the commissioning of the first three Justices of
the High Court illustrate the overt comparisons already being made between that
institution and the one on which it was modeled: the US. Supreme Court. While the
High Court of Australia was fashioned largely in the image of the US. Supreme Court,
there are many differences in practice and procedure between the two courts (Hunt 1930).
Because most theories of judicial decision-making behavior have been designed to study
Justices on the US. Supreme Court (Atkins 1991), institutional differences make it
inappropriate to transplant these theories wholesale to judges in other courts (Tate 1983).
After a brief review of the structure of the Australian government, some important
institutional characteristics of the High Court are outlined. These include the broad
jurisdiction of the High Court, the way Justices are selected, the structure of the legal
profession from which they arise, and the daily operating procedures that govern the
handling of the Court’s business.

The Institutional Context

The structure of the Australian government has been described as a “Washminster”
system—part Westminster parliamentary government and part Washington-inspired

separation of powers (Thompson 1980). This arrangement is detailed in the

Commonwealth Constitution, which was enacted by an act of the British Parliament1 and

 

1
Commonwealth of Australia Constitution Act 1900 (Imp).

took effect on January 1, 1901. Australian independence came as a peaceful agreement
between Britain and the Colonies, and the new government maintained its ties to the
British crown. As such, Australia is a constitutional monarchy with the Queen of
England as its nominal sovereign. So friendly was this relationship that the new
Australian government invited King George V to open its inaugural parliamentary
session on May 9, 1901 (Hirst 1998).

Australia adopted a federal structure similar to the American government. The
national legislature is a bicameral parliament made up of a lower house (the House of
Representatives), an upper house (the Senate), and the Queen, who is typically
represented by a Governor-General chosen on the advice of the Australian Prime
Minister. Members of the House of Representatives are elected in single-member
districts based on population and using preferential voting. In keeping with the
Westminster tradition, the leader of the party or coalition with a majority in the House
becomes the Prime Minister. The Prime Minister forms his government by assembling a
Cabinet and other ministers from among the members of his own party in Parliament.
This combination of parliamentary government with a maj oritarian electoral design
encourages a strong two-party system, where party discipline is maintained through the
unitary nature ofthe legislature and the executive (Galligan 1987). Because the House is
the origin of all supply (or appropriation) bills, the party in control of the House
essentially controls the government.

The Senate was modeled not on the British House of Lords but on the US. Senate

(Meek 1999). Each of the six states is allotted 12 senators, and the mainland territories

 

each have two.2 Senators serve six-year terms. The intent was for the Senate to represent
the interests of the states (Hirst 1998). Unlike the US. Senate, Australian senators are
chosen through a proportional voting system using statewide multi-member districts.
This arrangement promotes the proliferation of political parties, such that the party
composition of the Senate is much more diverse than the largely two-party House
(Lijphart 1999). This diversity contributes to the Senate’s role as a house of review, as it
often results in divided government—something otherwise rare in parliamentary systems
(Sawer 1999). The government cannot pass legislation—including supply bills—without
the assent of the Senate. When the Senate refuses to pass supply bills, the government
may be forced to dissolve Parliament in order to keep from a government shutdown. In
this way, the structure of the Senate adds a separation of government power that is not
found in the British Westminster system.

Some argue that the most important feature of the Australian government is its
federal arrangement (Galligan 1995). Each of the six Australian states has its own
parliamentary government.3 The head of government in each state is the Premier. The
state Premier is the leader of the party or coalition with majority control in the lower
house of that state’s parliament. Each state also has a Governor who is appointed by the
Queen on the advice of the state’s Premier. The distribution of power between the

national and state governments has been one of the major sources of political

 

2 The six states are New South Wales, Queensland, South Australia, Tasmania, Victoria and Western
Australia. The mainland territories are the Australian Capital Territory and the Northern Territory. The
Constitution does not guarantee representation for these territories, so their representation in the
Commonwealth Parliament is by statutory arrangement.

3

Three of the territories are also self-governing: the Northern Territory, the Australian Capital Territory,
and Norfolk Island. The unicameral legislatures in the territories are called “legislative assemblies,” and
are parliamentary in design. The remaining territories are governed by the national government.

disagreement since Australia’s federation (Blackshield 1972). The authors of the
Australian Constitution rejected the more centralized wording found in the Canadian
Constitution4 in favor of an American-style arrangement, where the states retain all
powers not delegated to the national government (Hodgins et al. 1978). Just as in the
United States, though, the boundaries of national power are incompletely delineated in
the Constitution.

The High Court has played an important role in refereeing the distribution of power

between the two levels of government (Selway and Williams 2005).5 It has been a
distinctly political branch of government from the start, even if the Court itself has been
slow to acknowledge this fact (Galligan 1987). While Westminster-style governments
are characterized by parliamentary sovereignty, Australia has been characterized by
constitutional sovereignty since its founding document took effect (Gleeson 2003b). The
architects of the Australian Constitution recognized this, and some acknowledged the
crucial role the High Court would play in maintaining constitutional sovereignty (Mason
2003a)

Staffing the High Court

In countries with a civil law tradition, judges and lawyers begin their careers on
separate paths. Would-be judges and lawyers attend separate training institutions, and

there is little lateral movement between the two professions (Guamieri 2004). In

 

The relevant provrsron rs Section 91 of the Canadian Constitution, which specrfies that the federal
government retains reserved powers. Section 52 of the Australian Constitution enumerates the powers of
the Commonwealth Government, leaving residual powers to the states.

5 Most decisions of the Australian judiciary were originally appealable to the Privy Council in Britain. The
Constitution has always stipulated that the High Court is the final arbiter of inter se questions—~questions
dealing with the distribution of power between the federal and state governments. This arrangement is
addressed in detail later in this chapter.

 

common law countries like Australia and the United States, judges are often chosen from
the ranks of the legal profession. As such, the process is far less self-selecting than in
civil law countries. As a preliminary step to understanding how judges function on the
bench, it is necessary first to examine the way in which people become judges. In
Australia, as in the American federal judiciary, judges are given their jobs by politicians.
This is the main opportunity for the government of the day to exert dejure influence over
the judiciary. Understanding how and why judges are selected for the job, then, is a very
important step for understanding what they do once they arrive (Ostberg and Wetstein
2004)
The Legal Profession

The legal profession in Australia dates back to well before Australian Federation.6
As in colonial America, the British system was imported as the basis for the legal
profession in Australia. Each colony developed its own Bar admission and regulation
structures, and these structures are still largely in place today. This is a similar system to
the United States, where the Bar is organized on a state level. The current state systems
are all derived from a shared legal tradition, but the state legal societies do differ to a
degree. There is no nationalized legal profession currently, although the Law Council of
Australia (LCA) and others have been working to unite the various state—based systems
(LCA 2003).

The legal profession in Australia is a multi-billion dollar industry, generating more

than AUD$1O billion in 2002 (LCA 2003). In the states of Victoria, Queensland, and

 

6 The term “Australian Federation” refers to the culmination of the process by which the six self-governing
British colonies in Australia united in a federation. This is akin to the use of the term “American
Founding.”

New South Wales, unlike the unified American systems, the legal profession is divided.
As in the British tradition, these three states divide practitioners into two types of
lawyers: solicitors and banisters (Meek 1999). A solicitor handles all non-trial matters.
This includes the drawing up of wills and real estate transactions. Additionally, the
solicitor conducts the initial legal research in preparation for a case, and then turns the
matter over to a barrister. The solicitor has no direct parallel in America, although the
role performed by a solicitor combines certain aspects the duties of a paralegal and a
research attorney. The solicitor is the main point of contact between a litigant and his or
her legal representative, and may join law firms and distribute work among colleagues
(LCA 2001). In most cases, a client must hire a solicitor, who will then provide initial
legal advice and procure a barrister on the client’s behalf.

Barristers are expert advocates, and plead cases at both the trial and appellate
levels. They are solo practitioners, and are not permitted to join forces to form law firms
(LCA 2001). Barristers appointed to argue cases in front of the High Court receive the
title of Queen’s Council (QC). This is also commonly called “taking silk.” It is a
professional ranking, as is becoming a member of the US. Supreme Court Bar, but is
bestowed on barristers by the Crown.7 This honor is generally attained by only the most
exceptional barristers after ten to fifteen years as a practicing junior counsel (Meek
1999). Like members of the US. Supreme Court Bar (McGuire 1994), QCs also tend to
be elite members of both their profession and their society.

In Victoria, Queensland, and New South Wales, barristers and solicitors are

regulated under different bodies. In these states, the Bar is the professional organization

 

7
In the Australian context, “the Crown” refers to the head of the British monarchy, who is also the
Australian head of state.

for barristers. In other states, both types of practitioners are regulated by the state Bar. It
is from the ranks of the Bar that most judges are selected (Gleeson 2003a). Some of the
distinctions between the barrister and the solicitor, however, are common among all of
the states and territories. In 1993, the state of New South Wales passed legislation
allowing practitioners to be listed as both barristers and solicitors. Prior to this, however,
the banister was not allowed to perform the duties of the solicitor, and vice versa. In the
states with an undivided profession, lawyers can be admitted as practitioners, enabling
them to perform the duties of a solicitor, a barrister, or both.

With the introduction of a mutual recognition scheme in 1992, it has become
relatively simple for a lawyer in one state to gain permission to practice law in another
state. In addition, standards for admission to the profession are similar across states. An
applicant must have a degree in law. This is roughly an equivalent to an American
bachelor’s degree. The Australian law degree typically takes five years to complete, and
law students also earn a degree in an additional subject. Admission to the profession also
requires practical training, which involves serving an apprenticeship or articles of
clerkship.

The membership of the profession, however, remains relatively homogeneous. As
of 1999, there were more than 30,000 members of the Australian legal profession (LCA
2001). The majority of these worked in private solicitor practices. Despite the fact that
there are now nearly equal numbers of male and female law graduates, the profession
remains dominated by men. Nearly 90% of barristers, for example, are men. Legal aid
offices and community legal centers are staffed largely by women, while men tend to

dominate the other, more lucrative sectors of the profession (LCA 2001).

 

The Selection Process

Unlike the American Constitution, the Australian Constitution provides explicit
instructions for the selection of all Commonwealth. judges. The process is governed by
section 72, which provides:

The Justices of the High Court and of the other courts created by the Parliament —

(i) Shall be appointed by the Governor-General in Council; (ii) Shall not be

removed except by the Governor-General in Council, on an address from both

Houses of the Parliament in the same session, praying for such removal on the

ground of proved misbehaviour or incapacity; (iii) shall receive such remuneration

as the Parliament may fix; but the remuneration shall not be diminished during their
continuance in office. (Commonwealth Constitution, § 72)

In 1977, this section was amended following a national referendum to prescribe a
mandatory retirement age of 70 years for all judges.

Appointments to the High Court are formally made by the Governor-General, who
is the British Crown’s representative in Australia. The Governor-General’s choice is
almost always based on the recommendation of the Prime Minister’s Cabinet through the
Attorney-General (Evans 2002). The High Court of Australia Act 1979 (Cth)8 added the
additional requirement that the Attorney-General consult the states before the
recommendation is made. The role of the states in the process, however, is minimal at
best (Moens 1997). There is no formal judicial confirmation process. Procedures for the
selection of judges are based on tradition as opposed to hard and fast legal requirements.
The Commonwealth Attomey-General typically advises the state Attorneys-General of

vacancies and solicits nominations.9 Nominees for the High Court are selected by the

 

8 The abbreviation in parentheses represents the jurisdiction in which the act was passed. Following the
Australian Guide to Legal Citation (Kirwin and Masters 2002), “Cth” is the abbreviation for
“Commonwealth,” in this case the Commonwealth Parliament.

9
The Attorney-General in a parliamentary system is a Member of Parliament by definition. This is in
contrast to the American Attorney General, who is appointed by the President from outside Congress. Any

government from the legal profession. They usually come from lower levels of the
judiciary or the Sydney or Melbourne Bar. The Commonwealth Attomey-General
presents this list, along with his or her additions and recommendations, to the Cabinet.
The Cabinet then makes a decision, and the Governor-General carries out the formal
appointment.

The Resulting Judges

In the current system, the list of names presented to the Cabinet is not on the public
record, and this fact concerns many observers. There have been a number of calls for
reform of the current selection system, most calling for a more formalized merit selection
process. Some politicians have made clear their desire to shape the Bench to reflect their
own political leanings. In one example, former Deputy Prime Minister Tim Fischer made
it known that the Howard government intended to appoint only “capital ‘c’
conservatives” to the Bench (Jackson 2003, 3).

The Australian Democrats, a minor party, proposed a merit-based selection plan
intended to minimize the impact of political consideration on the resulting appointments.
This plan was dismissed as unnecessary by the Law Council of Australia and the Liberal
Coalition government, among others (LCA 1997). Many important people, including
Justice Michael Kirby, have called for greater transparency in the process, with an eye to
keeping political forces out of the process (Childs 2001).

Justice Mary Gaudron has called attention to the potential danger of invoking the
“merit” justification for appointments made in the current selection system (Gaudron

2002). Because the pool of potential nominees is small and lacking in diversity, the

 

nominee from the ranks of Congress would have to resign his position, as per Article 1, section 6 of the
US. Constitution.

10

 

Australian judiciary tends to be a highly homogenous group (Gleeson 2003a). This is
especially true in the High Court. Justice Gaudron, appointed in 1987, was the first
female to sit on the High Court bench. Justice Gaudron had high hopes that her 2003
retirement would lead to the appointment of a female replacement, but, in her words, “we
muffed it” (Mazzocchi 2003, 1). The return to an all-male Court sparked renewed debate
about the appropriateness of affirmative action-type policies for High Court nominees.
With Gaudron’s retirement and subsequent replacement by Justice Dyson Heydon, there
were no female Justices on the Bench until the appointments of Susan Crennan in 2005
and Susan Kiefel in 2007.

Former Chief Justice Sir Harry Gibbs has expressed concern that, at least in
Queensland, merit has been surpassed by gender as the deciding factor in nominations for
judicial appointment Richards 2000). National Bar Association President Tom Glynn
points to the importance of strictly merit-based selection procedures, ruling out a “female
seat” on the bench (Mazzocchi 2003). The Attorney-General has said that John
Howard’s Liberal Coalition government used exactly such a merit-based program for
selecting new judges. In response to questions about Justice Gaudron’s male
replacement, he responded as follows:

Gender was not a basis for identifying whether people were suitable for the

appointment or not. What we seek to do is appoint the best person for the job.
(Attomey-General Daryl Williams, as quoted in Rubenstein 2002, 3)

It is clear, however, that there is a lack of consensus on this issue. Many observers
are highly critical of the notion that there is a single “best person” for a position on the
nation’s highest court (Rubenstein 2002). Alexandra Richards, former president of

Australian Women Lawyers, explains this position:

11

There are, at any given time, a number of lawyers who have the requisite ability,
character, temperament and experience to be suitable for a particular judicial
appointment. There is no serious argument that appointments should be made
otherwise fi'om those so qualified as are willing to accept an appointment.
However, to suggest that the “best” appointment from within that group can be
identified on the basis 0 “merit” is to give a false objectivity to something which
can only be a matter of opinion and judgment. (Richards 2000, 1)

This is what Justice Gaudron has called “the merit fiction,” and she cites it as one
reason why women are still highly underrepresented at the Bar and in Australia’s senior
judiciary (Gaudron 2002). She admits that this problem “wouldn’t matter a fig,” except
that having women in the judiciary is crucial to administration of justice (Gaudron 2002,
5). This sentiment is echoed by other women in the legal profession, who often note that
a diverse bench is one that can better administer justice in Australia’s diverse society
(Rubenstein 2002).

Similar arguments are advanced on the need for more diversity in other areas. One
complaint is based on the lack of geographic representation on the High Court. Of the
first 44 appointees to the High Court, more than half have come from the state of New
South Wales (Dominello and Neumann 2002). Over 80% have come fi'om either New
South Wales or Victoria. Another complaint is the lack of racial diversity on the Bench.
All of the Justices—current and former—are white.

Some judges are cautiously optimistic. Chief Justice Murray Gleeson notes that the
professional base from which judges are selected has been expanded, so that now “there
is a wider pool of legal talent than used to be drawn upon” (Gleeson 2001a, 3). Gleeson
argues that the judiciary will inevitably reflect the population of the legal profession—
admittedly with some delay (Gleeson 2003 a). Because of the increased diversity in the
younger generation of lawyers, the ranks of the judiciary are bound to follow suit. Justice

Ian Callinan has also noted the increased number of female members of the highest ranks

12

of Australia’s legal profession. Addressing the newest Queen’s Counsel appointees in
2000, he made mention of the number of females in the group:
There are among you only five women. Tonight is not the occasion to debate
whether that number should be regarded as a reasonably sufficient number in all of
the circumstances. But on any view, that you are at last here and that your numbers

are increasing show that we have moved a distance from some of the
discriminatory practices of the past. (Callinan 2000, 3)

In general, though, many of the Court’s onlookers do not share in this confidence.
Justice Gaudron spoke at the inaugural meeting of the Women Lawyers’ Association of
New South Wales, noting that little progress has been made since the first female
Australian lawyers emerged—nearly a century ago (Branson 1997). Federal Court Judge
Catherine Branson spoke to the group a few weeks later. She argued that the obstacles
facing women in the profession start not with the judicial appointment process, but much
earlier in the legal career (Branson 1997). The subtle bias experienced by female
practitioners from law school to judicial vetting often hinders them in their quest to
achieve the top positions in the profession (Branson 1995).

Kim Rubenstein argues that the problem is not just with the homogenous pool of
potential judges, but also with the group of people doing the deciding:

At the moment it is a predominantly male conservative cabinet deciding who the

"best" person is for the job. Indeed, our century-old experience of judicial selection

has shown that when male politicians gaze at the available gene pool of potential

High Court appointees, they only see reflections of themselves, and what they
understand as depictions of merit. (Rubenstein 2002, 1)

The replacement of Justice Mary Gaudron with Dyson Heydon increased the prevalence
of this kind of criticism. Many female practitioners saw this as an affront to the cause of
promoting diversity, noting that there were plenty of well-qualified females from which

to choose (McDonald 2002).

13

High Court Jurisdiction

Jurisdiction is the power and authority of a court to hear and rule on controversies,
and forms the crux of the institution’s role and function. The American Constitution was
undoubtedly the model for a great deal of the Australian Constitution, especially where
the judicial power is concerned (Cowen and Zines 2002). The Australian Constitution,
however, was drafted more than a century after its American counterpart. The lessons of
the American experience informed the Australian framers. The differences between the
two documents reflect those lessons, as well as the fundamental differences between
presidential and parliamentary governmental structures.

Original Jurisdiction

Like the American Constitution, the Australian Constitution lays out particular
situations in which the highest national court hears cases at first instance. The original
jurisdiction of the High Court is laid out in section 75, which reads:

In all matters (i) arising under any treaty; (ii) affecting counsuls or other

representatives of other countries; (iii) in which the Commonwealth, or a person

suing or being sued on behalf of the Commonwealth, is a party; (iv) between

States, or between residents of different States, or between a State and a resident of

another State; (v) in which a writ of Mandamus or prohibition or an injunction is

sought against an officer of the Commonwealth; the High Court shall have original
jurisdiction. (Commonwealth Constitution, § 75)

This section confers an original jurisdiction that is similar—but not identical—to
the original jurisdiction that Article 111 Section 2 of the United States Constitution confers
on its Supreme Court. Many of the specific subject matters conferred by the Australian
Constitution were probably adopted without much thought as to their suitability for the
Australian situation (Cowen and Zines 2002). One similarity is the use of the term
‘matters’ in describing the Court’s jurisdiction. This has been interpreted to mean that

the High Court, like the American Supreme Court, lacks jurisdiction to provide advisory

14

opinions (Cowen and Zines 2002). Additional restrictions were implemented in the
Boilermakers’ Case,10 when the Court held that non-judicial functions could not be
delegated to judicial bodies.

The Australian Constitution contains a specific provision allowing the
Commonwealth Parliament to confer additional matters to the original jurisdiction of the
High Court. This is a significant departure from the American system. The provision in

section 75(v) of the Australian Constitution is similar to the provision in the American

Judiciary Act held invalid in Marbury v. Madison. 11 Under the Judiciary Act 1903 (Cth),
the High Court’s original jurisdiction was exclusive in matters involving suits between
states, between the Commonwealth and a state, or matters in which a writ is sought
against an officer of the Commonwealth. Original jurisdiction in matters of
constitutional interpretation, admiralty, and federal law are not constitutionally
entrenched, but are instead left for Parliament to confer. The Judiciary Act 1903 (Cth)
provided this jurisdiction. Additionally, Parliament has conferred original jurisdiction in
a number of other matters through such legislation as the Commonwealth Electoral Act
1918 (Cth), the Income Tax Assessment Act 1936 (Cth), the Patents Act 1952 (Cth) and
the Admiralty Act 1988 (Cth).

As a result of Parliarnent’s tendency to add to—but never subtract from—the
Court’s original jurisdiction, it grew to proportions much larger than that of the US.
Supreme Court (Cowen and Zines 2002). By the 19605 this extensive nondiscretionary

original jurisdiction, coupled with the burdensome nondiscretionary appellate jurisdiction

 

l
0 R v Kirby; Ex Parte Boilermakers ' Society of Australia (1956) 94 CLR 254

H (1803) 1 Cranch 137

15

discussed in the next section, resulted in nearly insurmountable workload pressures
(Bennett 1980). As the scope of Australian statutory law increased, the High Court was
unable to continue its role as federal trial court while still maintaining its more important
role as interpreter of the constitution (Beaumont 2002).

Section 77(iii) of the Australian Constitution provides Parliament the option to
parse the federal jurisdiction among the various state courts. In the 19705, the High
Court increasingly remitted cases to appropriate state supreme courts for determination
(Cowen and Zines 2002). Matters arising under laws made by the Commonwealth
Parliament were routinely handled by state courts under this provision, but this still left a
great deal of work to be handled by the High Court alone (Australian Law Reform
Commission 2000). The High Court was still the preferred venue for adjudicating on
first instance in many matters because it provided a level of uniformity and consistency
that the numerous state courts could not match (Mason 2000). This left the High Court
with an incredibly high caseload.

As Attorney-General, Garfield Barwick was instrumental in presenting a solution to
this problem. He first advocated the creation of an intermediate federal court in the early
1960s, but the Federal Court of Australia did not begin to exercise jurisdiction until
February of 197 7 (Beaumont 2002). This new statutory court was created through
amendments to the Judiciary Act 1903 (Cth). It was to be a small court, with jurisdiction
conferred in special areas only (Australian Law Reform Commission 2000). The Federal
Court began by handling mostly bankruptcy, administrative law, industrial law, and trade

practices law.

16

Figure l - Original Jurisdiction Cases as a Percent of Reported Cases: 1965-2002

 

50

40*

30‘

20~

10‘

 

 

Percentage of Cases

 

*7 i w a f fi‘v f v

1%4 '85 '88 '70 '72 '74 '76 '78 '80 '82 '84 '% '88 '3] '92 '94 '% '98 '0] 2CD?

Soon, the Federal Court began to assume jurisdiction concurrent with much of the
High Court’s original and appellate jurisdictions. Some of this transfer arose from the
increasing tendency of Parliament to confer original jurisdiction through legislation
(Australian Law Reform Commission 2000). In Phillip Morris,12 the High Court also
expanded the Federal Court’s original jurisdiction by refining its doctrine of accrued
jurisdiction to include the Federal Court (Australian Law Reform Commission 2000).

Some scholars have speculated that the establishment of the Federal Court
alleviated much of the workload stress due to original jurisdiction cases (Gageler 2002a).
Figure 1 shows the original jurisdiction cases as a percentage of all reported cases.
Among the reported cases from 1970 through 2002 there seems to be little lasting shift in
the overall percentage of original jurisdiction cases. At the most, original jurisdiction

cases have constituted a maximum of 13 cases of the Court’s business in a year.

 

12
Philip Morris v. Commissioner of Business Franchises (Vic) (1989) 167 CLR 399

17

Appellate Jurisdiction

The majority of the High Court’s caseload comes from its appellate jurisdiction.
One of the main differences between the Australian High Court and the American
Supreme Court is the breadth of appellate jurisdiction conferred by the respective
constitutions. While the American Supreme Court is limited to adjudicating federal
issues, the High Court may hear appeals from state courts even in the absence of a federal
question. The American founders were deeply suspicious of the new centralized
government, and were loathe to allow it interfere in the affairs of the states. By contrast,
the Australian experience provided no impetus for such a complete separation (Quick and
Garran 1901). Section 73 of the Australian Constitution therefore confers a general
appellate jurisdiction to the High Court, subject to such restrictions as Parliament cares to
legislate. This section also prevents Parliament from restricting the jurisdiction to hear
appeals from the state supreme courts.

The availability of appeals from judgment is not derived from the common law
tradition, but instead is largely a product of statute. Because of this, the nature and extent
of the availability of appeal varies depending on the statute conferring the right
(Australian Law Reform Commission 2000). The appeals process serves to correct trial
court errors, and also to provide consistency and clarity in the law and its proper
application. The role of the High Court in this process is largely the latter, especially in
recent times. Before 1984, many of the appeals to the High Court came as of right. This
right to appeal was conferred by the various statutes under which the litigation was
commenced, and was usually related to the amount of money at issue in civil litigation.
In addition, interlocutory judgments could be appealed by leave to appeal, granted by

either the state supreme courts or the High Court. Such leave would be granted if the

18

judgment in the lower court was arguably incorrect (Australian Law Reform Commission
2000). The Court had a largely nondiscretionary appellate jurisdiction for that reason,
and many of the cases before it tended to be similar and straightforward (Gleeson 2000b).
Amendments to the Judiciary Act 1903 (Cth) in 1976 altered this situation slightly.
The new default mechanism for taking an appeal to the High Court was to obtain special
leave to appeal. Appeals as of right remained, however, in constitutional issues or issues
involving property in excess of AUD$20,000. This nondiscretionary appellate
jurisdiction was removed in 1984 through further amendments to the Judiciary Act.
These amendments provide that appeals will not be heard by the High Court unless the
High Court itself grants special leave to have an appeal heard on the merits. Criminal
appeals have always required special leave, but most civil litigation enjoyed a right to
appeal until the 1984 amendments. The result of these amendments was to abolish the

Court’s nondiscretionary appellate jurisdiction, in service of the Court’s role as

interpreter of laws. ’3

High Court Jurisdiction and the Privy Council

The Judicial Council of the Privy Council in England sat above the High Court at
the pinnacle of the Australian legal system for most of Australia’s history. The Privy
Council is a division of the British House of Lords, and is also called “Her Majesty in
Counsel.” The Privy Council is partially staffed by English Law Lords. There is no limit

to the number of Privy Counsellors, and there are currently around 550 members. It is

 

1 3 Additionally, the Court has changed its interpretation of what constitutes an appeal. Section 73 of the
Constitution provides High Court jurisdiction “to hear and determine appeals from all judgments, decrees,
orders and sentences” of a number of different types of courts. Recently, the Court has held that
determinations can be made on questions posed to the court during the course of trial proceedings without
the trial court judge’s assent (O’Toole v. Charles David (1991) 171 CLR 232).

19

not an appellate court for England proper, but instead is a body of judges charged with
maintaining a degree of unity in the English common law around the world (Matson
1993). In addition to the British jurists, judges from many of the Commonwealth
countries are regularly appointed as Privy Counsellors (V oigt et al. 2006), including a
number of former High Court Justices.l4 Traditionally, the Privy Council was made up
of special advisors, and their decisions were phrased as advice to the British monarch.
For this reason, there is a long tradition of unanimous judgments on the Privy Council—a
tradition that Barwick helped to eliminate during his service (Blackshield et al. 2002).

The constitutional provision allowing appeals to the Privy Council was a major
point of contention at federation. The Privy Council was one mechanism through which
the precedents of the British judiciary were “continually injected into the caselaw of the
Australian states” (Castles 1971, 139). As early as 1906, the Privy Council experienced a
hostile Australian reaction to its intervention in Australian litigation (Blackshield et al.
2002). The relations between these institutions were unpredictable, due in part to the
ever-changing membership of the Privy Council. At times, the Privy Council might
promise not to interfere in certain situations, only to have a differently constituted Privy
Council forget or ignore such arrangements (Blackshield et al. 2002).

The Constitution remains silent on appeals from the state supreme courts to the
Privy Council, tacitly leaving no limitations on such appeals. There are, however, two
constitutional limitations on Privy Council appeals from the High Court. First, the High

Court has always retained the right to determine all questions regarding the distribution

 

1 This lrst mcludes Barwrck, Gavan Duffy, Gibbs, Griffith, Krtto, Knox, Latham, McTrernan, Menzres,
Owen, Rich, Stephen, Taylor, Walsh, and Windeyer (Blackshield et al. 2002).

20

of federal and/or state powers, often called inter se questions. The relevant portion of the
Constitution reads:
No appeal shall be permitted to the Queen in Council from a decision of the High
Court upon any question, howsoever arising, as to the limits inter se of the
Constitutional powers of the Commonwealth and those of any State or States, or as
to the limits inter se of the Constitutional powers of any two or more States, unless

the High Court shall certify that the question is one which ought to be determined
by Her Majesty in Council. (Commonwealth Constitution, §74)

Over the years, the exact definition of an inter se question has been the subject of
much debate and litigation. In the end, potentially important federal issues were
excluded from this category, including interstate trade, separation of powers between
branches of the Commonwealth government, and many questions of inconsistency

between state and Commonwealth legislation (Blackshield et al. 2002). While the High
Court only issued a single prerogative writ to appeal under section 74,15 the Privy
Council was still able to hand down numerous important constitutional decisions. This is
especially true in the interpretation of section 92, which provides for free trade among the
states (Blackshield et al. 2002).

It is through the second limitation on Privy Council appeals that the High Court
eventually obtained final jurisdiction over all matters. This second limitation is found in
the last provision in section 74, which reads:

Except as provided in this section, this constitution shall not impair any right which

the Queen may be pleased to exercise by virtue of Her Royal prerogative to grant

special leave of appeal from the High Court to Her Majesty in Council. The

Parliament may make laws limiting the matters in which such leave may be asked,

but proposed laws containing any such limitation shall be reserved by the
Governor-General for Her Majesty’s pleasure. (Commonwealth Constitution, §74)

 

15 . . . . .

The High Court issued a section 74 certificate in Colonial Sugar Refining Co v. Attorney General (C th)
(1912) 15 CLR 182. Justice O’Connor’s absence resulted in an evenly split vote among the remaining four
Justices.

21

The first exercise of this provision was undertaken by the Commonwealth
Government when it passed the Print Council (Limitation of Appeals) Act 1968 (Cth).
This legislation limited potential Privy Council appeals from the High Court to those that
did not potentially deal with an issue of federal law. In Kitano v. Commonwealth,16 the
Privy Council agreed that this was a valid exercise of parliamentary power under section
74. Several years later, the Commonwealth Parliament passed the Privy Council
(Appeals From the High Court) Act 1975 (Cth). This legislation abolished all remaining
appeals from the High Court to the Privy Council without prior permission of the High

Court. This provision was challenged in the High Court in Attomey-General (Cth) v.

T &G Mutual Life. ’7 The High Court rejected the challenge and declared that it would no
longer certify cases for appeal to the Privy Council. As per the 1968 legislation, no
appeal to the Privy Council remained (Blackshield et al. 2002).

Although appeals from the High Court and other federal courts were abolished,
cases could still be appealed to the Privy Council from the decisions of the state courts.
This resulted in a unique situation where litigants who were unsuccessful at the state
court level could effectively choose their appellate venue (Blackshield et al. 2002). This
provided wealthy appellants the ability to “shop around” for their preferred remedy,
giving them a distinct advantage over respondents (Gleeson 2002a). Litigants continued
to appeal to the Privy Council from the state courts long after the High Court refiised to

certify firrther appeals. Although the total number of Australian cases in front of the

 

6
1 (1976) AC 99,135 CLR 1.

17 (1978) 144 CLR161

22

Privy Council subsequently declined, the Commonwealth Law Reports continued to

report cases appealed to the Privy Council into the 19805 (see Figure 2).

Figure 2 - Number of Privy Council Cases in Commonwealth Law Reports by Year

 

 

 

'0
9.
8
7
6i
5

U, i

8

to 4i

O

‘5 3i

8

or 2«

‘3?

e 1-

a)

0. 0 T fl f j fl i fi fi f

 

1%4 '88 '88 '70 '72 '74 '78 '78 '80 '82 '84 '88 '88 '90 '92 '94 '98 '98 '00 2002

This also resulted in a good deal of confusion at the lower court levels. Various
High Court Justices advised lower court judges to abide by High Court precedent, but
these judges effectively had two different — at times conflicting —- masters. This dilemma
was finally solved in 1986 with the passage of The Australia Act 1986 (Cth). This was
one of a pair of statutes, the other of which was passed by the United Kingdom
Parliament. The effect of these statutes was to give the Australian states independence
from Britain. They served to end all remaining appeals from the Australian judiciary to
the Privy Council, among other things (Blackshield 2002b). This placed the High Court
at the apex of the Australian legal system and made its judgments the highest source of
precedent (Twomey 2002).

These formal developments, according to some scholars, also helped cultivate a

new mindset amongthe High Court Justices (Blackshield et al. 2002). Before the

23

abolition of these appeals, the Privy Council had a direct and tangible effect on the
decisions of the High Court, as well as on the development of law in Australia (Gleeson
2000b). As the formal structures of this influence were removed, the Justices were no
longer bound by English authority. The Justices used this opportunity to exert control
over issues previously controlled by English precedent, including the rapid development
of a new approach to section 92 (Gleeson 2002a).

Daily Operating Procedures

In many significant ways, the daily operation of the High Court of Australia is
similar to that of the US. Supreme Court. Given the shared common law tradition, many
of the same concepts are applicable to both institutions. There are, however, a number of
procedural differences. While seemingly negligible, they can produce many significant
differences in terms of agenda, decision making, and the reporting of decisions. The
theories of judicial decision making that are most well known in American circles assume
a particular set of operating procedures. Even procedural rules that seem very similar can
produce very different effects (Glendon et al. 1999). Only by examining the rules and
traditions that shape the daily operation of the High Court can the appropriate
adjustments be made to our existing theories.

Setting the Agenda

The current agenda-setting process in the High Court is roughly comparable to the

American Supreme Court’s certiorari process, but with several important differences. In

the American Supreme Court, a writ of certiorari18 is granted when at least four of the

 

18 To complicate the situation further, the term writ of certiorari means something substantively different
in the Australian legal context. A writ of certiorari in Australia serves to quash a lower court decision. It
is granted as a remedy on the grounds that 1) there was error on the face of the lower court record, 2) the
lower court acted outside its jurisdiction, or 3) the applicant was denied procedural fairness at the lower

24

Justices agree that an appeal should be heard on the merits. The decision is made in a
conference that all Justices attend, and is made based on examination of the written
submissions alone. The Supreme Court clerks have a large role in this procedure, as they
are responsible for drawing up memoranda for each of the applications (Perry 1994).

In the High Court, the agenda is set through the special leave process. When there
is no leave as of right, all parties wishing to appeal must apply for special leave to appeal.
These special leave petitions are heard by a subset of the Bench. The panel is chosen by
the Chief Justice, and is now typically made up of two to three Justices (Jackson 2002b).
The panel size is a matter of practice, and is not dictated by law (Australian Law Reform
Commission 2000). The practice in the past was to have a three judge panel in civil
cases, five judges in criminal cases, and seven Justices in cases deemed important by the
Chief Justice (Gibbs 1986). This has been abandoned in favor of the smaller panel sizes,
largely due to workload concerns (Australian Law Reform Commission 2000).

The applicant traditionally has a right to appear in person for an oral argument on
the application, which is limited to twenty minutes of argument and five minutes of reply
for the applicant and the respondent (Clark 2003 ). These oral arguments are increasingly
being heard via video link between Canberra and the other capitol cities. While oral
arguments take up valuable time, the Court has been reluctant to eliminate this tradition
altogether (Clark 2003). These oral arguments serve to supplement the written briefs
submitted by counsel, which have been required for special leave applications since 1994
(Jackson 1996). Recently, however, Chief Justice Gleeson has advocated eliminating the

right to appear in person as a way to save the Court valuable time (Clark 2003). Changes

 

court level. In the American context, a writ of certiorari serves to grant the applicant a hearing on the
merits before the Supreme Court.

25

to the High Court Rules in 2004 now allow the Court to determine special leave
applications on the papers only, if it is deemed appropriate by two Justices (Doogan
2004)

Until recently, the special leave process was not the primary way by which cases
came to be heard by the High Court. In its earlier years, the Court had a substantial non-
discretionary jurisdiction. The Court, for example, was obligated to hear appeals from
state supreme courts in civil matters involving £300 (Jackson 2002b). Appeal as of right
was also guaranteed in cases in which the outcome may affect a person’s status in terms
of citizenship, marriage, divorce, bankruptcy, and insolvency. In addition to the abolition
of these appeals as of rights, the creation of the Federal Courts allowed the High Court to
share its original jurisdiction with lower courts. These two developments, taken together,
have virtually eliminated the High Court’s non-discretionary jurisdiction. The High
Court has interpreted these acts of Parliament as a grant of complete discretion as to the
cases it hears (Solomon 1999). Thus, the High Court’s agenda became almost wholly

dependent on the special leave process. Although the practice was challenged in the

Smith Kline & French 19 case, the Court upheld the practice unanimously.

In the High Court, like the American Supreme Court, the refusal of a special leave
application is not equivalent to a denial of the appeal. A special leave application heard
by a panel of five or more Justices is often disposed of during the special leave process.
The High Court often grants special leave with specific terms. On some occasions, the
Court will require one litigant to pay the other litigant’s costs of appeal. Additionally,

special leave is subject to rescission at or before the hearing on the merits (Jackson

 

l
9 Smith Kline & French Laboratories v. Commonwealth (1991) 173 CLR 194.

26

2002b). In 1999—2000, the High Court decided 387 special leave applications, of which
66 succeeded. Due to the ever-increasing number of special leave applications being
heard, many hearings are now being conducted via a closed-circuit video link system.

Both Courts follow similar guidelines when deciding whether a case will be heard.
The American Supreme Court has written its own guidelines in Rule 10. These
guidelines reflect the role of a secondary appellate tribunal, which is to clarify important
points of law, and generally not to simply correct trial court errors. The criteria for the
High Court are set out not by the Court itself, but by Parliament. The provision in
Australia that corresponds to the Supreme Court’s Rule 10 is Section 35A of the
Judiciary Act 1903 (Cth). This section requires the Court to consider the need to promote
uniformity throughout the legal system, the public importance of the question, and the
individual-level interests of justice and fairness.

The Panel System

In the United States, it is rare to see a Supreme Court case on which fewer than
nine Justices sit. In Australia, however, this is a very common occurrence. Instead of
sitting en banc, the High Court sits in panels. Neither the Australian Constitution nor the
Judiciary Act gives much help in determining how these panels are assigned (Schubert
1969a). This panel tradition comes largely from two factors. First, the Court has always
operated on a circuit system. In the early years, the Justices seldom lived in the same

city, and travel was long, harsh and difficult. Because of this, it was much more efficient

to allow for a subset of Justices to comprise “The Court and exercise 1t5jurrsdrctron. 0

 

20 While the US. Supreme Court abandoned the circuit-riding practice after 120 years, the High Court of

Australia still travels to the various state capitals to hear cases each year. Because of the role circuit riding
plays in keeping the Justices abreast of the legal and societal differences across a large and diverse nation,

some American scholars argue that the Supreme Court should readopt the practice (Stras 2007).

27

Additionally, a High Court Justice was usually serving on the Privy Council at any given
time. This required one Justice to travel to England, often for months at a time. Instead
of putting the work of the Court on hold, the panel system allowed for the continued
completion of day-to-day business (McMillan 2003).

The size of the panels is dependent on a number of different factors. The Judiciary
Act and the High Court Rules require that a Full Court consist of no fewer than two
Justices. In appeals from the Full Court (or Court of Appeal) of a state supreme court, a
Full Court must consist of no fewer than three. Aside from these rules, the jurisdiction of
the Court may be exercised by any number of Justices (Brennan 2002a). A single Justice
in chambers may exercise the Court’s jurisdiction in cases dealing with practice and
procedure. The single Justice, when necessary, may refer a question to the Full Court.

Traditionally, appeals brought from decisions of single state supreme court judges
were heard by a panel of three Justices. Since the 1984 changes to the judicial system,
however, appeals from single state supreme court judges are extremely rare (Brennan
2002a). Most of the appeals from state supreme courts have now been heard by a Full
Court or Court of Appeals division of the state supreme courts, and are heard in the High
Court by a panel of five or more Justices. Special leave petitions may be heard by panels
of three Justices, but criminal case petitions are often heard by five Justices and disposed
of during the special leave process.

The assignment of panels is arranged by the Chief Justice. The Chief also assigns
Justices to a schedule of sitting in chambers or in original jurisdiction cases (Mason
2002d). While the Chief may organize the panel assignments, any Justice is entitled to sit

on any case if he or she so desires, although this privilege is seldom invoked (McMillan

28

2003). The Chief Justice typically selects panels of five Justices unless the case is
deemed especially important (McMillan 2003). Typically, these especially important
cases involve interpretation of the Constitution. In such cases, all available Justices are
assigned to the panel (Brennan 2002a). There is little information available about how a
Chief Justice might capitalize on this power, as Schubert observes:
The apparently available explanation is not very satisfactory: reference is made to
‘settled practices’ and to ‘common sense’ — which remain equally undefined — as
the basis for the exercise of discretion which takes into account such self-evidently

weighty measures as the ‘relative importance of a case’ and the number of j ustices
‘available’ for assignment. (Schubert 1969a, 10)

It seems clear that many contemporary Australian legal scholars do not see
evidence that the Chief Justices use the power to assign panels to affect the outcome of a
case (McMillan 2003). As the current Chief Justice has noted,

I’ve never heard any complaints by people about the system by which we select the

five to sit. I’m the one who proposes who sits, and it’s usually done on the basis of
evening out the workload. (Gleeson, as quoted in Clark 2003, 10)

Schubert (1969a), however, finds evidence that at least one Chief Justice——Dixon—did
not choose panels randomly.

As the jurisdiction of the High Court has changed, the number of Justices on the
average panel has also changed. With the elimination of non-discretionary jurisdiction,
the Court now hears more of these “especially important” cases, and far fewer cases that
are heard by three-judge panels (see Figure 3). According to Chief Justice Brennan,

The Court tries to ensure that the opinions of all available Justices are secured to

finally settle any questions of these kinds. It is unsatisfactory to have finely

balanced questions resolved by fewer Justices than a majority of the whole Court.
(Brennan 2002a, 61)

29

This mle of thumb, however, is not always followed perfectly. In Woods v. Multi-
Sport Holdings,21 only five Justices sat on the panel. Justice Kirby was not one of them.

In a subsequent case involving a similar question,22 Justice Kirby expressed his

disappointment about the majority decision in Woods, and implicitly about his absence
from the panel. In Ex Parte Aala, 23 Kirby noted that the seven-member panel might
have been overkill:
In this matter, this Court has been involved, not in the elucidation of some
important question of constitutional, statutory or other legal significance. The

applicable principles are clear. This Court has been engaged in nothing more than
the elucidation of the facts and the application to them of settled rules of law.

Although the size of the panel is related to the type of case involved, it is not clear that
the membership of any given panel is related to the Chier assessment of any individual
Justice’s ideology, predilections, or likely vote on a case (McMillan 2003). Issues like
illness, vacation, preparation for retirement, service on the Privy Council, and
geographical location may affect the panels, but Australian scholars anticipate that, as far
as they can tell, concern for efficiency dominates this process (McMillan 2003). Because
the most important cases tend to involve all of the Justices, the Chief Justice’s ability to
assign panel membership would not likely benefit him much on major policy decisions.

Hearing Cases on the Merits

Like the US. Supreme Court, the High Court rules on the merits of only a fraction
of the cases brought to it. The abolition of appeals as of right has allowed the Court the

discretion to choose cases to be heard on the merits. This has allowed the Court to hear

 

2’ (2002) 208 CLR 460.
22 Graham Barclay Oysters v. Ryan (2002) 208 CLR 460.

a
-3 Re: Refitgee Review Tribunal; Ex Parte Aala (2000) 176 ALR 219.

30

fewer and fewer cases on the merits over time (see Figure 4). While the procedure for
hearing and deciding cases on the merits is similar between the US. Supreme Court and

the Australian High Court, there are some important differences.

Figure 3 - Average Panel Size by Year

 

100
90 r
80 l
70 <
80 4
50 .
40
30 <
20 -

10*

0
1%4 '88 '88 '70 '72 '74 '78 '78 '80 '82 '84 '88 '88 '93 '92 '94 '98 '98 '00

 

 

Number of Cases

 

Submission of written briefs

Written briefs are a relatively new addition to the High Court’s decision-making
procedure. It was not until the 19505 that the Court provided a non-compulsory
procedure for the submission of written cases or detailed briefs. Before this, the Court
relied almost exclusively on oral arguments. Chief Justice Latharn applauded the
measure, indicating that it would increase clarity and efficiency (Bennett 2002).

In 1982, the Court began to require barristers to present rough outlines of their
arguments to the Justices at the beginning of arguments (Brennan 2002b). The
submission of cases fully briefed “on'the papers” prior to argument was finally mandated
by the High Court in 1997. This was largely an attempt to increase the Justices’

understanding of the cases prior to oral argument. In recent years, the Court has

31

increasingly adopted the stance that oral argument must follow from the information
provided in the written briefs (Young and McLeish 2004). Especially when hearing
appeals on the merits, though, the Justices do not see the cases on the papers as a

replacement for oral argument (Leigh 2000).

Figure 4 - Total Number of Cases by Year

 

 

 

too
90-
so«
70-
sol
501
any

0')

(D

8 30«

o

‘5 20«

a

g 10'

z 0

 

1964 a as '70 '7'2 '7'4 is “is ab '62 '54 as '88 '50 '52 E {as as do

As in the United States Supreme Court, litigants are not always the only parties to
submit briefs. Interveners and amici curiae are less common in the High Court than in
the US. Supreme Court. As in the United States, amici curiae do not become actual
parties to the proceedings. Instead, they appear before the court or submit briefs in an
attempt to assist the Court when the actual litigants are unwilling or unable to provide
certain information. In the High Court, interveners become full parties to the
proceedings, although sometimes limited to specific issue areas.

Until recently, there was no statutory right to intervene. Instead, leave to intervene
was granted at the discretion of the Court. In 1976, amendments to the Judiciary Act

gave attomeys-general of the Commonwealth and the states the right to intervene in cases

32

where constitutional interpretation is involved (Burrnester 2002). Now, government
interveners are nearly always granted leave to intervene or appear as amicus curiae. The
Court, however, retains its discretion as to whether or not to allow other parties to
intervene. Traditionally, the Court has been reluctant to allow interest groups or other
non-govemment parties to intervene (Williams 2000). As a result, most of the
intervening appearances and written brief submissions occur in constitutional cases,
usually about the distribution of power between the Commonwealth and the states (Parish
2000)

In the United States, the Justices on the Supreme Court have four clerks apiece, and
these clerks have a crucial role in processing and summarizing incoming briefs (Perry
1991). Because of the huge workload, much of the Supreme Court’s work is delegated to
law clerks. In Australia, however, this is not the case. Although a High Court associate’s
job description is similar to that of a Supreme Court clerk, Australian Justices tend to rely
on them less (Leigh 2000). One reason for this might be the smaller workload. The
associates may create summaries of written briefs, but this is not the norm. This means
that the associates have less influence at this stage of the process than their American
counterparts (Leigh 2000).

Oral arguments

After the submission of the written briefs, litigants present oral arguments to the
Court. The High Court spends an average of about one day hearing oral arguments for a
single case. This is quite extensive, especially when compared to the two hours spent per
case by the United States Supreme Court. Chief Justice Gleeson thinks that this
comparison is not very helpful, though, because the US. Supreme Court does not “have

our tradition of oral advocacy” (Clark 2003, 10). While most of the work of the highest

33

courts of Canada and the United States are done “on the papers,” the Australian judiciary
has adopted the British focus on oral arguments over written briefs (Leigh 2000). As
Chief Justice Brennan notes,
The practice of the Court has been to rely heavily on oral argument. The dialectic
of advocacy and exchanges between Bench and Bar illuminate the issues for

decision and usually identify the material on which the decision is based. (Brennan
2002b,197)

For this reason, the High Court has declined to adopt formal time limits for oral
arguments, except in the case of special leave applications (Bennett 2002). Despite the
lack of formal limits, the length of oral arguments has declined over time. Before written
briefs were commonplace, the Court relied almost exclusively on oral argument. The
evolving process of getting access to briefs prior to oral argument has helped to shave
significant time off of the actual hearings (Brennan 2002b).

This focus on oral argument makes it far more relevant to the decision-making
process than is typically ascribed to oral argument in the United States (Glendon et a1.
1999). Although some research has shown oral arguments to be marginally important in
the US. Supreme Court (Johnson 1997), the brevity of these arguments necessarily limits
the amount of information the Justices can expect to extract from the parties. Chief
Justice Gleeson observes of the US. Supreme Court system that “most of their work is
done on the papers” (Clark 2003, 10). Nonparty litigants are rarely allowed to participate
in oral arguments in the US. Supreme Court (Segal and Spaeth 2002). The lack of time
limits on High Court oral argument leaves significantly more time for the Justices to
engage counsel. It is perhaps for this reason that the High Court does allow for oral

presentations from interveners, and less often for amici curiae (Burrnester 2002).

34

The courtroom environment during oral argument has changed over time, largely
influenced by the composition of the Bench and the leadership of the various Chiefs
(Bennett 2002). At times, the Bench has been known for intense interrogation of counsel,
rarely allowing them to finish a sentence without interruption. Justice Dixon resolved to
curtail such judicial Socratic methods, but it was revived under Chief Justice Barwi ck
(Gageler 2002c). Because the Justices use oral arguments as an important part of the
decision-making process, counsel will often be faced with difficult questions about the
quality of their arguments (Hayne 2004). For this reason, barristers are encouraged by
the Bar to “turn judicial questioning, including in particular hostile questioning, to his or
her advantage” (Young and McLeish 2004, 16).

Some J ustices—both past and present—have been exceptional in their displays of
wit and humor on the Bench (Davis and Simpson 2002). Others, however, find such
banter distasteful and distracting (Ackland 2003). Gleeson has argued that the display of
humor on the bench can have unintended consequences, largely because “judges and
legal practitioners may underestimate the seriousness which litigants attach to legal
proceedings, and they can become insensitive to the misunderstandings which might arise
if the judges appears to be taking the occasion lightly” (Gleeson 1998). Additionally, the
Justices expect counsel to refrain from attempting to appeal to the emotional or otherwise
extra-legal sensibilities. As Justice Hayne warns, “the Court is not a jury let alone a talk-
back radio announcer's audience” (Hayne 2004).

The spontaneous nature of the oral argument process, however, often produces
comedy despite attempts at restraint. Some argue that this judicial humor is an

inextricable part of the judicial decision-making process (Heerey 2004). Common

35

examples are wry comments to counsel about the poor quality of their arguments (Davis
and Simpson 2002) or playful jabs at their fellow Justices (Ackland 2003). In a recent
case,24 for example, counsel for the appellant produced a pistol as an exhibit. Justice
Gleeson commented to counsel: “If you are going to point it at us, would you mind
pointing it in the direction of Justice Callinan?” While humor might be taken to an
extreme in the judicial context, it can be used to relieve the stress and frustration of the
courtroom environment when used judiciously (Davis and Simpson 2002).

The Decision and Reasons

The decision-making process in Australia is much different from that of the US.
Supreme Court. Because of the influence of the British system, the Justices follow a
seriatim tradition of individual opinion writing. This has important consequences for the
process of negotiation and opinion creating. In addition, the traditions and norms on the
Court make the opinion-writing effort look foreign to American onlookers. The
differences, in turn, necessarily change the motivations driving the decision—making
behavior of each individual Justice.

The seriatim opinion-writing tradition

Like the early US. Supreme Court, the High Court of Australia follows a seriatim
opinion-writing tradition. Latin for “one after the other,” the seriatim tradition calls for
individual opinions or reasons to be produced by each judge. This method of individual
opinions was inherited from the English tradition. Under its first Chief Justice, John Jay,
the US. Supreme Court adopted this procedure. Under Chief Justice John Marshall,

- however, the tradition was abandoned in favor of a single per curium opinion (Ginsburg

 

24 Gillard v. The Queen (2003) 219 CLR 1.

36

1994). Only one opinion of the Court was typically handed down, and any dissenters
remained silent (Gleeson 2003b). This change was largely an effort on the part of
Marshall to reduce the appearance of discord and to protect the Court’s legitimacy
(Seddig 1991).

This practice was not universally applauded. As Thomas Jefferson wrote in a letter
to a sitting Supreme Court Justice,

The practice is certainly convenient for the lazy, the modest and the incompetent.

It saves them the trouble of developing their opinion methodically and even of
making up an opinion at all. (Jefferson 1822, 1)

Indeed, the practice of publishing a single majority opinion was later abandoned in the
Supreme Court in favor of a hybrid system. While the Supreme Court retains the
majority opinion, other Justices are free to express concurring or dissenting views in
separate opinions. The majority opinion still maintains importance, though, because
precedential weight is only attached to opinions joined by a majority of the Justices.

The High Court, however, never strayed from the British seriatim opinion-writing
process. Although there have been intermittent attempts to consolidate the majority into
a single opinion, these attempts have always been short-lived (Coper 2002c). Instead, the
rugged individualism of the Court and its members has encouraged them to produce
individual opinions (Mason 2003c). While Justices do occasionally join opinions, the
solitude and independence of a barrister’s work likely works to feed the seriatim opinion
process (Wheeler 2003).

This process, however, is not without its own difficulties. In terms of clarity of
law, the tradition of producing multiple “majority” opinions makes discerning the ratio

decidendi far more difficult. As Chief Justice Mason observes,

37

The Court is a collective institution; it is the Court, rather than the individual
Justices, that decides the case and declares the law. There is now an expectation
that the Court will declare the law in clear terms in order to provide guidance for
trial judges, the legal profession and the community. (Mason 2002d, 91)

This tradition does provide individual accountability, but provides little help for those
wishing to discern the common thread among majority opinions. Though there may be a
majority agreement on the result, the reasoning used to get to the decision can vary
widely (Willheim 2002). While legal academics like the intellectual challenge of
discerning the ratio decidendi, practitioners often lament this tradition (McMillan 2003).

Judicial conferences

The US. Supreme Court has long followed a tradition of formal judicial
conferencing. In some of these conferences, the Justices meet to discuss cases heard on
the merits. After a discussion of the case at hand, apparently proceeding in descending
order of seniority, the Justices cast a vote (Segal and Spaeth 2002). The most senior
Justice in the majority assigns the authorship of the majority opinion, and discussion
moves to the next case. The most junior Justice opens the door for the refreshments,
answers the telephone, and keeps records of the conference proceedings (Ginsburg 1994).

In Australia, however, the process is far more obscure. It has varied wildly over
time, largely dependent on the personalities of the Justices and the leadership of the Chief
Justice (Simpson 2002). The closest thing to opinion assignment probably happened
under Chief Justice Latham after the appointments of F ullagar and Kitto. The Privy
Council looked unfavorably on this practice, and it soon faded. The practice of holding
conferences waxed and waned over the years. It seems as though incoming Chiefs were
likely to attempt to begin such a tradition, but most abandoned it altogether in short stead

(Mason 2002d).

38

Some former High Court associates indicate that there is no formal conference
process. Following the extensive oral arguments, oftentimes the Justices seem to
understand where their colleagues will stand. Recent alumni indicate that there is now a
monthly “judgment meeting.” Indeed, Chief Justice Gleeson has acknowledged this
monthly meeting, as well as informal meetings over tea (Clark 2003). While the process
remains secretive, the result of the monthly conferences is called the “judgment list.”
This is a list of the matters before the Court along with the status of all active business.
The list indicates who has written or joined opinions. After the business is done, these
lists are destroyed. One thing that is certainly missing from this list — if it does exist — is
a majority opinion assignment.

The writing and circulation of drafts

In the US. Supreme Court, the process of writing and circulating drafts is very
important. Because of the weight a majority opinion holds, there is a high level of
motivation for its author to recruit other Justices to join. Many of the Justices permit
their clerks to write the first draft (Tarr 2003). The Justice will then edit the first draft,
distribute it to his peers and respond to any concurring or dissenting opinion drafts that
have been circulated. During the course of this process, Justices may change their votes
or decide to write concurring opinions.

The process in the High Court is somewhat different, due largely to the different
opinion writing tradition. When it comes to opinion writing, High Court Justices work
« more like a set of independent law firms than a collegial court. It is very rare that a
Justice would allow an associate to create the first draft of the opinion (Craske and Haigh

2002). The Justices edit this draft to their liking, and then circulate it to the other

39

members of the panel. Oftentimes the original drafts will later be edited based on

discussions with colleagues, or in response to other circulated opinions.

Figure 5 - Percentage of Cases with at least One Joint Opinion

 

 

 

100
80-
60-

8

U') 40-

or

O

'5

8)

m 20h

E‘

a)

8

61’ U f

 

1954 as as 7'0 '7'2 '71: "is 'is 'so at '84 'as as so '52 '54 '55 as '60 2002

This is not to say that the opinion circulation process is not important in the High
Court. Clearly, the ability to have the first opinion in circulation allows a Justice to frame
the case according to his perspective. This gives the first opinion author the
benefit of influencing the way his peers interpret the case (Kirby 2003). The other panel
members may decide to ad0pt some—or the entirety—of the first author’s reasoning.

The pressure to negotiate and bargain for agreement, however, is almost non-
existent in the High Court. Because the majority opinion will have no less precedential
value if it is divided among various individual opinions, the Justices may express their
particular thought process without worry of losing support for the majority outcome.
Although there is no tradition of a single majority opinion, the Justices do join with their

colleagues. Sometimes this involves actually signing multiple names to a single opinion.

40

In this situation, it is impossible to discern which Justice is the main author.25 Other
times, a Justice may submit a very short opinion expressing concurrence with the reasons
and result of another’s opinion. When considered together, these two types of j oint
opinions occur considerably often on the Court (see Figure 5). On average, 30.2% of all
Opinions handed down are joint opinions. Between 1965 and 2002, an average of 77% of
cases had at least one joint opinion. As Figure 5 illustrates, however, joint opinions are
becoming increasingly common.
Proliferating the decisions

As in the US. Supreme Court, one of the most important duties of the High Court
is its proliferation of precedent through written opinions. The High Court’s authorized
reporter is the Commonwealth Law Reports (CLR). Volumes of the CLR have been
published privately since its inception in 1903. The CLR has contained select decisions
of the High Court, as well as Australian cases decided by the Privy Council in London.
Additionally, cases can be found in several other places, including the Australian Law
Reports, the Australian Law Journal Reports and more recently the Australasian Legal
Information Institute website26 and the High Court’s official website.27

Editors of the various reports have always reported only a selection of the High
Court’s business (Meek 1999). Only cases deemed “important” are typically included.
The presence of constitutional interpretation is a flag for the editors, and these cases are

always included (Jones 2002). The editors, however, have not always been able to

 

25 The lack of an authorship designation, however, has not stopped legal scholars from attempting to
discern which of the Justices penned particular joint judgments.
26 ..

http://www.austlrr.edu.au

27
http://wwwhcourt. gov.au

41

appreciate the future importance of a case. In an attempt to fill in the blanks, the CLR
issued a special volume in 1994. This volume included cases from up to thirty years
earlier that had gained importance in High Court jurisprudence over time.

As the Court’s discretionary jurisdiction has increased dramatically over the years,
a higher percentage of the cases handed down have been reported. The practice of
handing down ex tempore or “spur of the moment” judgments has declined, as the agenda
has become more heavily weighted to difficult legal and constitutional issues (Kirby
1992). Currently, nearly all judgments of the High Court are reserved judgments, or
judgment accompanied by written opinions handed down some time after oral arguments.
These reserved judgments provide more detailed reasons, and are thus more suitable for
inclusion in the reporters.

As the proportion 0 “important” decisions has increased, so has access to the
entirety of the High Court’s business. In 1980, legal scholar David Solomon began
publishing The Legal Reporter, in which he included all of the High Court’s judgments
(Jones 2002). Additionally, since 1997, the High Court itself has made available the full
text versions of all opinions on its official website. The media coverage of the Court has
increased over time, and the major Australian newspapers each have journalists who
spend a significant portion of their time covering the High Court’s business (Williams
1999)

Practice and Procedure in Comparative Perspective

The High Court and the US. Supreme Court are alike in many respects. Each court
sits at the apex of a British common law-inspired system, and each is charged with
interpreting and clarifying the law for its respective judiciary. Each exercises

discretionary jurisdiction, and they are both understood to be important political as well

42

as legal institutions. In some important ways, though, these two institutions differ. These
differences are critical for conceptualizing the ways in which the behaviors exhibited by
members of the High Court bench might be different from what we know about US.
Supreme Court Justices. Differences in judicial selection, jurisdiction, agenda setting and
opinion production must be considered before an attempt is made to use existing judicial

behavior literature to explain the production of j oint opinions on the High Court.

43

CHAPTER TWO: A BRIEF HISTORY OF THE HIGH COURT OF AUSTRALIA
Judicial biographies in Australia are as rare as hen ’s teeth.
— Justice Michael D. Kirby (1998a, 1)

The High Court of Australia shares much in common with the American Supreme
Court. From a number of distinct state systems of courts, the enactment of a constitution
created a new federal jurisdiction. The High Court, like the US. Supreme Court, has
seen periods of calm, times of turbulence, and a number of very interesting characters. In
order to understand the functioning of the contemporary High Court, it is important to
conduct at least a cursory review of the High Court’s history. This history can help to put
various current issues into context. Additionally, the understanding of the behavior of
recent members of the bench can be illuminated by examining the tradition of decision
making and interpersonal relations among Justices throughout the Court’s hundred-year
history.

The Early History of the High Court

The High Court of Australia first sat on October 6, 1903. This first sitting took
place in Courtroom Number One of the Melbourne Supreme Court, and it was met by
relatively little popular excitement. It was reportedly a slightly awkward affair, and one
Melbourne newspaper quipped that “the most experienced man in the court was the crier”
(Bennett 1980, 23). That there was little public attention, however, did not mean that the

opening of the Court was not the subject of intense debate among lawyers and politicians

(Galligan 1987).28

 

28 The Court’s first case, Hannah v. Dalgamo (1903) l CLR I, asked the Justices to decide when the
Court actually came into existence. The decision on the merits hinged on whether the Court was in
existence as of the ratification of the Constitution on January 1, 1901, or not until the passage of the
Judiciary Act in August of 1903. The Court ducked this question, instead rescinding the grant of special

44

The eventuality of a supreme court of appeal was taken without much comment
during the process of drafting the Commonwealth Constitution (Bennett 1980). In the
late 18905, the most pressing disagreements had to do with the maintenance of appeals to
the Privy Council. By 1903, however, additional serious complaints were lodged during
the Parliamentary debate over the Judiciary Bill, which would create the High Court
(McHugh 2002). Among these concerns were gloomy prophecies that the Court would
be a waste of resources and would have little or no work to keep it occupied (Bennett
1980). Many observers also assumed that the long-established state supreme courts,
coupled with the Privy Council, would eclipse the High Court in prestige and importance.
In this way, the High Court would be rendered largely impotent (Bennett 1980). In the
years that followed, however, the Court would prove its critics to be profoundly
mistaken. In a series of measured steps and giant leaps, the High Court ultimately
cemented itself as a competent and powerful tribunal.

Establishing an Identity (1903-1919)

The personalities on the first High Court bench—along with impassioned pleas
from the ambitious Attorney-General Alfred Deakin—helped to assuage much of this
criticism (McHugh 2002). Indeed, the three men selected by the Protectionist Party
government were quite distinguished. All had worked together in various capacities to
draft the Commonwealth Constitution, and were all largely in agreement as to the

political and economic role of the newly constituted government (Bennett 1980).

 

leave on the grounds that the negligence question at hand “was not a question of sufficient public
importance” (Coper 2004).

45

The Chief Justice, Samuel Griffith, was the Premier of Queensland prior to

Federation. Griffith was widely seen as the main architect of the constitution.29 While
some of his contemporaries thought him too self-serving and ambitious for the position
(McHugh 2002; Clark 2003), his appointment was generally heralded by the press as a
positive step (Gibbs 2002). He had proven himself a great legal mind in his role as the

Chief Justice of the Supreme Court of Queensland.

The puisne Justices30 were also among the most upstanding of Australian citizens.
In fact, many legal scholars — past and present — argue that either one would have been a
superior choice for the position of Chief Justice (McHugh 2002). Richard O’Connor had
been solicitor-general for the state of New South Wales and was the first Leader of the
Government in the Commonwealth Senate. Edmond Barton resigned his position as the
first Prime Minister of the newly constituted Commonwealth of Australia in order to
serve on the High Court. Both O’Connor and Barton were chosen for the position by
Attorney-General Alfred Deakin. Deakin was criticized for his choice of Barton, as
onlookers correctly anticipated that Deakin wished to succeed Barton as Prime Minister
(Bennett 1980).

During the first few years of its existence, the High Court made a significant mark
on the Australian legal world. This was due largely to the leadership of its first Chief
Justice (McHugh 2002). The members of the Bench enjoyed a friendly personal rapport,

and these agreeable relationships translated into a high degree of ideological unanimity in

 

2 . . . . . .
9 According to Botsman (2000), It was not until John Reynolds’ 1958 publication of Andrew Inglis
Clark’s draft constitution that Griffith’s role has been diminished, although only minimally so.

0 The term puisne (pronounced ‘puny’) is used to refer to the ‘junior’ judges, or any High Court Justice
other than the Chief Justice (Blackshield 2002f).

46

the Court’s output (Simpson and Simpson 2002). Each of the members of the first
Griffith Court was dedicated to the protection of state’s rights (Botsman 2000). The
puisne judges typically assented to opinions, which were mostly written solely by Griffith
(Gibbs 2002).

The Griffith Court managed to prove wrong those High Court critics who
anticipated a weak and useless tribunal. In fact, Griffith managed to solidify the Court’s
position as the paramount Australian legal institution in just a few years (Gibbs 2002).
The first Court successfully confronted the Attorney-General (Priest 2002), the
Commonwealth Parliament (Mason 2002b), and even the Privy Council (Bennett 1980).
The Australian political climate was strongly majoritarian at the time; citizens and
politicians alike became wary of the large role the High Court was assuming, especially
in exercising judicial review of legislation and executive action (Galligan 1987). Many
onlookers, especially Commonwealth Parliamentarians, began to wish for the impotent
tribunal against which they once campaigned. It was the Commonwealth Parliament
itself, in the Judiciary Act 1903 (Cth) that provided for the High Court’s power of judicial
review (Mason 20020). While these concerns were not center stage in the parliamentary
debates over the bill, they began to reveal themselves as the Court began to assert its
substantial role (Bennett 1980).

By 1906, the Court’s workload had increased such that its members requested an
enlarged bench. Their request was answered, and the Protecti oni st Party government
appointed Justices Isaac Isaacs and Henry Higgins (Bennett 1980). With this addition,
however, the personal and professional goodwill on the Bench deteriorated markedly

(Gibbs 2002). The main point of professional contention was the intergovernmental

47

immunities doctrine established by the foundation members of the Court (Mason 2002b).
This doctrine held that the Constitution contained an implied protection of the reserved
state powers against Commonwealth interference, and vice versa (Coper 2002b). This

arrangement was derived from American Supreme Court principles in cases like

McCulloch v. Maryland (1819),31 and followed nicely from Griffith’s vision of the
appropriate structure of coordinate federalism (Zines 1997). In its first year, the High

Court cited the US. Supreme Court’s decision in McCulloch when it struck down a state

tax on a federal officer’s income.32 The newer Justices disagreed, and began to write a
large number of dissenting opinions (Gibbs 2002).

Part of the problem centered on the method of constitutional interpretation
instituted by the Court’s founding members. The reasoning behind the doctrine of
intergovernmental immunities involved distinctly political considerations (Patapan
2000c). This was largely a result of the characters on the Bench at the time. The first
three Justices were all senior statesmen, heavily involved in the creation of the federal
system found in the Constitution (Galligan 1987). Griffith, Barton and O’Connor shared
a common political vision, namely of a strong system that divided powers between the
states and the Commonwealth (Patapan 2000c). For this reason, they often agreed on a
balanced, political approach to distributing powers between levels of government.
Because the first Justices were among the framers of the document, their interpretive
approach “relied more heavily on the sagacity and authority of the Justices than on

logical consistency or a literal construing of the constitution’s text” (Galligan 1987, 80).

 

31 17 US. (4 Wheat) 316.

32
D 'Emden v. Pedder, 1 CLR 91 (1904).

48

They tended to approach the question as politicians allocating power, and not as lawyers
interpreting the law.

Neither Isaacs nor Higgins accepted the doctrine of intergovernmental immunities
established by the founding members of the Court (Cowen 1955). There is some
scholarly disagreement, however, on the substance of their complaint. It is clear that
Isaacs and Higgins were both involved in the framing of the constitution. Some argue,
however, that these men saw the Constitution in terms of its legal consequences, and not
in political terms (Goldring 1997). According to this view, their legalistic perspective
kept them from agreeing with the holistic interpretation style of the founding members of
the Bench, instead preferring a strictly legal approach (Patapan 2000c).

Others argue that it was not the interpretation style that troubled Higgins and
Isaacs, but the political vision that was supported by its use. Galligan notes that “they
were both ardent federalists intent upon expanding commonwealth powers,” 3 position
distinctly opposed to the existing High Court doctrine (Galligan 1987, 84). Isaacs and
Higgins were both unimpressed by the doctrines of immunity and reserved powers.
Higgins was particularly and vocally critical of the doctrine of implied immunities, and
took to berating American Chief Justice John Marshall for his landmark opinion on the
issue (Zines 1997).

Soon after their appointments, Isaacs and Higgins also expressed dismay at the

methods of constitutional interpretation favored by the other members of the Bench. In

Baxter ’3 Case,33 the Court was charged with determining the extent to which the High

Court must be bound by the decisions of the Privy Council. The majority in this case

 

33 (1907) 4 CLR 1087.

49

refused to apply the Privy Council’s decision in Webb v. Outtrim (1907).34 Griffith’s
majority opinion in Baxter’s Case, joined by Barton and O’Connor, embraced a
contextual strategy for interpreting the constitution:
There is no doubt as to the meaning of the words used, but the circumstances under
which the power was intended to be exercised must be discovered from some other
source. That source is to be found in a consideration of the whole purview of the

Constitution, and the answer to the question cannot be given without having regard
to its history. (Baxter’s Case, 1103)

In his dissenting opinion, Isaacs argued that such external information is at the very
least unnecessary in this case:
I know of no clearer words in the Constitution, or words having a more obvious

purpose. Sec. 74 is one of the pillars of the Constitution. Unless it stands firm,
much of the true meaning of that document is lost. (Baxter ’5 Case, 1151-52)

Higgins conceded that he would agree with the majority as a statesman, but
asserted that his role on the Bench required something decidedly different. In his words:

I admit that, before taking my seat on this bench, I thought that it would be wiser to

leave, as far as possible, the interpretation and the application of Australian laws to

Australian Courts But my duty here is to accept the law as I find it. (Baxter’s
Case, 1166)

This increase in disagreement on the Court was a matter of great concern for the
press (Bennett 1980). There were only four dissents before the expansion of the Bench in
1906 (Lynch 2002b), but the appointment of Isaacs and Higgins led almost immediately
to an increase in individual concurring opinions and, eventually, dissenting opinions
(Mason 2002b). This new tradition of disagreement and individual opinion authorship is
one that continues to the present day (Lynch 2002b). Griffith’s hold on the direction of
Court doctrine was further diminished with the replacement of O’Connor by Frank Gavan

Duffy and the 1913 expansion of the Bench (Mason 2002b). Charles Powers and George

 

34 (1906) 4 CLR 356.

50

Rich35 filled the two newly created seats on the Court. All three of these Justices were
appointed by the Australian Labor Party government. By the time of Griffith’s
resignation in 1919, the Court’s thinking had changed substantially, especially in terms of
constitutional interpretation (Bennett 1980).

Growing Pains (1919-1950)

Settling on a replacement for Griffith was difficult. Griffith’s resignation date was
postponed twice as the government debated whether to elevate a sitting puisne Justice or
to choose from outside the Court (F ricke 1999). The Cabinet preferred to give the job to
Justice Barton, but his old age and ill health led Griffith to recommend against the idea.
Finally, the Nationalist Party Cabinet settled on Adrian Knox. Shortly thereafter, Barton
died. Hayden Starke was appointed by the Nationalist government to replace him.
Although Knox was the Chief, Isaacs had a commanding personality and largely
dominated the Court during this time (F ricke 2002d). On the Griffiths Court, Isaacs was
a frequent dissenter, but things had changed by the early 1920s. Isaacs’s expansive view
of Commonwealth power began to win a majority on the Court (Kirby 2005b). He
penned a long line of joint opinions, to which Knox, Rich, and Starke generally
concurred.

In the aftermath of World War I, the Commonwealth government was enjoying a
strengthened position in Australian federal politics. The founding members of the Bench
had generally sought to protect the states from the Commonwealth’s encroachment into

their reserve powers (Bennett 1980), but had also interpreted the Commonwealth defense

 

35 Before Rich’s appointment, Albert Bathurst Piddington was appointed to the second new seat. He
resigned before taking his seat, however, amidst criticism that neither he nor Powers were of the keenest
contemporary legal minds (Bennett 1980). Rich was then appointed to replace Piddington. Powers did not
seem to be dissuaded by the popular criticism.

51

power in extraordinarily broad terms (Galligan 1987). The expansion of Commonwealth
defense power, followed with the introduction of Justices who favored the expansion of

Commonwealth power over the states, led to what is largely regarded as the foundation of

Australian constitutional law: The Engineers’ Case.36

The Griffith Court had largely approached constitutional interpretation in terms of
political and legal context, and the doctrines of intergovernmental immunities and
reserved state powers were the results. Engineers’ swept these doctrines aside and
replaced them with a doctrine derived from a literal approach to constitutional
interpretation (Blackshield 2002d). Substantively, the decision gave the commonwealth
government the power to bind the states under legislation passed concerning the
conciliation and arbitration of industrial disputes. The joint opinion found no legal basis
for the Court’s earlier decisions protecting the states from the legislative power of the
Commonwealth (Bennett 1980). While this result might have been achieved by
augmenting the doctrines of immunity and reserved powers, the Court dismissed these
doctrines altogether, claiming:

They are sometimes at variance with the natural meaning of the text of the

Constitution; some are irreconcilable with others, and some are individually rested

on reasons not founded on the words of the Constitution or any recognized

principle of the common law underlying the expressed terms of the Constitution but

on implication drawn from what is called the principle of ‘necessity’, that being

itself referable to no more definite standard than the personal opinion of the Judge
who declares it. (Engineers ’, 141-42)

This decision served to greatly expand the Commonwealth govemment’s
legislative powers under section 92 of the Constitution (Zines 1997). This is especially

true in regard to trade, commerce, and arbitration (Fricke 2002d). Engineers ’ marked the

 

36
Amalgamated Society of Engineers v. Adelaide Steamship C 0. Ltd. (1920) 28 CLR 129.

52

introduction of a newly legalist approach to constitutional interpretation, and largely
resulted from the replacement of the early politicians on the Court with purportedly
apolitical lawyers (Galligan 1987; Patapan 2000c). Engineers ’ reliance on legalism,
however, is seen by some as a strategic decision rather than a philosophical one:
One cannot escape the view that in the Engineers’ case, as elsewhere, reliance on
‘express provisions’ and the ‘golden rule’ of interpretation was emphasized because

it suited a purpose and was a means of advancing the vision he had of Australia and
its governmental institutions. (Cowen and Zines 2002, 426)

Whatever the reason it was employed, however, legalism remained the dejure method of
interpretation on the Court for decades to come. As one scholar notes, “Engineers ’ can
be seen as the decisive juncture where conventional rules of interpretation asserted their
dominance over the innovation of federalism” (Patapan 2000c, 250). Justice Kirby
agrees with this assessment, arguing that Isaacs merely used the rhetoric of the legalistic
style of interpretation “and applied it with uncompromising rigour” (Kirby 2005b, 27).
During the Knox Court, several procedural changes were introduced. Attorney-
General John Latham proposed an amendment to the Judiciary Act. This amendment
provided pensions for Justices on the High Court. It also discussed the High Court’s need
for a permanent home in the seat of government, although action would not be taken on
this recommendation for more than fifty years (Bennett 1980). The Court also adopted
several changes to the High Court Rules, the most memorable of which was the method
for conducting the delivery of opinions. Instead of requiring the Justices to read the text
of their written opinions aloud in open court, these changes allowed the Justices to simply
state their opinion briefly and then deliver the written reasons to the clerical staff in open

court (F ricke 2002d).

53

As the 1920s drew to a close, Isaacs lost his hold on the majority of the Court
(Fricke 2002d). Subsequent challenges to state law under section 92 were largely
handled through statutory interpretation as opposed to constitutional interpretation.
Many of these cases began to whittle away at the extreme centralization of government
power advocated by the Engineer ’s decision (F ricke 2002d). With the death of Higgins
and the resignation of Powers in 1929, Isaacs’ views became more isolated (Fricke
2002d). Owen Dixon was appointed by the Nationalist government to replace Higgins,
but the Labor Party was back in office before Powers could be replaced. Soon after,
Chief Justice Knox retired.

The new Labor government appointed Herbert Evatt and Edward McTieman as
replacements for Knox and Powers, and elevated Isaacs to the position of Chief Justice.
Evatt was, at age 36, the youngest ever to be elevated to the High Court (Bayne 2002).
McTieman and Evatt had both been involved in Labor politics, and the pair of
appointments attracted significant criticism from the media (Bennett 1980). Neither
Evatt nor McTieman sat on the Court during Isaacs’ tenure as Chief Justice, as he would
remain in this position for a mere 42 weeks (F ricke 2002c). Effectively, the Isaacs Court
was a Bench of five. As the Great Depression took hold, the once steady flow of taxation
cases was replaced by an increasing number of bankruptcies (F ricke 2002c). Overall,
though, the work of the Court diminished significantly during this period.

As Isaacs was uncharacteristically ill during much of his Chief Justiceship, the
considerable influence of Owen Dixon was allowed significant space to flourish (F ricke
2002c). Isaacs resigned in 1931 in order to accept the vice-regal position of Govemor-

General. Despite rumors that Evatt would be elevated to succeed him, the position

54

instead went to the seventy-eight year old Gavan Duffy.37 Some historians believe that
this choice was the Labor govemment’s reaction to criticism of its previous
appointments, which were perceived as politically motivated (F ricke 2002a). During the
same year, under pressure to economize, the Labor government officially reduced the
number of seats on the High Court Bench from seven to six, leaving unfilled the empty
seat vacated by Isaacs (F ricke 2002b). The circuit system, which required that the judges
travel regularly to each of the state capitals, was suspended during this time in an attempt
to save money (Del Villar and Simpson 2002).

The Gavan Duffy Court was unspectacular by most accounts. Tensions rose as
alliances between the Justices shifted, and the rivalry between the Court’s two leaders,
Dixon and Evatt, flourished (F ricke 2002b). The work of the Court was still dominated
by Depression-related litigation, including many bankruptcy, taxation and corporations

law cases. Gavan Duffy’s retirement in 1935 led to the appointment of John Latham by

the United Australia/Country Party coalition government.38 For the first time since
Griffith’s departure, the Court enjoyed a capable and effective Chief Justice (Cowen
2002)

The Latham Court enjoyed an unmatched measure of membership stability. All
four of the longest-serving Justices sat during this time, and the Court saw only one
retirement and two appointees during the fourteen year period (Douglas 2002). Evatt .
retired in order to stand for the 1940 commonwealth parliamentary elections. He was

successful, and went on to serve as Attorney-General and Minister for External Affairs

 

37 Frank Gavan Duffy may have been seventy-eight years on the earth, but had only celebrated 20
birthdays, as he was born on February 29, 1852,

38 . . . . .
The United Australia Party changed rts name to the Liberal Party 111 1944.

55

for the Labor government until 1949 (Bayne 2002). He was replaced on the Court by
Dudley Williams, a United Australia Party appointee. William Webb was appointed in

1947 by the Labor government to fill the newly reinstated seat on the High Court bench,

bringing the Court back to its previous size of seven.39 Additionally, Justices Latham
and Dixon each took leave from the bench in order to serve as Foreign Ministers
(Douglas 2002).
The stability of membership on the High Court, however, was accompanied by
stable conflicts between members of the Court. As Roger Douglas explains,
The most bitter conflicts were those between Starke on one hand and the two
former Labor politicians, Evatt and McTieman on the other — but Starke was also
resentful towards Dixon, somewhat contemptuous of Rich, and cool towards

Latham. Evatt reciprocated Starke’s hostility and his relations with Latham could
sometimes be prickly. (Douglas 2002, 421)

This severe fragmentation effected a notable decline in the frequency of joi nt
opinions, which until this time had been relatively frequent (Cowen 2002). Votes
dissenting from the case outcome were present in 42% of cases during the Latham Court,
and in an astounding 72% of cases in 1944 (Blackshield 1972). This, coupled with the
even number of judges on the Bench until 1946, meant that a large number of decisions
were handed down by an evenly divided Court (Bennett 1980). The idea of holding
conferences among the Justices was abandoned as an impossibility, in great contrast to
the daily lunches enjoyed by the Court’s first three members (Bennett 1980). Starke

seems to have been the instigator of much of this trouble, refirsing to circulate draft

 

39 The relevant amendment to the Judiciary Act 1903 (Cth) was introduced by the Commonwealth
Attorney-General, Herbert Evatt. Thus, the new appointment would be made by the Labor government, of
which he was a member.

56

opinions and making “no attempt to conceal his personal animosities” (Cowen 1967,
117)

The Latham Court did manage to make a mark on several important social issues
brought forward by World War II. These issues were important to lawyers, but they were
also the most widely salient issues of the day among the citizens generally (Hayne 2003).
Many of the Court’s decisions in these cases knocked back Labor government plans for

increasing the power of the Commonwealth to nationalize industry (Galligan 2002). In

the Bank Nationalisation Case,40 the Court struck down Labor Prime Minister Chifley’ 5
plan to nationalize all of the banks in Australia. In the Communist Party Case (1951) the
Court held the Communist Party Dissolution Act 1950 (Cth) unconstitutional, despite
Dixon’s best efforts to uphold the anti-communi st legislation (Dawson and Anderson
1996). In this decision to limit the use of defense power during peacetime—and to
prevent its use to curtail the expression of pro-communi st sentiment—Latham was the
lone dissenter (Botsman 2000). Through his tenure, Latham was seen as the epitome of a
legalistic judge. Some blame this legalism for his position as a dissenter in the critical
cases of his day (Cowen and Zines 2002, 429), but Dixon criticized Latham’s opinion in
the Communist Party Case as being an abrogation of the judicial duty to uphold the law

(Dawson and Anderson 1996).

 

40 Bank of New South Wales v. The Commonwealth (1948) 76 CLR 1.

57

Time to Regroup (1950-1964)

In the twilight years of the Latham Court, both Rich and Starke resigned. They
were replaced in 1950 by the Liberal Coalition41 government appointments of William
Fullagar and Frank Kitto. On Latham’s retirement in 1952, Owen Dixon was elevated to
the Chief Justiceship and Alan Taylor was appointed by the Liberal Coalition
government.

Dixon presided over a notably harmonious bench, and much credit is often given to
the new appointees (Simpson and Simpson 2002). As most of the dissenters had been
replaced with more “likeminded” Justices, a spirit of friendliness and cooperation grew
quickly, and helped to support the stature of the Court in the public eye (Bennett 1980).
The Dixon Court, however, did endure a great deal of change over the years. Before the
end of his tenure, Fullagar would be replaced by William Owen, Webb by Douglas
Menzies, and Williams by Victor Windeyer. All were appointees of the Liberal Coalition
government, though, furthering the homogeneity of thought and experience on the Bench
(Simpson and Simpson 2002).

Although Dixon’s leadership ability had been apparent nearly from the time of his
appointment, it flourished to an unprecedented degree during his tenure as Chief (Bennett
1980). He was adept at forging consensus on the Court, both through his facilitation of
professional cooperation and his negotiation of personal relationships (Smyth 2003b).
Dixon is often associated with the term ‘strict and complete legalism.’ Indeed, Dixon

noted that “[t]here is no other safe guide to judicial decisions in great conflicts than a

 

41 The Liberal Coalition at this time was made up of the Liberal Party (once the United Australia Party)
and the Country Party. The Country Party adopted the name “National Country Party” in 1975, and further
amended its name in 1982, becoming simply the “National Party.” The Liberal Coalition today is made up
of the Liberal and National Parties.

58

strict and complete legalism."42 Some interpret this remark to mean that Dixon favored
an approach to constitutional interpretation that sidelines the consideration of practical
policy consequences (Bhattacharya and Smyth 2001b). The legalist reputation of the
Dixon Court is enhanced by the perception that Douglas Menzies’s appointees also
favored this approach (Sawer 1975).

Others argue that this reputation for legalism may be due more to the famous
comment being taken out of context than a mechanistic view of the judicial task on
Dixon’s part (Hayne 2002). In this view, Dixon rejected the notion that judges had to
ignore information about the implications following from judicial decisions (Cowen and
Zines 2002). The Dixon Court, according to many scholars, remained flexible and even
creative in the face of difficult challenges (Zines 2002a). During this period, the
Commonwealth government was dominated by the Liberal Coalition. Unlike Labor
governments past, the Liberal Coalition government was generally happy to stay within

the express powers granted by the Constitution (Zines 2002a). For this reason, the Court

had little reason to strike down Commonwealth legislation.43 This light use of the
judicial review power contributed to the Dixon Court’s reputation as a legalistic bench.
This is not to say that the Dixon Court was inactive. Great innovations were made
during this time, especially regarding the interpretation of section 92 of the Constitution.
This section provides for freedom of interstate commerce, and had previously been
interpreted as a limitation only on protectionist mechanisms among the states. The Dixon

Court continued the evolution begun by the Bank Nationalisation Case, giving section 92

 

7
4" (1952) 35 CLR xi, xiv.

4 . . . . .
3 One notable exception rs the Boilermaker’s Case (1956), in which the Court expanded the operation of
separation of powers at the Commonwealth government level.

59

a distinctly laissez-faire flavor (Sawer 1957). The Court overturned previous rulings,
instead holding that the Constitution confers an individual right to free trade, subject only
to reasonable regulation (Zines 2002a). In the Second Uniform Tax Case (1957), the
Court affirmed the power of the Commonwealth government to tax to such a degree as to
make it politically impractical for the states to levy additional taxes (Bennett 1980). All
of this, coupled with the increase in Commonwealth defense powers during the Latham
Court, contributed to a further erosion of state powers.

The Modern High Court

With the appointment of Chief Justice Garfield Barwick in 1964, the High Court
began a period of transformation. The membership of the Court shifted dramatically
through Barwick’s tenure, and he instigated several important changes in jurisdiction and
procedure during this time (Mason 2002a). The Court physically took its place among
the other branches of Australian government, leading to “a new national vision of the
Court” (Kirby 1996). More than his predecessors, Barwick emphasized the High Court’s
role as the final arbiter of the important questions of the day (Mason 2003b).

The Barwick Court (1964—1981)

Chief Justice Barwick’s term was a time of transformation for the High Court in a
number of important ways. The Barwick Court saw many changes in jurisdiction and
procedure. Some of these changes were due largely to the work of Barwick himself, both
as Chief Justice and in his previous role of Commonwealth Attorney-General. During his
tenure as Chief, the Court gained greater independence from the Privy Council,
dramatically decreased its nondiscretionary docket, and obtained administrative control
over the affairs of the Australian judiciary (Mason 2002a). Barwick was even

instrumental in securing a permanent home for the High Court, which prior to 1980 was a

60

traveling tribunal. The new retirement age of 70, however, was the brainchild of the
Commonwealth Parliament, not Barwick.

The Court also saw huge membership shifts during this time and ended the period
much younger than it began (see Table 1). Secondly, many jurisdictional and procedural
changes dramatically altered the nature of the Court’s work. Finally, due in part to the
first two changes, the role of the Court and its Justices began its own transformation.
These changes began to build momentum as politicians, the public, and the Justices
themselves began to re-examine the position of the High Court in the government and
society.

Garfield Barwick was appointed Chief Justice by Liberal Coalition Prime Minister
Robert Menzies. Although Barwick had an impressive resume, he had a difficult time
fitting in on the High Court. As one scholar has noted,

Barwick did not slip easily on to the bench. Rather, he was an intruder in the club.
(Marr 1980, 213)

His reputation at the bar was extraordinary, so much so that he was knighted in
1953 (W interton 2002). Immediately prior to his appointment to the High Court, he was
serving as the Liberal Party Member of Parliament for Parramatta, Minister for External
Affairs,44 and Commonwealth Attorney-General.45 Many of his contemporaries believed

that Barwick had aspirations to assume the Prime Mini stership on the retirement of

 

Barwick had mixed reviews at best in terms of his role as Minister for External Affairs. In one instance,
he mistakenly claimed that America would be necessarily drawn into any potential Australian-Indonesian
conflict as a result of its treaty obligations. This caused a stir among Australians and members of the
Johnson administration alike (Galligan 1984).

45 In Australia, the Commonwealth, state, and territory Attorneys-General are members of their respective
parliaments, and are ministers of state. For this reason, the Commonwealth or State Attorney-General is
always an elected politician. Barwick was not the first former Commonwealth Attorney-General to serve
on the High Court, as Henry Higgins, Isaac Isaacs, and John Latham had also previously served in this
position. The last such appointee to date was the controversial Justice Lionel Murphy in 1975.

61

Robert Menzies, but the retirement of Owen Dixon forced Barwi ck to choose between the
two prestigious jobs (Winterton 2002).

Following Barwick’s appointment, there were several more changes in the Court’s
membership. In fact, not one of the Justices who began the Barwick Court would remain
until its end. Two Justices served their entire High Court careers within the space of
Barwick’s tenure. This was due largely to Barwick’ s unprecedented seventeen-year stint
as Chief Justice. For this reason, legal scholars tend to split the Barwick Court into
distinct periods (Mason 2002a). Near the beginning of the Barwick Court, the Liberal
Coalition government made several appointments to the Bench. Cyril Walsh was
appointed after Taylor’s death in 1969. After graduating at the top of his class at the
University of Sydney, Walsh had acquired a substantial law practice, sharing chambers
with Barwick. Unlike many Liberal party appointees, Walsh had never been involved in
legal academia, and he had never taken silk (F ricke 1986). Shy and modest, Walsh was
appointed to the New South Wales Court of Appeal at the recommendation of his
admiring colleagues at the Bar (McLaughlin 2002). He held this position until his
elevation to the High Court.

The following year, Kitto resigned and was replaced by Harry Gibbs. The son of a
successful lawyer, Gibbs excelled in his academic career. Shortly after beginning his law
practice, Gibbs enlisted in the military during World War 11. Subsequently, he helped to
plan a post-war government for Papua and New Guinea. After returning to practice,
where he quickly developed great prestige, he was appointed to the Queensland Supreme

Court (Jackson and Priest 2002). He was then promoted to the Chief Justiceship, and

62

later to the Supreme Court of the Australian Capital Territory and the Federal Court of
Bankruptcy. It was from this position that he was elevated to the High Court.

In 1972, Windeyer resigned and Owen died. They were replaced by Ninian
Stephen and Anthony Mason. Stephen, like Gibbs, enlisted in the military at the
beginning of his law career. He returned to join the Melbourne Bar, and from there was
appointed to the Supreme Court of Victoria. Two years later, he was elevated to the High
Court. Mason served in the Royal Australian Air Force before completing law school.
After a distinguished career in legal practice, he was appointed to the New South Wales
Court of Appeal. He was elevated to the High Court shortly thereafter.

The Labor Party came into power in 1972. Labor’s first new appointment to the
Barwick Court was Sydney Jacobs in 1974, in the seat left vacant after Walsh’s death.
Jacobs enlisted during World War II, and was appointed as a military intelligence officer.
Upon his return, he completed law school and developed a legal practice. He stood for a
New South Wales Parliament vacancy, but his campaign failed by a single vote
(Blackshield and Mackrell 2002). He was then appointed to the Supreme Court of New
South Wales, where he worked with Mason and Walsh. He became that court’s president
before being elevated to the High Court. Soon thereafter, Menzies died at a New South
Wales Bar Association function and the Labor government appointed the Commonwealth
Attorney-General Lionel Murphy to replace him. Murphy, who first earned a degree in
organic chemistry, sat for the New South Wales Bar exam before actually graduating
from law school. He passed, and proceeded to make a name for himself in labor law. He

was elected to the Commonwealth Senate in 1961, and subsequently was selected to be

63

 

$582 toned SE52 2:: DZ; 5::8 do 5250 2: $3 $562 3&on or

:80 x2255 Eiwco 2: mo mconEoE 3:365 *

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

07: EB: impasse; ocoz 82 Hansen 32 mm 82 tonnes
07: a ccoméaéezé 9;: 0-0 oooz Baas; <3 om 8% 82:5
024 N. Emboéé oooz a; E asmeoi >62 8 8% in»,
oz; N. szozoé oooz a E 23885 52 cm 32 12%.;
024 m. .£02.3\mm0m5>o 0:02 .mbn N ESmBOr—m U~> xv mNQ— cosmoam
oz; Eon: ocozefifi 252 a; a “cameos >32 5 on: 225
mi ,5 :53 amassed EB o-< 82 2658 32 S S2 39::
oz; :23: ens—2.2253 252 2oz Ensues; 65 cm 82 22.252
5< m2 ES Emsofisoo 9&6 o-< 252 29:6 >32 3 N3: 225%:
024 a Eméoaéa 29 on a; m 22:6 262 2 82 cos:
024 a images 252 onoz Essex. 52 3 82 *ozg
S< Eon: amass E 252 a; E asses amz om. :2 £82
075 a 20229:; ocoz a; x _ 59865 30 m :2 Bee
024 N. .eozjoséo 252 252 Benson 0; oo 22 canon

m2 EB: .3228; 2Q o-< 8oz nausea amz $23.3

523;

 

.23.. _. EC

:7;
__..5:::.

:29 Tax

.5391

:30 its: 2: no 3.2:: 8:2... - _ 2...;

Es:

 

64

leader of the opposition and later Attorney-General. Against Barwick’s protests, Murphy
was appointed to the High Court.

After sustaining a fractured hip at the age of 84, McTieman was confined to a
wheelchair.46 Because the High Court had no accessible entrances, McTieman ended
hisretumed to Australia and built a reputation as a capable practitioner. He declined an
offer to replace Taylor on the Court, but accepted another invitation a few years later.
After another two years of the Barwick Court, Jacobs retired and was replaced by Ronald
Wilson. Wilson left school after his tenth year and started work as a messenger in his
local Western Australia courthouse. He enlisted in the Royal Australian Air Force and
flew Spitfires in England. After the war, when he could not secure a position as a flying
clergyman, he abandoned hope at joining the ministry and chose to study law instead.
After graduating, he worked in Western Australia’s state prosecutor’s office. He studied
briefly in the United States on a Fullbright Scholarship. He returned to the Crown
service, and was appointed Solicitor-General of Western Australia. He served in this
position for ten years before being the first Western Australian ever appointed to the High
Court.

These changes in membership resulted in a Bench that resembled very little of the
original Barwick Court. These changes affected a shift in approach that was at first
subtle, but set the groundwork for the more marked departures that would occur in later
Courts (Botsman 2000). The Barwick Court, though, is probably best known for its

extraordinarily literal interpretation of tax law under section 90 of the Constitution. Most

 

46 Apparently, McTieman sustained his injury while “trying to squash a bug on the floor of his bedroom”
(Callinan 1999). Chief Justice Barwick denied McTiernan’s request for a ramp to be built, leaving him
physically unable to continue his record-breaking tenure on the High Court.

65

notably, the Barwick Court majority gutted section 260 of the Income Tax Assessment Act

1936 (Cth) (Mason 2002a). This section was meant to be an anti-avoidance provision. In
cases like F C T v. Commonwealth Aluminium47 and Slutzkin v. F C 7148 the Barwick Court
set the groundwork for the creation of numerous tax avoidance schemes by reading
section 260 very narrowly (Kobetsky and Krever 2002). This process added complexity
to the tax code, and purposefirlly failed to consider the intent of the legislatures. Barwick
himself argued against any civic or moral duty of citizens to comply with tax laws, citing
only a legal obligation (W interton 2002). Criticism of this interpretation came from the
Court’s onlookers, but also from within the Court. Justice Murphy was the most
outspoken of the dissenters. Later, he noted that:

The [Barwick] Court used to read Acts of Parliament absolutely literally - the

words were all important - the spirit was often ignored. This made it easy for any

competent lawyer or accountant to devise schemes to turn profits into tax losses.

Because of this many of the rich ceased to pay tax and the burden fell on the
workers and the scrupulous. (Murphy 1983, 1)

Chief Justice Barwick’s large personality shaped a good deal of the public’s
opinion of the Court (Mason 2002a). He twice gave informal constitutional advice to the
Governor-General, both instances seen by the public as morally dubious at best. The
second and most famous of these is the advice he gave to Governor-General John Kerr
surrounding the 1975 dismissal of Prime Minister Gough Whitlam. When it appeared
that Whitlam would not be able to get the budget passed in the Senate, the Govemor-
General solicited Barwick’s extra-judicial advice on the situation. Barwick responded in

writing that he agreed with Kerr’s unprecedented plan to sack the Prime Minister and

 

47 (1980) 143 CLR 646

48(1977) 140 CLR 314

66

invite the opposition leader to form an interim government (Blackshield 2002c). Barwick
continued to defend his role in the dismissal, but many of the Court’s observers maintain
that the advice was controversial at best (W interton 2002).

Additionally, the modicum of civility maintained prior to Barwick’s tenure eroded
significantly over this period. Especially tense was the relationship between Barwick and
Murphy. They differed in a firndamental way in their approach to law and their
conception of the nature of democracy (Hocking 1997). Barwick had opposed Murphy’s
appointment, and was unabashed in his criticism after Murphy joined the bench. One
scholar has observed that the Court was “. . .dominated by the increasingly erratic
behaviour of its Chief Justice, Garfield Barwick” (Hocking 1997, 227).

The Gibbs Court (1981-1987)

As compared with the Barwick Court, there were fewer changes in personnel
during the Gibbs Court (see Table 2). Upon Barwick’s 1981 retirement, Liberal
Coalition Prime Minister Malcolm Frasier appointed Harry Gibbs to the position of Chief
Justice. Gerard Brennan was selected to fill the resulting vacancy. This pair of
appointments was meant to help depoliticize the Court after the controversy of the
Barwick Court (Galligan 1987). The son of a Justice of the Supreme Court of
Queensland, Brennan passed his high school exams at age 16. After university, he
worked as an associate to his father. He had a very successful practice in Queensland
before his appointment as the first President of the Administrative Appeals Tribunal. He
was also appointed to the Federal Court, but stayed only a short time before being
elevated to the High Court.

Soon thereafter, Justice Stephens resigned in order to take up the Govemor-

Generalship. Justice Keith Aickin died as a result of injuries sustained in a car accident

67

(Holloway 2001). William Deane succeeded Stephens, and Daryl Dawson joined the
Bench shortly thereafter, both appointed by conservative Prime Minister Malcolm
Frasier. Daryl Dawson was appointed to the High Court from his position as Solicitor
General for the state of Victoria. In this role, he had argued 60 cases in the High Court,
some of them critical in the development of Australian constitutional law.

William Deane, who was knighted by the Queen of England in 1988, has gone on
to become a very popular figure in Australian public life. Born to a strict Catholic
family, Deane excelled in a string of prestigious private schools and colleges (Holloway
2001). He went on to become a university lecturer, a prominent lawyer, and finally a
judge before his appointment to the High Court Bench. His only participation in party
politics was his short-lived membership in the Democratic Labor Party—a ri ght-win g
anti-communist party founded in the late 19505 (Stephens 2002). He, like his
predecessor, would later resign in order to accept an appointment as Governor-General.

These High Court appointments, along with the growing tension between Gibbs
and Murphy over Murphy’s legal troubles, changed the dynamics of the Bench. In
November of 1983, information was leaked to the press about illegally recorded
telephone conversations between a Sydney solicitor and an unnamed judge. It was not
until three months later that the judge on the tapes was identified as now High Court
Justice Lionel Murphy. At this time, a series of allegations—most politically
motivated—were leveled against Murphy (Blackshield 2002a). Among these charges
was the allegation that he gave legal advice to and attempted to influence the judge in the

trial of an acquaintance.

68

 

A _ com oct— wcm :oamEmUV 853m HEEEEooAEm 2; do :32 a 3 meme 2: E acorn bus; csoan-o_§ooEoQ of 02:3 2.on *
.350 76%? team 582 Enzgmzlx 2: :o 83 8:28ng 366:3,5955 or

 

 

 

 

 

 

 

 

 

 

 

07: a ==oo.<3_:\o=€._ 23v 0-0 0.52 asses <3 3 82 823
3< N. .<3_:\o=o56 252 a; 2 32:8 <3 on or: 32.83
024 Nt .n_®2.3\mnoeo>o 0:02 .mbm N ESmBOgL U—> wv Mm©_ cosmoum
A3< m: :53 202235 29 0-< 2:2 22:8 3mz mm S2 3%:
024 a Engage; 29 0-m a; m 2658 3% 2 82 Ea:
07: ._. Embozoaoo 252 .23 m 2658 3mz s :2 2.80
07: a o_§.£o2.0\o=o§ 8oz 5; 2 nausea 0; we 22 8min
07: a. mo§_§oe\o=o§0 oooz a:2 m 2056 30 mm 32 52:55
07: a. £22225 oooz ocoz asses 0; 8 22 €32
07: 5 20.0225 252 a; : assess So 9 :2 3&0

:20 :35 o... .5 3.2:: 822:. - N osfi

 

69

Two separate Senate committees looked into charges against Murphy, the second
of which resulting in his prosecution. At the first trial, in 1985, he was convicted of
attempting to pervert the course of Justice. On appeal, the conviction was quashed and a
new trial ordered. At the second trial in April of 1986, Murphy was acquitted of all
wrongdoing. In light of the fact that Murphy had recently been diagnosed with terminal
cancer, impeachment hearings against him had been halted (McGarvie 1999).
Nevertheless, Chief Justice Gibbs wrote to Murphy and suggested that he refrain from
returning to the Bench. He indicated that he thought Murphy’s return undesirable, and
that he thus had no intention of listing Murphy to sit (Campbell and Lee 2001).
Notwithstanding the wishes of the Chief Justice, Murphy returned to the Bench until his
death later that year (Blackshield 2002a).

It was during this turbulent time that some of the most provocative cases in the
Court’s history made their way to the Justices (Twomey 2002). The often-divided Court
began to expand the powers of the Commonwealth government at the expense of the

states. The proposal to dam the Franklin River may not sound politically controversial,

but it lead to the infamous Tasmanian Dam Case (1983).49 In this case, the majority of
the Court dealt a crushing blow to state sovereignty by preventing Tasmania from
building a dam on land entered in a ‘World Heritage List’ at the federal govemment’s
request (Twomey 2002). The opinions in the case, for both the majority and the minority,

were “replete with policy considerations and value judgments” (Zines 2002c, 14).

 

49 Commonwealth v. Tasmania, 158 CLR 1

7O

In Koowarta ’5 Case (1982),50 the Court expanded the Commonwealth
govemment’s external affairs power. The Commonwealth govemment’s conciliation and

arbitration powers were also expanded during this time, especially in the C YSS Case

(1983).51 The Court’s role in the centralization of government power did not go
unnoticed, even by a previously disinterested public (Walker 2002b; Zines 2000).

During the Gibbs Court, the High Court’s place at the pinnacle of the Australian
legal system was finally secured. The Court severed its remaining ties to the Privy
Council. In Caltex Oil v. X1 Petroleum (1984),52 the Court asserted that its decisions
should prevail over Privy Council decisions on the same subject matter. Additionally, the
Australia Acts 1986 (Cth) were enacted at the end of the Gibbs Court, abolishing appeals
to the Privy Council from the state courts. In 1984, by an amendment to the Judiciary
Act of 1903, appeals as of right were replaced with a system requiring special leave to
appeal. This gave the High Court an almost completely discretionary agenda. Perhaps in
part because of this newfound docket control, the Court began to take on a more flexible
approach to jurisprudence. In the process, Australian scholars point to the Gibbs court as
the beginning of the end of strict legalism in the High Court, and a move toward a “more
expansive view of Commonwealth legislative power” (Twomey 2002, 305).

The Mason Court (1987-1995)

Ifthe Gibbs Court began to move away from strict legalism, then the Mason Court

might be said to have abandoned it altogether (Doyle 1996). While the eight years of the

 

50 Koowarta v. Bjelke-Petersen, 153 CLR 168
51 R v. Coldham; Ex Parte Australian Social Welfare Union, 159 CLR 297

52
155 CLR 72

71

 

Mason Court enjoyed a relatively stable membership (see Table 3), very little else was
without tumult. Lionel Murphy lost his battle with cancer in October of 1986, leaving a
vacancy on the High Court. Shortly thereafter, Gibbs reached the newly instituted
retirement age of 70. This left a good deal of space for the new Labor Party Prime
Minister, Bob Hawke, to alter the composition of the Court. Anthony Mason, by this
time knighted by the Queen of England, was elevated to the position of Chief Justice to
replace Gibbs. Mary Gaudron and John Toohey rounded out the first wave of Mason
Court appointments.

Mary Gaudron was the first woman ever appointed to the High Court. Born in a
country town in New South Wales, Gaudron’s interest in the Australian Constitution
developed early. She witnessed protests against a proposed amendment meant to ban the
Australian Communist Party (Kalowski 2002). She attended law school on scholarship
after getting married and becoming a mother. By most reports, she faced a good deal of
resistance in a legal profession that was quite chauvinistic at the time (Blackshield et al.
2002). Despite the obstacles, she had developed a blossoming legal practice in New
South Wales. She was appointed a Deputy President of the Arbitration Commission by
age 31. She left the commission to lecture at the University of New South Wales, and

was quickly appointed Solicitor-General of New South Wales in 1981. Her appointment

to the High Court as Murphy’s replacement was greeted with near universal prai se.53
Like Mary Gaudron, John Toohey grew up in the country. Only the second
appointee from Western Australia, he left his profitable law practice in Perth to establish

the Aboriginal Legal Office in Port Hedland (Edelman and Gray 2002). At the same

 

53 Among the few opponents of this appointment was a local gossip sheet accusing her of having “an
emotional disposition inappropriate in a holder of judicial office” (Kalowski 2002, 294).

72

time, he served on the benches of both the Federal Court and the Supreme Court of the
Northern Territory. For the remainder of his time on the bench—and beyond—he has
been best known for his passionate defense of Aboriginal Rights. Moreover, he has
proven to be willing to bend the traditional judicial role in order to achieve his goals. In
an interview shortly before his retirement, he noted that:

the decision not to change or not to develop the law is just as ‘activist’ as a decision

to change the law and can have consequences just as dramatic. (Edelman and Gray
2002,675)

The only other personnel change during the Mason Court was the retirement of
Wilson in 1989. Michael McHugh was appointed by Prime Minister Bob Hawke to
replace Wilson. McHugh’s career got off to a shaky start when he dropped out of school
at age fifteen to pursue various blue-collar jobs. At age 22, he began taking night classes,
and soon finished his law degree (Guilfoyle 2002). Taking time to build a successful
career in Newcastle, he was eventually appointed Judge of Appeal in the New South
Wales Court of Appeal.

The Mason Court reinvented the role of the High Court, all the while shaping the
interpretation of the Australian Constitution for a new era (Lavarch 1996). The
establishment of the video link system in 1988 allowed the Court to hear special leave
petitions from the comfort of their own Canberra building. This reduced the need for the
Court to travel, while still maintaining accessibility (Jackson 1996). With the passage of
the Australia Acts of 1986, the last avenues for appeal to the Privy Council were severed.
The High Court, finally the only final court of appeal in all Australian Cases, turned its
back on the Dixonian doctrine of legalism in favor of what has been described as a more
practical and realistic approach (Galligan 1995). Leading the charge was Mason himself.

He was not afraid to speak in public about judges and judging, and in the process

73

 

.88 3:: 035230 on 00: 2:03 02:95. .380 ,2me 2: 9 0850090 22000 Eu 2: £3 08030 H
.EoEmEam 00 $852 been 593 a 0883 082 0:3 00.0000 03052 02:08 amaze—z *

 

 

 

 

 

 

 

07: 2.. 22.<3.:\23§ 230 0-0 202 Ease; <3 2 82 823
mi 2.. .<3.2£220 202 a; 2 £220 <3 2 3.2 280
S< .. 2022230 202 a; m £230 3% mm 22 2:082
312 2.. 223.5 @2230 A390 0-m a; a £25 32 3. 0.2 H 2230
07: 2022230 202 a; w £230 3% a :2 280
024 2.. 83.22.2232 2oz .2» S 22865 02> a. me 825
07: 2222092230 252 a; m £230 50 9. £2 2255

 

 

:2: 5;; z.

 

55?.

202222

300 01m .23 m
,7;
5;. ., 55 2:32;.

£230

2:03:01

262

25:11.02

50

mag

Es:

0822

0.0.1: _.

taco fleas—Z 2: no 03:55:. 00:2; - n 933,—.

 

74

introduced the idea that policy has an important role to play in the Court’s decision-

making process (Brennan 1996a).54

The Chief Justice’s outspoken personality came as a bit of a surprise to onlookers,
especially given Anthony Mason’s cautious beginnings on the Bench (Doyle 1996). As
apuisne Justice, Mason and Barwick issued many legalist-minded joint opinions (Walker

2002c). At the beginning of his tenure as Chief, the Court decided one of the most

important cases in its history: Cole v. Whitfield.55 In this case, the Court
abandoned the various legalistic tests which had previously been developed for the
application of section 92 of the Constitution. Instead it adopted a test based on the

underlying purpose of section 92, which was, of course, to provide a free trade area
throughout Australia. (Lavarch 1996, 15)

In its unanimous decision, the Court considered the political and economic history dating
back to Australia’s pre-federation experience (Selway 2002). They also considered
modern context and values, arriving at a simple and lucid solution to the longstanding
quagmire of free-trade jurisprudence (Coper 2002a).

Then, in 1992, the Court addressed the controversial native title issue in Mabo.56
In this case, the Court incorporated the native title doctrine into Australian law. In the
process, the majority overturned the longstanding policy that Australia was terra nullius
on the arrival of the European settlers. Instead, they gave legitimacy to the society of

indigenous people that existed here before colonization. This decision overturned 200

 

54 The one sitting Justice most clearly opposed to this new approach was Daryl Dawson, who expressed his
distaste for the new radical approach on several occasions (Dillon and Doyle 2002).

55 (1988) 165 CLR 360

56 Mabo v. Queensland [Na 2] (1992) 171 CLR 1

75

years of ostensibly settled law and unleashed a good deal of social and political unrest,
albeit with a delay of a couple of years (Solomon 1999).

The Mason Court’s new approach to constitutional interpretation was also used to
secure implied rights and freedoms (Dillon and Doyle 2002). As Leslie Zines observed:

The decisions of the Court in respect to implied rights and freedoms, together with

the minority views of Deane and Toohey JJ, led to some speculation whether

Australia would finish up with something like a Bill of Rights without the necessity

of actually enacting one. (Zines 2000, 226)

In the Free Speech Cases,57 the majority held that a fundamental freedom of speech
could be implied from the principle of representative government laid out in the
Constitution. In these cases, the Court overturned legislation that was intended to limit
political speech. The resulting doctrine falls far short of the American First Amendment
standard, and is instead meant to encompass mainly political speech. Because of this,
politicians were the most vocal critics of these new implied rights (Solomon 1999).

The Court also increasingly referenced international law in its opinions. As the
Australian Constitution has no written statement of individual rights, Australia’s
commitment to various international treaties provided the Mason Court with another
avenue through which to secure the protection of such rights (Patapan 2000a). While

international treaties are not automatically incorporated into Australian domestic law, the

ratification of the treaty can be seen as an obligation under international law (Joseph and

Castan 2001). In T eoh,58 the Mason Court prevented the deportation of a convicted drug

dealer. He was the father of several children, and his deportation would have left them

 

57 Nationwide News v. Wills (1992) 177 CLR 1;.4ustralian Capital Television v. Commonwealth (1992)
177 CLR 106

58 Minister for Immigration and Ethnic Affairs v. Tech (1995) 183 CLR 273

76

without any parents. Although the UN Convention on the Rights of the Child had not yet
been specifically incorporated into law, the Court held that ratification had generated a
legitimate expectation that the treaty would be honored (Solomon 1999). This new

legitimate expectation approach allowed the High Court to police the Commonwealth

government, effectively bypassing the Parliament in the incorporation of these treaties.59
Through all of the doctrinal tumult, the Mason Court enjoyed a collegial
workplace. Cooperation among judges on opinions was common and, as Brennan later
recalled:
Suggestions for changes in a draft judgment were freely given or received with full
recognition of the independence and intellectual integrity of the other Justices. It is

no wonder that the members of the Court remained on the most agreeable terms,
though we often divided on issues of the greatest importance. (Brennan 1996a, 13)

Although the Court faced some of the most difficult issues in Australian law, it
openly acknowledged its perennial law-making role. This allowed the Court to avoid
haphazardly squeezing novel situations into tired and outdated precedent (Doyle 1996).
This fell in line nicely with Mason’s developing belief that the Court’s primary firnction
is to keep Australian law up to date with the changing conditions of the country (Galligan
1995)

The Brennan Court (1995-1998)

On Mason’s retirement in 1995, Gerard Brennan was elevated to the Chief Justice
position by Labor Prime Minister Paul Keating. During the course of Brennan’s short
tenure as Chief, the Court went through a transitional period, both in terms of

membership (see Table 4) and of administrative difficulties. This is not to say that the

 

59 See also Dietrich v. The Queen (1992)l77 CLR 292.

77

Brennan Court did not accomplish great things, but simply that the discontinuity of
membership made the process more difficult (Jackson 2002a).

The first addition to the Brennan Court was William Gummow. He was appointed
in April of 1995 by Prime Minister Keating to fill the vacancy left by Brennan’s
elevation. Gummow attended private school until university, when he attended the
University of Sydney and graduated with honors. Gummow went on to teach in the
Faculty of Law at the University of Sydney until his appointment to the Federal Court in
1986. While active in academic circles, he also maintained a diverse law practice before
becoming a judge. His work included serving as junior counsel to then Commonwealth
Solicitor-Genera] Maurice Byers (Lehane 2002).

The following year, Justice Deane announced that he would retire early in order to

serve as Governor-General. His replacement, Justice Michael Kirby, would soon be the

most talked-about judge since Justice Murphy.60 Like most High Court appointees,
Kirby was an exceptional student. After graduating with numerous degrees from the
University of Sydney, he maintained an active academic life. Additionally, he was
chairman of the Australian Law Reform Commission. Following this, he served as the
special representative of the Secretary-General of the United Nations for Human Rights
in Cambodia. The assignment allowed Kirby to pursue one of his most passionate
causes: the fight against discrimination and human rights abuses. His experience as an
openly homosexual professional “taught him the pain and distress that discrimination

causes” (Sheller 2002, 394).

 

60 Justice Kirby was awarded the “Judicial Oscar” by journalist John Henningham for Kirby’s excessive
mentions in newspaper articles in 2003. According to Henningham, Kirby was mentioned in 303 stories
about the High Court (Albrechtsen 2003).

78

Justices Dawson and Toohey also retired during the three years of the Brennan
Court. Their replacements were appointed by Liberal Coalition Prime Minister John
Howard. Justice Dawson was replaced by Kenneth Hayne. Hayne was a Rhodes
Scholar, but defied the predictions of his peers by eschewing a career in academia. After
a wildly successful commercial law practice spanning two decades, Hayne was appointed
to the Victorian Supreme Court and Court of Appeal. From here, he was elevated to the
High Court.

Justice Toohey was replaced by Ian Callinan. Callinan took an unusual path to the
High Court bench. After completing the equivalent of a high school education, he
worked as an Immigration Department clerk. He subsequently studied law part-time at
the University of Queensland, and later embarked on a very successfirl nation-wide

practice in civil litigation. He was also called upon in 1985 to join the prosecutorial team

in the case against Justice Lionel Murphy, which he did.61 His interest in the law did not

keep him from a second passion—fiction. He authored two novels and a series of plays
which drew on his experience with jurisprudence and the related issues of ideology and
morality (Hasluck 1998). In 1998, he became the first High Court Justice in two decades
to be appointed directly from private practice.

The replacement of three Justices in as many years caused a great deal of technical
difficulty on the Court. High Court protocol requires that the Court hand down opinions
prior to the departure of any participating Justice. This made scheduling hearings very
difficult, and resulted in a number of important constitutional issues being decided by a

six-member Bench. Additionally, the members had little opportunity to familiarize

 

61 Callinan secured a conviction on the first instance, but in the retrial mandated by the appellate court,
Murphy was acquitted.

79

themselves with the style and idiosyncrasies of their colleagues, and thus endured times
of noticeable tension (Jackson 2002a).

Despite these diffi culties, the Brennan Court managed to accomplish a lot more
than many onlookers had expected (Haultain 1997). As a Chief Justice, Brennan was
“generally non-interventionist and invariably polite” (Kirby 2003, 6). The Court under
Brennan’s leadership worked mainly to consolidate the law. In this, it was quite
successful (Baker and Gagel er 2002). One of the most notable areas clarified by the
Brennan Court is the freedom of political speech. In Lange v. ABC and Levy v.
Victoria,62 the Court unanimously confirmed an implied constitutional protection of
political speech related to both defamation and non-verbal political protest. This move
helped to clarify the deep divisions among the members of the Mason Court regarding the
effect of implied free speech on the defamation of public figures and non-verbal political
speech (Blackshield 2001).

The Lange decision, despite its acceptance of an implied freedom of
communication, is sometimes seen as the beginning of the Court’s retum to a more
conservative, doctrinal approach to constitutional jurisprudence. This is due not to the
substance of the opinion, but the more traditional, legalistic arguments surrounding the
substantive conclusions (Zines 2000; Stone 2000). Observers of the Court anticipated
that this shift in tone and language was an attempt by the Brennan Court to distance itself
from the perceived activism of the Mason Court, while sustaining the substance (Stone

2000)

 

62 Lange v. Australian Broadcast Corporation (1997) 189 CLR 520', Levy v. Victoria (1997) 189 CLR 579

80

 

woom 00 mm 2300 :wi 05 :0 8:2; 3:088:20: 22:03 22:0 0:0 5:0 05 E Eth sh

.288 £29 22:2 .23 22 9 e22 22 38:22 352 *

 

 

 

 

 

 

 

 

 

 

 

2.2 2.. .<3.02£22e0 2oz a; 2: £220 <3 2 082 2283
2:2 20022222220 252 22,. m £22220 3mz mm 22 222%:
22.2 ._. Embezeee 2oz .22 2 2382922 3mz 2 3.2 Z2252
024 2.. 28202220202295 252 a; m 28855 02> S 9.9 232
22.2 2.. 03229.5 252 a; e asses 3mz mm as. 32:20
5< 2 2.235 £26.20 236 02 .22 e £230 3% 3. 0.2 8230
02.2 202022230 252 a; m £230 3mz : :2 280
02.2 2.. 2.332020222222222 252 a; S 23855 02> we .82: 22838
07: 2. .Eobesée 252 2oz seaweed So 8 :2 2220
07: 2.. 22222232002220 252 .222 m £220 30 mm 2282 2220

9:50 53:95 2: .5 335.532 8:2... . v 035.

 

81

The Brennan Court also attempted to refine the application of the doctrine of
intergovernmental immunities. The Cigamatic63 case held that the Commonwealth
government maintained a wide exemption from state-levied taxes. Since the Cigamatic
Case, the Court had not yet managed to unify behind a specific area of governmental
immunity (Dominello 2001). It was not until Henderson’s Case“ in 1997 that the Court
began to resolve the issue. Here, the majority reduced the extent to which the
Commonwealth would be free from taxes levied on its agencies by the various state
governments. In Ha v. NS W,65 though, the Brennan Court found that state-instituted
tobacco taxes were to be classified as excise duties. Thus, the power to tax these items
remained solely with the Commonwealth government. These decisions taken together
worked to keep the states “out of the traditional areas of taxation” (Haultain 1997, 2).

Another important topic clarified by the Brennan Court is the area of native title.

In Mabo,66 the Mason Court had determined that pastoral leases did not necessarily
extinguish native title. If this were true, then up to 42% of Australia might be
redistributed through native title claims (Patapan 2000b). This very question came to the
Brennan Court in the Wik67 case. The majority held that, to the extent that pastoral leases

were in conflict with native title rights, the leases would prevail. In this way, the

 

63 Commonwealth v. C igamatic Pty Ltd (1962) 108 CLR 372

64 Residential Tenancies Tribunal (NS W), Re; Ex parte Defence Housing Authority (1997) 190 CLR 410
65 Ha v. New South Wales (1997) 189 CLR 465

66 See 11. 56.

67 Wik Peoples v. Queensland (1996) 187 CLR 1

82

Brennan Court acknowledged native title rights, but reigned them in (Baker and Gage] er
2002)

The Gleeson Court (1 998-present)

Chief Justice Brennan retired in 1998. His replacement, Murray Gleeson, was
elevated directly to the Chiefs position from the Supreme Court of New South Wales.
While some early assessments assumed that the Gleeson Court would continue the
consolidation of the jurisprudential changes of the Mason Court (Patapan 2000b), the
substantial imprint of John Howard’s more conservative appointees leads some to
question this assumption (Came 2002). Although Australian legal scholars had once
declared the demise of legalism, the Gleeson Court has at least resurrected the debate
(McHugh 2004) much to the chagrin of some observers (Canigan 2003).

Murray Gleeson was raised in a Catholic family in New South Wales. Not only did
he excel in his academic studies, but as a youth he also played cricket and raised piranhas
(Kirby 1998b). After graduating with honors from Sydney University, he began a very
successful law practice. He very quickly became a highly sought-after barrister. In 1980,
he won the last appeal by leave from Australia to the Privy Counsel in England (Walker
2002a).68 In 1988, he was appointed Chief of the Supreme Court of New South Wales.
This appointment was popular among members of the law profession (Walker 2002a).
Ten years later, he was elevated to the High Court Bench by Liberal Coalition Prime
Minister John Howard. He was certainly not a new face in the High Court, as he had

appeared as counsel in 70 reported cases (Blackshield et al. 2001).

 

68 See Port Jackson Stevedoring v. Salmond & Spraggon (1980) 144 CLR 300.

83

 

 

 

 

 

 

 

 

 

244< 33.322230 252 a; m £230 32 mm 39 233222
244< 3302233 252 .222 2 3332822 32 em 22 254
024 2.. 35320232033222 252 a; m 3332822 04> mm 22 23:
L114. K... fizmfi<3w>tm 0:02 .850 acmwmoaoum amz NW NVO~ 2505530
e4< 2.. 3.32235 @2230 2320 0-m 3: 20 £230 32 a. 9.2 83230
024 3 35.033222 252 252 22332822 040 8 22.: 23230
024 2 33.02230 e2 .3; 2 £230 32 em 2282 25320

«.500 .8305 2: :0 38.3332 332:. - m 03:.

 

84

The Gleeson Court is mostly seen as relatively conservative, quite a change from
the more adventurous Benches of the recent past. Generally, the main point of departure
cited is a renewed emphasis on the text and structure in statutory and constitutional
interpretation (Zines 2002b; Patapan 2002). Indeed, Chief Justice Gleeson has been
anything but shy in advocating a return to Dixonian legalism (Gleeson 1998, 2001b,
2002b), and Sir Anthony Mason has taken note of this shift (Mason 2000). This focus is
not shared by all members of the Court, however, and “there is an interpretation debate
occurring on the High Court and how it unfolds may have a significant impact on the
trajectory of our future constitutional jurisprudence” (Meagher 2002, 14).

The High Court Centenary and Beyond

The analysis of the High Court presented here extends from Chief Justice
Barwick’s 1964 ascension to the High Court through Justice Gaudron’s retirement in
2003. While the three subsequent appointees are not included in the analysis, a few
words about them can help to paint a picture of the future of the High Court under Chief
Justice Gleeson and beyond. Indeed, these changes to the Court have likely consolidated
the Gleeson Court’s return to legalism (Pelly 2008). While it is too soon to know for
sure, some observers argue that all three new appointees are likely to follow Gleeson’s
favored path of Dixonian legalism (Ackland 2007).

Dyson Heydon was appointed in 2003 after Justice Gaudron’s early retirement. An
avid crusader for a return to a more traditional style of judicial decision making, he gave
a lecture entitled “Judicial Activism and the Death of the Rule of Law” mere months
before his appointment to the High Court (Heydon 2003). In this piece, he lamented two

tendencies working to pull the judiciary fidrther from Dixonian ideals.

85

The first is the desire to litter judicial decisions with the judge’s opinions on every
subject which may have arisen, however marginal. The second is the desire to state
the applicable law in a manner entirely unconstrained by the way in which it has
been stated before because of a perception that it ought to be different. (Heydon
2003, 12)

As for Justice Susan Crennan, the picture is much less clear. Crennan was
appointed to replace Justice McHugh upon his mandatory retirement in 2005. She was
one of “McHugh’s Angels,” as the ten women named by McHugh as suitable
replacements have been dubbed (Pelly 2005). There has been some early speculation that
she may be more of an activist than John Howard’s Liberal Coalition bargained for. In
her swearing-in speech she noted that legal developments in the High Court have
occurred against a background of significant social change, and major shifts in public and
private values.

But the images to which I have referred, of a judiciary which transfuses fresh blood

into our polity and of the law as a living instrument conjure up the human qualities
needed for the impartial dispensation of Justice according to law. (Donald 2005)

The most recent appointee to the High Court was a favorite among McHugh’s
Angels to replace McHugh, and had also reportedly missed the nod to replace Justice
Gaudron by a small number of cabinet votes (Pelly 2005). After being passed over twice,
Susan Kiefel was appointed in 2007 to replace Justice Callinan. Justice Kiefel’s
appointment has been regarded as largely uncontroversial and relatively apolitical
(Bradford 2007). She was appointed to the High Court by Liberal Attorney-General
Philip Ruddock, but was previously appointed to the Federal Court bench by Labor
Attorney-General Michael Lavarch. One reporter explains that, while she is “regarded as
conservative, Kiefel is not expected to be captive to either side of politics” (Marriner
2007, 1). She has been described by a conservative commentator as having “earned a

reputation as a fine black-letter lawyer” (Albrechtsen 2007, 14).

86

In 2004, the High Court celebrated its centenary amidst the thunderous roar of two

Royal Air Force F-l 1 1 jets setting excess fuel ablaze in the Canberra night sky. The

reaction to this exercise was mixed.69 Justice Kirby later pondered the symbolism of the
two streaks of light stretching across the horizon in a speech at the University of Notre
Dame. He indicated that, in many respects, this display represented a distinct change
from the past to the future (Kirby 2004).

It remains to be seen whether the centenary becomes an historical turning point. In
the span of its first hundred years, though, the High Court has changed dramatically.
From a roving band of three constitutional framers traveling the Australian continent, the
High Court has become a highly respected and very important part of the Australian
political system.

That said, the increasing willingness for members of the Court to embrace a less
legalistic vision of the High Court’s decision-making process has raised more than a few
eyebrows. While some members of the current Bench are making a concerted effort to
reinstate a less activist reputation, they still find themselves needing to defend their
actions to an ever-more-critical media and public. Legal scholar Anthony Blackshield
noted in a recent panel discussion that the Court “has been subjected to what really were
unprecedented political attacks” over the course of the last decade (Blackshield in
Jackson 2003). In part as a reaction to this, the High Court has appointed a Public
Information Officer to help improve the Court’s relationship with the media.

Another suggestion aimed at improving the Court’s approval ratings is to alter the

format of its opinions. One prominent legal scholar advises the Court to “consider

 

69 . . . .
Havrng been an unsuspecting wrtness, the author was among those extremely startled by the srght of the
fiery sky.

87

making a greater effort to produce joint judgments” (Williams 1999, 145). Observers
lament that the Court’s adherence to a seriatim opinion tradition makes the Court’s output
unnecessarily confusing, especially given that the Court’s most important role is to
clarify the law (Willheim 2002; Campbell 2003). After a review of the High Court’s
practices and procedures, the remainder of this analysis will examine the role of j oint

opinions on the High Court of Australia.

88

CHAPTER THREE: EXISTING TI-IEORIES OF JOINING BEHAVIOR

Despite much progress during the past few decades in the study of courts and
judicial behavior, most theory and data developed by social scientists for
understanding legal systems still remain very much the product of: and thus bound
to, the inevitable peculiarities of the US. context.

—— Burton Atkins (1991, 881)

The study of j oining behavior has become an important part of the judicial behavior
scholarship. Certainly, joining behavior on the United States Supreme Court is the main
focus of what Epstein and Knight call the “strategic revolution” in the study of law and
courts (2000). As Segal and Spaeth point out, the decision whether or not to join the
majority opinion is “among the most important decisions a justice makes” (2002, 387). A
good deal of evidence has been amassed in the study of j oining behavior on the Supreme
Court. Researchers have identified numerous factors that contribute to the formation of
opinion coalitions. These include the policy positions of the individual judges (Rohde
19723), the selection of a majority opinion author (Brenner and Spaeth 1988), the content
of the opinion itself (Schwartz 1992), workload considerations (Spriggs et al. 1999), and
issue specialization (Brenner and Spaeth 1986).

The mere presence of opinion coalitions, of course, is related to the norms and
traditions of the Court and the interpretation of these by the individual judges (Grossman
1968). As the previous chapter demonstrates, the High Court of Australia has
institutional differences—most notably the seriatim opinion tradition—that may give
joining behavior a significantly different purpose and meaning.

The American Literature

The quest to fashion a complete theory of judicial decision making has been going

on for quite some time. In many respects, it has been successful. The major contributors

89

have been proponents of the traditional legal model, the personal attribute model, the
attitudinal approach, and various rational actor models. Together, these have yielded an
enormous list of factors that have been hypothesized to contribute to the understanding of
judicial behavior, with various degrees of empirical support.

The legal model of Supreme Court decision making is traditionally associated with
the legal positivist tradition, emphasizing the mechanical role of the judge in the
application of the law. More recently, judicial behavior scholars have developed a more
sophisticated contemporary version of this tradition, most often referred to as post-
positive legalism. This perspective shares one thing in common with its more
deterministic predecessor—namely, the assertion that decisions of the Justices should be
understood largely in terms of the dictates of the laws that speak to the dispute in
question. All told, the legal model’s understanding of judicial decision output is most
significantly derived from knowledge of the relevant laws (in the legal positivist
tradition) or, alternatively, from knowledge of the judge’s understanding of the relevant
laws (in the post-positivist tradition).

The traditional legal model, holding that case facts and precedent serve as the sole
determinants of the outcome of litigation, was called to task by Pritchett’s The Roosevelt
Court (1948). Pritchett identified policy preference as an important explanatory variable,
along with some of the traditional legal factors involved in understanding judicial
outcomes. He stopped short of an empirical test of such factors, but other researchers
quickly stepped in to conduct statistical analyses (Baum 2003).

The attitudinal model represents a significant break from the legalist traditions.

Developed in response to the American legal realist movement, the attitudinal model is

90

the culmination of a large and varied behavioralist tradition. Political scientists, intrigued
by Pritchett’s introduction of legal realist theory to the discipline, sought to find ways to
empirically validate or invalidate this politicizing revelation about the judiciary.

The earliest attempts to scientifically analyze judicial behavior led researchers to
probe such explanatory variables as social background characteristics (e. g., Nagel 1961;
Schmidhauser 1962), policy-oriented attitudes and values (e.g., Schubert 1965; Spaeth
and Parker 1969), conception of the judicial role (e.g., Grossman 1968), and the
dynamics of small-group membership (e.g., Snyder 1958; Ulmer and Nicholls 1978).
Each of these approaches arose out of the search for satisfactory statistical evidence.
Small-group dynamics research is arguably closer to the rational choice tradition than the
others are, while the role orientation approach is closer to the legalist tradition. This
diversity of behavioral measures illustrates the rapid innovation of the field in the late
19503 through the 1960s. It is also a demonstration of the willingness of behavioralist
researchers to pursue numerous avenues in an effort to gain an empirical handle on the
issue. In terms of Supreme Court research, the attitudinal model has survived the others,
largely because of its impressive ability to explain behavior. The attitudinal model
derives its leverage by concentrating on individuals—specifically on the dtflerences
between individuals in terms of policy orientations.

The rational choice tradition, like the attitudinal tradition before it, grew largely out
of theoretical advances in other fields of research. Rational choice theorists, most
notably William Riker (1962), were beginning to develop a positive political theory,
applying the assumptions of rational choice to political behavior and phenomena. These

assumptions, although stated differently by different researchers, largely center on 1) the

91

ability of actors to rank alternatives in terms of their goal preferences, and 2) the ability
of actors to select from available alternatives in order to maximize the attainment of these
goals (see Epstein and Knight 2000). The focus of the theory is on the individual, but the
theory emphasizes the similarities between individuals instead of their differences. In the
context of judicial decision making, this difference is usually manifested in the
concentration on strategy as shaped by institutions. Although this is different from the
attitudinal model, most incarnations of judicial strategy include explicit provisions for
individual attitudes and goals. Because of the Supreme Court’s majority opinion
tradition, eliciting joining behavior is a crucial part of judicial strategy.

While the seeds of rational choice can be found in the early work of Schubert
(1965) and Pritchett (1961), the first explicit application of rational choice theory to
judicial decision making is found in Murphy’s Elements of Judicial Strategy (1964).
Although the trend toward rational choice theory in the other American political science
sub-disciplines grew at a fever pitch, judicial scholars largely abandoned the approach.
The empirical rewards of the more behavioral strategies were more immediately
apparent. It was not until the work of economist Brian Marks (1989) that rational choice
once again grabbed the attention of the judicial subfield.

Marks, followed by reputable political scientists from various sub-disciplines,
reintroduced rational choice theory to the judicial subfield (F erejohn and Shipan 1990;
Ferej ohn and Weingast 1992). Some argue that there is now a distinct trend toward
‘ rational choice theory in the study of Supreme Court decision making (Epstein and
Knight 2000). Certainly, with a number of significant strategy-based models of decision

making emerging in the last decade, the approach occupies an important space in the

92

contemporary literature (Epstein and Knight 2000; Hammond et al. 2005; Maltzman et al.
2000)

The result is a renewed debate about the appropriate paradigm for understanding
judicial decision making on the Supreme Court. Traditional proponents of the attitudinal
model have been reinvigorated, responding to the challenge of the rational choice
theorists (i.e., Segal and Spaeth 2002). While compelling reasons for employing a
rational choice approach in lower court studies are well understood, the case for
abandoning the attitudinal model at the Supreme Court level is not nearly as clear-cut.

But while the attitudinal model tells us how Supreme Court Justices are expected to
vote on the merits, it has a harder time helping us understand processes that happen
before, during, and after the merits stage. Proponents of the rational choice approach to
studying judicial politics have developed many theories about the constraints on US.
Supreme Court Justices. Constraints are imposed either by external actors limiting the
expression of judicial preferences (Epstein and Knight 2000), or by the judges upon
themselves and each other (Maltzman et al. 2000).

Proponents of applying strategic, rational choice-based theories to the Supreme
Court largely argue that such an application affords researchers the ability to describe
more than simply the votes on the merits (Epstein and Knight 2000). For example, the
decision of a litigant to pursue a remedy through the filing of a petition for a writ of
certiorari is an external constraint on the judges that is not explicitly addressed by the
attitudinal model McGuire et al. 1999). Other such external constraints include the
effects of public opinion, congressional policy positions, and more. In addition to this,

there are internal constraints on the Justices such as collegiality, equity, efficiency, and

93

strategic considerations of the positions of others. The adoption of the rational choice
approach allows for an understanding of judicial constraints, while not abandoning the
idea that judges have (and use) policy preferences to make decisions (Shapiro 1995).

Proponents of the attitudinal model do not concede much to this argument. Segal
and Spaeth (2002) take up the cause against using rational choice models of Supreme
Court behavior. First among their criticisms of this approach is the stringency of the
assumptions that are inherent in rational choice models. This is a familiar argument from
critics of rational choice theory as applied to political behavior generally (Green and
Shapiro 1994). Perhaps more damning, however, is their contention that the empirical
evidence simply fails to bear out the hypotheses derived from these strategic theories ,
(Segal and Spaeth 2002).

The research has begun to come full circle. Several scholars have started to take a
fresh look at the traditional legal model variables, integrating them with the lessons
learned from other empirical work. Segal (1984) has examined Supreme Court search
and seizure cases from the perspective of the legal model. In this research, he uses a
multivariate model in an attempt to bring order to the seemingly chaotic realm of Fourth
Amendment litigation. His findings suggest that traditional legal characteristics do
matter in these particular cases. Aliotta (1988) applies a similar approach to equal
protection cases. Although her legal variables do not correspond in any meaningful sense
with those used by Segal (1984), she does find that they add explanatory power to the
attitudinal variables. George and Epstein (1992) also find evidence to support the

inclusion of traditional legal variables in models of capitol punishment cases.

94

The lesson here, of course, is that competing models of judicial decision making
are only “competing” to the extent that they are incompatible in their approach to a
particular task. A reliance on attitudes is important in most fiinctioning strategic models,
and this pairing can be extremely helpful in understanding decision-making behavior in
courts other than the United States Supreme Court. Ifnothing else, it reminds the
researcher to be aware of potentially significant internal, external, or legal positivism-
inspired constraints on judicial behavior. It also instructs the researcher how to go about
accounting for the impact of any of these constraints once they are identified. Judicial
scholars should not be deterred from considering these resources in other judicial
decision-making contexts; to the contrary, there is reason to believe that such
considerations would carry significant explanatory power across national borders. The
fact that these theoretical resources have been little used in the comparative context, it
seems, is mostly a sign of the relative infancy of the comparative judicial research
program.
Judging Across the World

The business of courts around the world is to settle disputes in accordance with
some sort of written or understood set of laws or norms (Tate et al. 1990). While courts
tend to have similar broad goals, they pursue these goals in diverse ways. Some courts
have the power only to decide small matters of legal application. Others are given the
responsibility of judging the laws created by the legislature against some higher or
constitutional norm. In any case, courts are always situated within a particular social and
political context, and they always have a particular set of institutional features and

procedures that shape the way they administer justice in their own society.

95

Even courts embedded within democratic systems of government have widely
different methods, responsibilities and contexts. As Australian High Court Chief Justice
Murray Gleeson put it, “democracy itself can hardly be said to contain an inherent
definition of the role of the judiciary” (Gleeson 1997). These differences can have
important consequences for the application of the standard American judicial behavior
models even to courts in other liberal democratic countries.

With a few notable exceptions, though, judicial behavior scholars have virtually
ignored courts outside the United States. While Tate (2002b) traces the tradition of
studying non-American courts as far back as the late 18005, the first modern examples of
systematic research on the judiciaries of other nations were written in the 19603. Some of
the eariiest works were translations of behavioralist work from native scholars. The first
research programs that actively compared more than one country in the modern tradition
began to emerge in the late 19605 (e. g., Sheldon 1967). Shortly thereafter, Glendon
Schubert began his global odyssey, applying jurimetrics techniques to courts in various
corners of the world, inspiring others to do the same. Two major collections of
comparative (and quantitative) judicial behavior research appeared at the end of the
decade, setting the stage for what seemed likely to be a major movement in the subfield
(Grossman and Tanenhaus 1969; Schubert and Danelski 1969).

From these dramatic beginnings, however, the resulting trickle of comparative
empirical work has been less than impressive (Tate 2002a). Despite the repeated pleas of
Tate and others, comparative judicial research remains a sub-sub-discipline, with little
acknowledgement from American judicial scholars generally. Indeed, from 1960-1991,

Tate (1992) found that comparative court research occupied less than five percent of the

96

published work in major political science journals. Many wondered what all of the fuss
was about in the first place, why studying courts in other countries may be important, or
even why it could be interesting. Despite the promise of travel to exotic locations, many
gave up on this line of research because of the difficulty in obtaining data—especially the
non-unanimous decisions that were critical to the existing American judicial theories of
the time.

The benefits of comparing judicial institutions across countries should seem
obvious to the student of comparative politics. By studying the American courts almost
to the exclusion of other systems, we have developed intricate models that may not be
generalizable to any other system (see Sartori 1970). Even if we desire only a way to
understand the American system, though, this myopic case-study method may rob us of
potentially useful information. By failing to observe more than one system, we fail to

gain a handle on the importance of a myriad of institutional, cultural and political

variables in the judicial decision-making process.70

One purpose of comparative judicial research may be to relieve the United States
Supreme Court of its most prevalent stigma: that of the Western high courts, it is alone in
its overtly political decision making. Outside of the United States, the legal positivist
model is widely held by lawyers and social scientists alike (Haynie and Tate 1998).
Without systematic empirical analysis, researchers will continue to be unable to identify
those countries that may share this dubious distinction. While our experience in America

has led us to suspect that ours is not the only political high court in town, empirical

 

70

In the early 19808, Tate observed that the study of comparative judicial politics had yet to be integrated
into the study of comparative politics more generally (1981a). Although he provides several concrete steps
toward achieving this goal, progress on this front has remained sporadic at best.

97

evidence is only now beginning to confirm this (Epp 1998; Magalhaes 1998; Smithey
2002; Stone 1992; Tate and Vallinder 1995; Volcansek 1993; Waltrnan and Holland
1987)

Comparative judicial research can also be a worthy pursuit in its own right. The
one area where the study of courts has been taken up by comparative politics scholars is
in the area of democratization research. One section of this literature examines the
relationship between institutional arrangement (i.e., presidential or parliamentary system)
and the independence of the judiciary (F erej ohn 1995). Ferej ohn’s findings support the
idea that presidential systems afford the judiciary more independence, which is associated
in other research with the destabilization of fragile democracies (Tate et al. 1990). Other
research shows that single party dominance begets low levels of judicial independence
(Ramseyer 1994; Ramseyer and Rasmusen 1997), which in turn presents complications
for democratic consolidation (Rose and Shin 2001). In short, the comparative study of
courts can assist not only in the validation and refinement of models of American judicial
behavior, but it can also assist in the understanding of political systems generally.

Australian Jurimetrics and Beyond

While many comparative judicial scholars lament the lack of attention to courts
outside of the United States (Tate 1996), it may be unfair to conclude that the Australian
High Court has been neglected completely. The field of jurimetrics was especially well-
received by a handful of Australian legal scholars. Jurimetrics, or the application of
quantitative mathematical methods to legal problems, arose from the behavioralist focus
on “attitudinal” variables in the study of judicial decision-making behavior. Very quickly
following the behavioral revolution in the study of American courts, like-minded

researchers abroad began to apply similar techniques to the Australian High Court.

98

After his book The Judicial Mind (1965) inspired the behavioral approach to
studying American courts, Glendon Schubert himself contributed much to the early
jurimetrics work in Australia. In the late 19605, Schubert published two articles on the
High Court (Schubert 1968; 1969a), largely in an attempt to verify the findings of his
work on the US. Supreme Court. He identifies two significant ideological scales for the
tenure of Chief Justice Dixon (1952-1964). The first is called the “X—scale,” and is made
up of issues including support for economic underdogs, religious morality, security,
socialism, and government regulation of business. The “Y-scale” is made up of more
broad theoretical divides including adherence to a strict interpretive theory of the
constitution, perspective on the role of judicial review, and judicial and federal
government centralization (Schubert 1969a). He finds that a judge’s position on these
two scales is related to that person’s age, state of residence, and score on a
conservatism/liberalism scale.

This work was closely followed by Australian researchers, some of whom
conducted their own research in this vein (Blackshield 1972; 1978; Douglas 1969). By
the end of the 19705, nearly forty year’s worth of High Court data had been analyzed for
decision-making patterns (Blackshield 2002c). The result of these studies had some
findings in common with Schubert’s — most importantly that at least two distinct
dimensions of decision making existed in the High Court during the analyzed years. One
of these dimensions, according to Blackshield (1972), roughly corresponds to the familiar
left-right dimension of policy preferences. His findings also pointed to a second relevant
dimension, this one corresponding with differences in attitudes about the distribution of

legal control of the institutions of government.

99

The systematic study of the Australian High Court was not limited to Schubertian-
style endeavors. A parallel line of Australian jurimetrics research was inspired by the
work of American researcher Fred Kort (1957; 1963). This research applies
mathematical methods in an attempt to create models of judicial votes, thus predicting
future court outcomes. Kort’s models place heavy emphasis on the importance of case
facts in the production of judicial outcomes. Prediction-minded researchers like Alan
Tyree (197 7; 1981) have analyzed the High Court in this way (Blackshield 2002c).
Tyree attempted to salvage the practice of using content analysis to model case facts by
relaxing the strict deterministic assumptions of previous research (Tyree 1981). In doing
so, he finds evidence to support the idea that case facts are at least related to decision
output, even in Australia’s highest court. He also offers an alternative to the Schubertian
classification of similarities and differences between judges. By using a fact-based
distinction, he created issue-specific multidimensional maps in ideological space. In this
way, he is more confident that he is, indeed, comparing apples to apples (Tyree 1977).
He concludes that this non-metric multidimensional scaling method allows structure to
emerge in data that were previously devoid of any obvious pattern.

In the 19705 and 19805, as interest in comparing courts waned among American
scholars (Tate 2002a), a similar trend formed in Australia. Australian researchers turned
away from this type of research for practical and theoretical reasons. The limits of the
factor analysis techniques had largely been reached, especially at a time where computing
power was at a premium. Additionally, Australian legal scholars were faced with a
general lack of statistical training and doubts about the ability of the methodology to add

significantly to the existing knowledge (Blackshield 2002e).

100

This lack of confidence in statistical methodology may be due to the ‘legalist’
culture among Australian scholars and High Court observers at the time. From the
legalist perspective, judges are constrained in large part by legal precedent, and outside
influences are irrelevant to the decision-making process. Legalism does not deny the
political nature of the questions before the Court, nor does it eschew the policy
implications of the Court’s decisions (Gageler 2002b). Instead, it simply denies the use
of such non-legal factors in the decision-making process. Members of the High Court
have been known to perpetuate the idea that decisions are made exactly this way
(Galligan 1987), and that creating law on the basis of extra-legal concerns is simply
beyond the scope of the High Court’s role (Gageler 2002b). The Court has traditionally
held fast to the legalist framework as its sole method for interpreting common law,
statutory law, and the constitution (Thompson 1982).

The High Court was not conceived until well after the famed Marbury v. Madison
(1803) case. The constitution that created the High Court modeled it largely after the
US. Supreme Court as it operated in the late 19th century, and gave it the power of
judicial review outright (Galligan 1991). Unlike the US. Constitution, though, the
Australian Constitution does not have an explicit list of individual rights. The existence
of the Bill of Rights means that “each Justice is inevitably drawn deeply into its boiling
brew of politics, the attributed or perceived social values of innumerable groups” of
American politics (Callinan 1999, 3). The absence of such a constitutional document in
Australia means that the Australian High Court Justices are not directly subjected to the
aforementioned fate. On the other hand, the role of the High Court in determining which

rights are implied in the document is necessarily more ambiguous than in the United

101

States (Kirby 1999). In most other ways, though, the Australian Constitution is much
more explicit than the American Constitution, making it less necessary for the High Court
to appear to be divining the true meaning of the document (Crawford 1993). In this way,
with the exception of civil rights cases, the judges in the Australian system have an easier
time maintaining the air of legalism. Before ascension of Sir Anthony Mason to the High
Court Chiefship, the High Court generally stayed away from addressing civil rights cases
beyond assenting to the will of Parliament, making their claims of legalism nearly
airtight.

For this reason, legalism is particularly important in the context of the High Court.
Not only is it the traditionally accepted interpretation strategy but it is also one of the few
defenses the Court has against its particular political context. Because of the history of
conflict between the Labor and Liberal parties, the liberal consensus enjoyed in America
is not a guarantee in the Australian context. The Labor party was virtually excluded from
the drafting and ratification process of the Australian Constitution, and their main
political goals remain outside the scope of the constitutional framework. Since the
ratification of the US. Constitution, political debate in America has generally taken place
inside the framework of the institutions prescribed by the Constitution. In Australia,
however, the main points of contention between the parties run much deeper—to the very
core of the liberal institutions contained in the Australian Constitution. The authors of
the Australian Constitution were decidedly anti-centralist, and the labor movement’s
interests in creating a strong central authority were generally excluded from the
discussion (La Nauze 1972). The labor movement saw the Australian Constitution as an

obstacle to its objectives, and approached it with a degree of hostility (Hirst 1995). This

102

debate has continued into modern times, and the High Court has often been drawn into
the debate (Walsh 1997). Because of this unique political difficulty, the Justices rely on
the popular legalist rhetoric in order to show that they remain impartial in the face of this
potentially divisive conflict (Crawford 1993).

This is not to say that the US. Supreme Court does not need the support of the
people in order to maintain its power. On the contrary, the constitutional weakness of the
Supreme Court means that it needs the support of the public in order to keep its
legitimacy (Franklin and Kosaki 1989). Research has shown that Justices, especially the
moderate ones, are sensitive to the ebbs and flows of public opinion (Mishler and
Sheehan 1993; 1996). Justice Kirby warns that any dissatisfaction with the US. Supreme
Court might foreshadow similar feelings about the Australian High Court:

Nowadays, abuse and insult are not unknown. Every judge can accept criticism

and acknowledge that sometimes it is warranted. But Australians must be careful

to avoid the perils of damage to our institutions that the judiciary faces in the
United States. (Kirby 2001)

That this observation comes from one of Court’s leading critics of Australian
legalism is of particular interest. The commonly accepted way to defer criticism of the
Australian judiciary has been to invoke a legalist perspective — whether in practice or
only in rhetoric. Indeed, analysis of mid-20th century Communist Party cases indicates
the High Court was particularly unresponsive to public opinion (Sheldon 1967).

Although the legalist culture has begun to wither among Australian legal scholars
and casual court observers alike, the concern for legitimacy remains. This has left a
significant rift among members of the Court, both past and present. While some Justices
adamantly believe that legalism is the only way to secure public confidence in the

judiciary (e.g., Gleeson 2000a), others argue that a failure to accept the duty to craft law

103

in the best interests of the Australian people might be detrimental to the Court’s
legitimacy (Kirby 1996).

No matter which of these camps is correct, experts and citizens alike have begun to
speculate that some set of extra-legal factors influences the policy output of the Court
(e.g., Galligan 1987; Haultain 1997). Some of the opposition to the traditional legalist
view likely comes from cues given by some members of the Court. Like the US.
Supreme Court, the High Court decides many cases in which all of the members fail to
agree on what should be the proper outcome. In these cases, both the majority and the
dissenters put forth logical arguments to support their positions. In most cases neither
argument is logically necessary, and each is logically feasible (Sawyer 1967). As
witnesses to the empirical studies of the US. Supreme Court, Australian legal scholars
are beginning to protest loudly against the idea that even a relaxed version of the
traditional legalist model is an appropriate way to characterize the work of the High
Court (Galligan 1987).

This realist critique of legalism is now coming not only from onlookers but also
from the members of the Court (Kirby 1983). The High Court under Chief Justice Mason
(1987—1995) moved down a path of innovation and activism. Some of the Justices began
to embrace their roles as policy-makers, and they tipped their hand by showing
willingness to undertake sharp departures from the status quo. Many of the Mason Court
Justices were taught by prominent legal realists. Some even publicly acknowledged the
work of jurimetrics scholars directly, taking seriously the patterns of social attitudes in

judicial decisions that these scholars had uncovered (Murphy 1980).

104

More recently, Australian scholar Russell Smyth has almost single—handedly
reinvigorated the empirical study of judicial behavior in Australia. The High Court itself
publishes basic information in its annual reports including workload data in various
breakdowns and budget expenditures, despite the strains this puts on the Court’s data
management system (Popple 1999). Sometimes, the judges themselves make use of these
data in their speeches or extra-judicial writings (Gleeson 2004b). A number of
interesting hypotheses, many derived from the American literature, have also been tested
using empirical data about the High Court. Research has found a relationship between
case complexity and the reliance on citations of authority in judicial decisions (Smyth
2003a). One analysis fails to find support for party capability theory (Smyth 2000), but
another finds evidence to the contrary (Haynie et al. 2001). The number and type of
interveners in the High Court have been evaluated over time (Pierce 2006), and these data
have been systematically compared to the Supreme Court of Canada (Williams 2000).

The role of the associates has been compared to the US. Supreme Court clerks
using survey data (Leigh 2000). The use of American precedents in the High Court over
time has been examined (von Nessen 1992). An analysis of voting blocs on the Latham
Court presents an interesting (if somewhat problematic) assessment of coalition success
(Smyth 2001), and a later incarnation of this work finds support for the application of
many American judicial behavior theories to the Latham Court (Smyth 2002b).

Another area of interest to scholars has been the prestige and productivity of judges
at different stages of the judicial life cycle. Valiant attempts have been made to quantify
judicial performance according to productivity, citations, extra-judicial writings and

reversals (Smyth 2005a). Judicial productivity has been found to follow a life-cycle

105

trend, rising to a peak and then declining again near the end of the judicial career
(Bhattacharya and Smyth 2001a). Judicial productivity, along with ideological
compatibility between the judge and the current administration, may also help to predict
the timing of retirements (Maitra and Smyth 2005). Previous judicial experience,
combined with youth and a conservative appointing prime minister, have been associated
with judicial prestige, measured in terms of citation rates (Bhattacharya and Smyth
2001b)

A very popular line of research attempts to define (Lynch 2002a) and to explain
variations in dissent rates (Smyth 2004) and other separate opinions (Groves and Smyth
2004) over time. Fluctuation in nonconsensual behavior have been explained as a
function of institutional changes and different leadership styles (Smyth and Narayan
2004), as well as attitudinal factors (Smyth 2003b; 2005b), acclimation effects (Smyth
2002a) and consensual norms (Narayan and Smyth 2005; 2007; but see Smyth 2002c;
Wood 2008). It is on this line of research that the present analysis builds.

Conceptualizing Joining Behavior

In a situation where a number of different people are asked to come to a decision on
a topic, there will likely be some level of disagreement between the parties. This is also
true in the case of judges. Unless we are willing to accept the legalist doctrine that the
outcome of each case is logically derived from the law and stare decisis, there is the
potential in each case for some level of disagreement between the judges. This level of
disagreement can be thought of as a continuous variable, with no necessary limits on its
value. The extent of these disagreements cannot be directly observed by the researcher,

though, or even by the Justices themselves.

106

Decision Patterns

In Australia, as in the United States, there is an institutionalized proxy measure for
such disagreement. If the level of disagreement reaches a certain critical level on a given
case, judges will cast dissenting votes. From a strict legalist perspective, dissents are the
result of the breakdown of the process of finding the legally necessary outcome. Critics
of the legalist model, however, see the dissent as a result of ideological disagreement
among the judges (Brace and Hall 1993). Because of the independence of the judiciary
and the design of the court system, it is not necessary for judges in the United States or
Australia to assent to decisions with which they do not agree. This leaves the researcher
with an observable measure of disagreement between the judges, the importance of which
has been well documented in the study of judicial behavior in the American court system
(Segal and Spaeth 1993).

In the American system, concurring opinions might also add an important piece of
information to this puzzle. Indeed, much of the research on Supreme Court judicial
behavior concentrates on voting behavior, typically omitting the distinction between a
concurrence and assent to the majority opinion. A scale could be constructed using
information about both concurrences and dissents to incorporate this additional opinion
agreement information (see Handberg 1978). The exact role of the concurring opinion in
the American context, however, is not altogether clear. Typically, concurring Opinions
are assumed to be a middling option, indicating less agreement with the majority than a
vote for the majority opinion, but more agreement than the filing of a dissenting opinion
(Segal and Spaeth 1993). The precise mechanism, however, is very difficult to
understand in the American context, especially from the perspective of current models of

judicial decision making (Maveety 2003).

107

The seriatim format of the Australian High Court decisions makes the
concurrence/maj ority opinion distinction nearly impossible. Each judge tends to write a
separate opinion, whether or not the opinion expressed supports the final majority vote.
In any case, a Justice writing a concurring opinion still votes with the majority of the
members of the Court, even though there is usually no single majority opinion with
which to concur. For this reason, concurrences and majority opinions typically look
exactly alike in the vote totals. This fact removes from practicality the option of using
concurring opinions as a refinement to an agreement/disagreement scale. It does
introduce another possible avenue: the phenomenon of j oint opinions. Without the
institutionalized motivation to produce a majority opinion, why do the judges engage in
such behavior? The decision to join or not to join is subject to far fewer constraints in the
absence of the majority opinion tradition. In this way, patterns of “inter-agreement in
opinions makes possible the discrimination of differences among the justices” in all
cases, be they unanimous or not (Schubert 1969b, 339).

Joining Behavior in the US. Supreme Court

The motivation to join the opinions of other judges is easier to explain in the
American context than in the Australian context. The majority opinion “is the core of the
policy-making power of the Supreme Court” of the United States (Rohde 1972b, 653).
When a joint opinion carries additional force beyond that of an individual concurrence or
dissent, the desire to achieve a joint Opinion is to be expected. In the US. Supreme
Court, there is a tradition “which generally encourages each justice to join the majority
opinion whenever possible” (Atkinson and Neuman 1969, 273). This pressure, coupled
with the need to establish and maintain a relatively coherent body of precedent, serves as

a force to keep judicial individualism at bay (Kelman 1985).

108

If a judge finds the majority opinion acceptable, there is “little need for a justice to
push assertively to change it or to write separately” (Spriggs et al. 1999, 489). When a
judge finds the majority opinion unacceptable, he faces a number of potential options.
The first and most obvious choice is to simply dissent from the majority. Dissenting
behavior is not uncommon on the Supreme Court, and is firmly rooted in the history of
Anglo-American jurisprudence (Bergman 1991). Historically, Supreme Court onlookers
saw dissenting opinions as a useless and unnecessary drain on the Court’s legitimacy
(Stager 1925). While there is still a bias against dissenting among some members of the
legal community (Campbell 1983), the practice is now generally understood to be a
natural part of the decision-making process (Maveety 2003). As William 0. Douglas
puts it,

When judges do not agree, it is a sign that they are dealing with problems on which

society itself is divided. It is the democratic way to express dissident views.
(Douglas 1948, 106)

The modern Supreme Court, with its discretionary jurisdiction and its position at
the apex of the judicial system, is far more likely to deal with these divisive issues than
most other American courts (Post 2001). Evidence has shown that judges become more
likely to dissent once they are elevated to the Supreme Court bench (Gerber and Park
1997). Indeed, some dissenting opinions have become “canonized,” or received by the
legal community as among the “highly authoritative legal texts that command special
reverence in the law” (Krishnakumar 1999, 781).

As a second option, the Justice may choose to author a concurring opinion. This
may take the form of either a regular concurrence or a special concurrence (Kinnan
1995). In a regular concurrence the judge agrees with the majority on the result of the

case and the reasons, but decides to write separately anyway. A concurrence is what the

109

term “concurrence” is more commonly used to describe. This type of concurrence, also
called a “concurrence in judgment,” indicates agreement on the disposition of the case
but not the reasons of the majority. The writing of a concurring opinion, however, is not
without its own costs. Besides the extra time it may take to assemble a concurring
opinion, the practice is seen as particularly distasteful, even among members of the Court
(Maveety 2003).

The third option is to bargain with the author of the opinion seeking
accommodation for his views. Ifthe differences are not extreme, the judge may have a
strong incentive to pursue this path. The higher precedential value of the majority
opinion may entice fence-sitters to attempt to shape that opinion to better reflect their
own preferences (Spriggs et al. 1999). Indeed, numerous researchers have found
evidence of such bargaining, or strategic behavior, on the Court (Epstein and Knight
1998; Murphy 1964; Wahlbeck et al. 1998). In an attempt to secure accommodation,
Justices can exercise various levels of pressure on the majority opinion author. These
actions, when combined with the majority opinion author’s desire to keep a majority,
may lead to accommodation. Absent the need for a majority opinion, however, this may
not be the case:

The collaborative nature of the opinion-writing process stems largely from the

author’s need to have a majority of justices join the opinion once it is circulated

before it becomes the opinion of the Court. This institutional rule therefore

provides incentives for justices to bargain with the majority opinion author and for
the author to sometimes accommodate their concerns. (Spriggs et al. 1999, 486)

Since the end of the seriatim opinion tradition on the US. Supreme Court, the
absence of a single majority opinion has been frowned upon (Novak 1980). In the event
that a majority opinion cannot be agreed upon, the Court typically produces a plurality

opinion — or an opinion assented to by a plurality of the Justices. Plurality opinions lack

110

the precedential force afforded to majority opinions, and are often criticized for creating
chaos in the legal community (Novak 1980). Concurring behavior itself has been
criticized for emasculating the Opinion of the Court, regardless of whether or not the
behavior leads to a plurality decision (Davis and Reynolds 1974). Participation in a
special opinion leading to a plurality outcome can work against a Justice’s self-interest
(Hammond et al. 2005).

This is not to say that all joining behavior on the Supreme Court is easily
categorized as a rational compromise in an attempt to further policy goals. In order to
qualify as the “Opinion of the Court”—or the majority opinion—the opinion must be
joined by a majority of the Justices. In most instances, this means that the minimum
number of j oiners necessary for a winning coalition is five (opinion author included).
The author of the majority opinion, then, will need to gather four additional coalition
members. As Rohde notes,

The opinion he drafts will be shaped by his own preferences, but he is not a free

agent. If he is to attain a winning coalition, he must gain the assent of at least four

other justices in the opinion. The opinion writer is thus forced to bargain with the
other justices. (Rohde 1972a, 214)

Once this minimum winning coalition number of five is reached, though, the
motivation for additional compromise or joining behavior is slightly more difficult to
explain. While the majority conference coalition needs to maintain a minimum winning
coalition, there is little need to entice more than the minimum number of j oiners to sign
the opinion. From the joiner’s perspective, why Opt to join the majority opinion once a
minimum winning coalition has already been reached? Rohde finds some evidence that
the existence of larger than necessary majority opinion coalitions is related to a threat or

perceived threat to the Court as an institution (Rohde 1972a). Other researchers, though,

111

have discounted the empirical evidence surrounding this claim (Brenner 1979; Giles
1977), suggesting that the minimum winning coalition hypothesis finds no support
regardless of the presence of a threat.

Saul Brenner (2003) outlines several possible explanations for unnecessary joining
behavior. Many of his hypotheses are specifically relevant only to joining a majority
opinion, and do not adequately explain the decision to join a special opinion—any
opinion that is not the majority opinion. The simplest explanation for unnecessary
joining behavior is an attitudinal one: that the judge agrees with the reasons expressed in
the opinion. One way to illustrate this is to look at situations where a judge decides to
join a colleague’s special opinion. While there is institutional pressure to achieve a
majority for the opinion of the Court, there is no such pressure to form coalitions for
other types of opinions. Indeed, Segal and Spaeth argue that, because there is no
institutional pressure to join in concurring or dissenting opinions, these opinions “reflect
the strongly held policy views of their authors” (2002, 395). The bargaining and
accommodation needed to attract joiners to a majority opinion is not likely to occur in the
absence of the institutional pressure to achieve at least a minimum winning coalition.

Because there are no institutional benefits to assembling a coalition for these
special opinions, the strategic motivations to join fall away and leave policy preferences
as the most obvious explanation. Judges are more likely to join a concurring opinion
when they are ideologically distant from the majority opinion writer (W ahlbeck et al.
1999). When judges do join a special Opinion, they usually join an opinion written by a
close ideological ally (Segal and Spaeth 2002). Even still, judges with similar policy

goals are likely to disagree to some extent about the precise details of the reasons.

112

Brenner (2003) also cites a number of strategic motivations for joining an opinion
when there is no institutional pressure to do 50. First, a judge may join an opinion if his
views have been accommodated by the writer. Indeed, this seems to imply strategy on
the part of the opinion author, and not on the part of the joiner. If the joiner’s views have
been accommodated, then joining the opinion is merely an exercise in pursuing policy
preferences. There is some reason to expect a majority opinion writer to accommodate
the views of an additional joiner once the minimum winning coalition has been
assembled. Evidence from the Warren Court indicates that minimum winning coalitions
were more likely to break up prior to the final disposition, giving majority opinion writers
a reason to solicit additional joiners (Brenner and Spaeth 1988).

Because the majority opinion sets forth the rationale for subsequent decisions, a
larger-than-minimum winning coalition may convey a consistency and certainty in the
reasons given by the Court (Murphy 1964). The 5-4 decision set a precedent “which
might be overturned if membership changes or if one of the justices in the majority
changes his mind” (Brenner et al. 1990, 309). Additionally, the compliance of lower
courts may be more likely when the majority opinion coalition is larger (Benesh and
Reddick 2002; Woodward and Armstrong 1979; but see Johnson 1979).

The classic example of an accommodation strategy to achieve a unanimous opinion
is the effort of Earl Warren in Brown v. Board of Education.71 In this case, it was the
likelihood of widespread external resistance to the Court’s ruling that motivated
Warren’s pursuit of a unanimous decision. There is less reason to believe, though, that

the author of a special Opinion would be interested in courting joiners (O'Brien 1999).

 

71(1954) 347 US. 484.

113

Additional signatures on a special opinion may say something about the ability of the
author to persuade his colleagues (Segal and Spaeth 2002); concerns about institutional
legitimacy and providing a unified front to external actors are less relevant to special
opinions.

A second strategic reason for unnecessary joining given by Brenner (2003)
suggests that the joiner may concede some policy ground in an attempt to achieve this
gain in legitimacy. He hypothesizes that a Justice may join a majority opinion after the
minimum winning coalition has been solidified as a way to strengthen the decision in the
eyes of outside actors. Because he lists this hypothesis separately from the previous one,
it implies that a Justice may join for this reason even if his policy positions have not been
accommodated. For the same reason that a special Opinion writer is unlikely to solicit
joiners, though, Justices would be unlikely to join a special opinion that does not reflect
his policy preferences.

Thirdly, Brenner (2003) suggests that a judge may strategically join a majority
opinion because the presence of a separate opinion may raise the salience of the majority
opinion. This reason seems to contradict the previous one. In the previous hypothesis, a
unanimous opinion is assumed to strengthen the holding in important and controversial
cases. In the present hypothesis, though, the presence of a special opinion is assumed to
call attention to the majority opinion, which supports something other than the potential
joiner’s ideal policy. There is empirical evidence indicating that winning coalitions in
nonsalient cases tend to attract more joiners than those in salient cases (Brenner et al.
1990; Maltzman et al. 2000). In the Burger and Rehnquist Courts, 80% of cases without

a majority opinion were in the highly salient area of civil rights and liberties (Segal and

114

Spaeth 2002). This hypothesis says nothing about joining behavior for special opinions.
Ifthe Justice believes that the presence of a separate opinion detracts from the majority
Opinion, joining a special opinion has essentially the same effect as writing his own.
There would seem, then, to be little reason to accept and join a sub-optimal special
opinion.

The fourth strategic explanation given by Brenner (2003) is that a Justice may join
a majority opinion in order to increase the chance that the author will return the favor in
future cases. Because the Justices on the Court are in an iterative game, it is reasonable
to expect that their long-term interest can be furthered by working to secure cooperative
relationships with their colleagues, even if it may mean compromising (Axelrod 1984;
1986; Murphy 1964). Empirical evidence suggests that such a cooperative strategy can
provide real payoffs for the joiner in future attempts to marshal a majority coalition
(Collins 2007; Wahlbeck et al. 1999). In this research, though, the measure of
cooperation is a firnction of “the percentage of time that the author joined another
justice’s separate opinions” (W ahlbeck et al. 1999, 500). In other words, this research
finds that joining a special opinion—not a majority opinion—can yield benefits in later
attempts to form a majority coalition. Ifthe Justices perceive that this is true, this may
provide a motivation to join a special opinion he might not otherwise join.

Small group theory implies that judicial behavior is motivated in part “by
considerations of group solidarity and intragroup harmony” (F esta and Vichules 1968,
540). The last group of hypotheses put forward by Brenner (2003) is derived from this
perspective. First, he suggests that a Justice may choose to join the majority opinion

because he has an interest in being on the winning side. Certainly, this is one common

115

characterization of Justice Sandra Day O’Connor’s approach (Maveety 1996). Early
small group research argued that an individual member’s power is related to the
probability that he would be the pivotal member of a minimum winning coalition in any
given decision (Schubert 1964; Shapley and Shubik 1954).

Maveety argues that O’Connor’s experience in the Arizona state legislature taught
her “that the nature of a compromise can be greatly affected by those who consistently
ally with the majority coalition (Maveety 1996, 4). As the majority opinion contains
legal rules with fiiture precedential value, Justices may decide to join in an effort to
influence the substance of these rules (Spriggs et al. 1999). Once the minimum winning
coalition has been assembled, all those remaining Justices who wish influence the text of
the final majority opinion would find it prudent to join the majority. This approach
worked for O’Connor at least, and her “propensity to join the winning coalition and to
retain majority-side membership has kept O’Connor from being marginal to the Court’s
discussions of policy” (Maveety 1996, 69). This interest in contributing to the majority
opinion, along with an institutional bias against unnecessary concurring opinions, might
encourage Justices to refrain from writing separate reasons when they agree with the
majority’s result (Rohde 1977; Caldeira and Zorn 1998). Implementation of this strategy,
though, seems to preclude the unnecessary joining of special opinions. If the goal is to be
a part of the winning coalition, then nothing is to be gained from joining in concurrences
or dissents.

The second small group hypothesis presented by Brenner is that a Justice may join
the majority opinion because everyone else has, and “the remaining justice feels

uncomfortable as the sole hold—out” (2003, 277). Howard associates this behavior with a

116

freshman effect, noting that Justice Cardozo “frequently [voted] alone in conference
before ultimately submerging himself in a group opinion” (Howard 1968, 45). Indeed,
empirical research on voting fluidity between the conference vote and the final case
disposition finds that a lone dissenter is twice as likely to join the majority than is a
Justice who is part of a three-member dissenting coalition (Maltzman and Wahlbeck
1996b)

In very important cases, this change in opinion may be a result of the
accommodation efforts undertaken by the majority Opinion author and/or the Chief
Justice. Howard argues that this happened in many cases, including the Japanese

internment cases, with Justice Murphy suppressing a dissent in Hirabayashi and Douglas

withholding a dissent in licortmt'azsu72 (Howard 1968). Joining as a way to avoid being
the lone dissenter provides an explanation for unnecessary joining of a majority opinion,
but does not help to explain the joining of a special opinion.

Brenner’s third small group hypothesis seems to predict very different behavior
from his second. Here, he asserts that a Justice may join because “she does not care a
great deal whether the Court’s opinion is written one way or another or because the case
is not particularly salient to her” (2003, 277). It is difficult to measure the salience of a
particular case to a particular judge, and researchers usually resort to general measures
deriving from contemporaneous (Epstein and Segal 2000), or post hoc assessments of
case salience (Brenner and Arrington 2002). As such, the empirical research on this issue
is generally at the level of the Court and not the individual Justice. Much of the evidence

from voting fluidity analyses finds that judges are more likely to sign on to majority

 

7
2 Hirabayashi v. United States (1943) 320 US. 81; Korematsu v. United States (1944) 323 US. 214.

117

opinions in less salient cases (Dorff and Brenner 1992; Spriggs et al. 1999). Again, this
hypothesis does not provide as much traction when we are considering unnecessary
joining behavior involving special opinions. Ifthe judge is relatively ambivalent as to the
outcome, other considerations may drive the decision to join the majority opinion,
including the first and second hypotheses derived by Brenner (2003) from small group
theory.

The last hypothesis concerns the practical cost of authoring separate opinions.
Brenner suggests that Justices may join to avoid spending “the time and effort necessary
to write a persuasive concurring or dissenting opinion” (2003, 277). Because this
hypothesis is separated out from the third small group hypothesis, we can assume that it
is not intended to convey simply a lack of interest in the material. Instead, this argument
implies that a Justice is responding to workload pressures. Baum (1997) includes
minimizing workloads among the important judicial goals. Unlike the other small group
hypotheses, the interest in easing workload pressures may apply to unnecessary joining
behavior involving both majority and special opinions. Some argue that Justice
Brennan’s attempt to organize joint opinions for the minority bloc succeeded because the
other members of the minority felt significant workload pressures (Cook 1995; O'Brien
1999). While the author of a special opinion may have less motivation to accommodate
potential joiners, the benefit to the joiner is likely the same.

While some empirical research supports the assumption that heavy workloads can
factor in to opinion-writing decisions, the evidence is far from decisive. Justices in the
minority are more likely to acquiesce to the majority opinion without first circulating

special opinions when their workloads are heavy (Spriggs et al. 1999). On the other

118

hand, majority opinion authors with heavy workloads may be less willing to
accommodate his colleagues (W ahlbeck et al. 1998). Individual workload does not seem
to increase the likelihood that a Justice will join the majority, although this likelihood
does increase at the end of the term (Maltzman et al. 2000). A large number of cases on
the Court’s docket does not seem to increase the likelihood that a member of the minority
will switch sides to join the majority opinion (Epstein et al. 2001).

As this discussion illustrates, most of the joining behavior research concerning the
Supreme Court has to do with the decision to join the majority opinion. Much of the
evidence outlined above suggests that, for various reasons, Justices will join majority
opinions even if they do not match their most preferred ideological preferences. This
makes sense, given the importance of the majority opinion in this particular institutional
context. The need to amass a majority opinion coalition changes the calculation for
judges. Because of the unique importance of the majority opinion, the existing American
research spends little time investigating other types of unnecessary joining behavior. In
addition, the rate of j oining behavior on special opinions is quite low on the Supreme
Court (Segal and Spaeth 2002). Because of the seriatim opinion tradition in Australia,
however, all joining behavior in the High Court resembles this type of unnecessary
joining behavior. As such, all joining behavior on the High Court takes place in the

absence of a clear institutional incentive to bargain and accommodate.

119

Joining Behavior in the Context of the High Court
As the previous chapter indicates, the High Court is producing joint opinions at a

much higher rate than it did several decades ago.73 Explaining the reasons behind the
decision to join is quite difficult, especially in the absence of a majority opinion tradition.
This growing number of joint opinions may reflect a desire of the Justices to clarify
important points of law for lower cOurts or to build legitimacy in contentious areas.
Certainly, it is easy to imagine that efficiency concerns may have instigated this shift.
Some of this recent increase may have to do with the personal relationships between the
judges (Coper 2002c). One High Court observer notes that Justice Gaudron often joined
with Justice Gummow because she thought he was quite clever (McMillan 2003). An
individual’s prior experience or issue expertise may well contribute to a tendency to join.

It is also important to consider the role of ideological agreement in the decision to
join. Once we take away the structure of opinion assignments, we should able to see a
clearer picture of the “justice’s attractiveness to other justices” in terms of opinion
writing (Handberg 1978, 373). Without the institutionalized pressure to negotiate a
majority opinion, there is less reason to expect the judges to accommodate the views of
his colleagues in order to obtain a joint opinion. Ifthis is the case, things like workload
concerns, issue expertise and clarity should not satisfactorily explain the incidence of
joint opinion production on the High Court. Instead, these joint opinions, like the special
opinions of the US. Supreme Court, would comprise “a much richer source of

information about judicial relationships” (Segal and Spaeth 2002, 395).

 

73 By contrast, the US. Supreme Court is producing fragmented decisions at a growing rate (Segal and
Spaeth 2002).

120

Before we can investigate these hypotheses, however, it is important to address
some of the recent empirical research on consensual norms in the High Court. Although
the High Court formally operates using a seriatim opinion—writing tradition, some
researchers have suggested that this tradition is being subverted by a defacto norm of
consensus (Narayan and Smyth 2005). If this is true, then High Court Justices are
increasingly subject to an institutional norm encouraging joint Opinions just as Supreme

Court Justices are. For this reason, the next chapter investigates this question in depth.

121

CHAPTER FIVE: CONSENSUAL NORMS ON THE HIGH COURT

Unlike their position in America, concurring judgments are simply the usual
method by which judges in the English tradition express agreement with each
other, despite occasional instances of welcome unanimity.

—Andrew Lynch (2003a)

This analysis begins with the premise that, with the existence of the seriatim
opinion tradition in Australia, there is no functioning consensual norm to encourage
judges on the High Court to produce joint Opinions. Ifthis is the case, the presence of
joint opinions would be unexpected given the standard explanations used in the American
attitudinal and strategic decision-making models. These joint opinions would either be
expressions of attitudinal congruence between judges, or attributable to some other
mechanism (such as workload pressures, deference to issue experts, etc.). The first step
to approaching the question, then, is to determine the extent to which the premise is
reasonable.

To do this, it is necessary to examine the extent to which the High Court as a whole
subscribes to some sort of consensual norm. A norm is “a social rule that does not
depend on government for either promulgation or enforcement” (Posner and Rasmusen
1999, 369). The consensual norm—at least in its American formulation—is a social rule
that causes judges to feel “obliged to mask their differences from public view by working
them out via a single majority opinion” (Epstein et al. 2001, 364). Certainly, the seriatim
tradition is also a norm, in that under it judges feel “a very strict obligation under the
constitution to express their individual opinions” (Clark 2003, 10). It is reasonable to
assume that the High Court’s seriatim opinion tradition precludes any type of functioning
consensual norm (Smyth 2002c), but this is an empirical question. This chapter will

approach this question by adapting the relevant Supreme Court research to accommodate

122

the High Court’s unique institutional setting. This is helpful for the present analysis, as it
provides an institutional-level baseline of normal joining behavior. It will also provide
evidence as to whether joining behavior is largely norm driven (as it seems to be on the
Supreme Court), or if it may be driven by some other mechanism.
Defining Types of Opinion-Writing Behavior

Before considering the factors influencing joining behavior in the High Court, it is
necessary to address an important practical concern. In the High Court of Australia, it is
nearly impossible to distinguish a joiner from an author. In the US. Supreme Court, a
measure of writing behavior is easy to create. With the exception of per curium opinions,
the writer of a Supreme Court opinion is typically identified in the written reports. This
makes it possible to attribute writing behavior to those judges who have written an
opinion, be it the majority opinion or some other opinion. The traditions and reporting
style of the High Court, however, create significant differences that make the process
more complicated.

Terminology

Even among Australian legal scholars and practitioners, the use of opinion-writing
terminology is far from precise (Lynch 2002a). To complicate matters further, the
American and Australian literature often use the same words to describe very different
things. The American use of opinion-writing vocabulary is more consistent, and provides
a solid point of comparison for settling on terminology to accurately describe Opinion-
writing on the High Court.

Describing opinion-writing on the US. Supreme Court
In the common law tradition, the written opinion is an explanation (or set of

explanations) of the Court’s disposition of a particular case. These reasons are published

123

in the relevant reports, and are made available to lawmakers, lawyers, and the general
public. The Court’s decision as to the disposition or outcome of the case—what actually
happens to the litigants—is called a “judgment,” while the associated reasons are called
“opinions.” In the Supreme Court, members of the majority vote coalition all agree with
the disposition of the case at hand. Justices who disagree with the majority as to the
outcome of the case will author or join a dissenting opinion. These dissenting opinions
support an alternative disposition for the case, and they provide reasons for this
conclusion.

Members of this coalition typically join (or author) the opinion of the Court. The
opinion of the Court is also called the majority Opinion, as it is a set of reasons that a
majority of Justices have joined. Ifthe majority voting coalition fails to secure a majority
of j oiners, the resulting opinion is a plurality opinion and is called a “judgment of the
Court.” While the reasons contained in the opinion of the Court carry the weight of
precedent, the reasons in a judgment of the Court do not (Hochschild 2000).

Ifthey are dissatisfied with the reasons of the majority opinion, members of the
majority vote coalition may choose to author or join a concurring opinion. If a Justice
agrees with the reasons contained in the majority opinion and simply wishes to add
additional observations or reasons, she may author or join a regular concurrence. Justices
writing or joining a regular concurrence are also part of the majority opinion coalition. If
a Justice agrees only with the outcome but not with the majority’s reasons, that Justice
may write or join a special concurrence. A special concurrence is essentially an
alternative set of reasons to that presented by in the majority opinion. When a Justice

authors or joins a special concurrence, he is not part of the majority opinion coalition. In

124

this way, a special concurrence diminishes the majority coalition in a way that a regular
concurrence does not (Maveety 2003).
A lexicon for describing High Court opinion-writing

First, it is important to note that the common Australian use of the term “judgment”
is different from the American usage. Although some purists protest (Kirby 2006), most
legal commentators and practitioners use the terms “judgment” and “opinions”
interchangeably. Justice Kirby points out that the word “judgment” is used in the
Constitution to refer to the orders handed down by the Court, and he suggests using either
“opinion” or “reasons” to denote the written explanation of the Court’s orders (Kirby
2007b). The term “opinion” will be used here to refer to the written explanations of the
case disposition, which will be termed the “judgment” or “outcome.”

All separate opinion-writing behavior on the Supreme Court is described relative to
the majority opinion. Regular concurrences concur with the majority opinion, special
concurrences concur with the majority’s disposition of the case, and dissenting opinions
disagree with both. In the absence of a majority opinion tradition, these terms lose their
moorings. The classic seriatim-style disposition of a case is a collection of separate sets
of reasons—one set for each participating Justice. Dissenting Opinions express the
reasons for disagreement with the outcome favored by the majority. The use of the term
“concurring opinion” is much more problematic. In a sense, a unanimous decision
handed down in the traditional seriatim style is much like a collection of what Americans
call special concurrences.

The main difference, of course, is that the seriatim tradition leaves no central
explanation of reasons with which to concur. When Australians use the term

“concurrence,” they are actually referring to a separate Opinion that supports the majority

125

vote coalition’s disposition of the case. Because precedential weight is distributed among
all of the opinions in support of the majority, the concurring opinions essentially are the
Australian version of a majority opinion. The term “concurrence” is somewhat
problematic in this context, as it implies the existence of a single majority opinion. The
other descriptor commonly used in Australia—the term “separate opinion”—is even more
problematic. This term is over-broad, as it incorporates separate opinions in the minority
as well as the majority. The more precise alternative may be “separate opinion in support
of the majority vote coalition,” but this is far too cumbersome. As such, the term
“concurrence” will be used here to denote the separately-delivered opinions that,
together, make up the reasons for the majority’s disposition of the case.

The High Court does not always dispose of cases using the classic seriatim style.
Indeed, the intent of this research project is to explain the Court’s production of j oint
opinions in the absence of an American-style majority opinion tradition. For this reason,
some additional terminology is needed to describe the full range of opinion-writing
behavior of the High Court Justices. On occasion, the Court will publish a single set of
reasons, with which all the Justices agree. In the Supreme Court, this would be a
majority opinion. The use of that term, however, implies the unique precedential weight
that American majority opinions carry. In Australia, these Opinions do not carry
additional weight. To avoid this confusion, the Australian version will be called a
“single-opinion disposition.”

Often, smaller opinion coalitions will form on the High Court. Instead of a single-
opinion disposition, many cases will have one or more Opinions coauthored by a subset of

the panel. These are often called “joint judgments” in the Australian literature but, for

126

consistency, they are referred to as “joint opinions” here. These joint opinions can be in
support of the majority or in dissent. On occasion, Australian researchers refer to joint
Opinions in support of the majority as “joint majority opinions” or simply “majority
opinions.” As this chapter will demonstrate, this can lead to serious measurement and
equivalence problems. Instead, the term “joint opinion” will be used to refer to any
instance where more than one Justice coauthors an opinion. The single-opinion
disposition is a specific kind of joint opinion, and is included in analyses of joint opinions
unless otherwise specified.

Consensual Norms in the Literature

At its heart, a consensual norm in the judicial context is an institutional tradition of
deference on the part of the individual judge to the decisions of the majority. At the
height of the consensual norm in the US. Supreme Court, “[i]ndividual expression was
kept to a minimum and within a relatively consistent range” (Walker et al. 1988).
Essentially, this means that dissenting and specially concurring opinions were somewhat
unusual, even if private difference of opinion was not (Epstein et al. 2001). In the
American context, the decision to author a concurring opinion is seen as “strategic
devices that further short-term individual ends at the expense of institutional integrity”
(Schwartz 1992, 241). Researchers have documented the decline of the consensual norm
in the Supreme Court (Best 2000; Caldeira and Zorn 1998; Haynie 1992; Walker et al.
1988). This decline has long been lamented as being harmful to the institutional
legitimacy of the Supreme Court (Hand 1958; Pritchett 1954).

The reasons given for the decline are varied. Much of the blame is laid on the
shoulders of the Court’s leadership (Caldeira and Zorn 1998), Chief Justices Stone

(Danelski 1960; Walker et al. 1988) and Hughes (Haynie 1992) especially. Others

127

attribute the decrease in consensual behavior to practical considerations, and not to an
underlying consensual norm. The increase in discretionary jurisdiction created by the
Judges’ Bill of 1925 has been cited as a reason (Halpem and Vines 1977), along with the
subsequent decrease in routine cases (Wood et al. 1998b). Others point to an actual
increase in ideological disagreement among members of the Court, especially in the areas
of civil rights and civil liberties (Hurwitz and Lanier 2004). Even technological and
organizational advances in Court administration have been cited as a cause for the
proliferation of separate opinions (Martin and Soroka 2004).

Whatever the cause, the declining levels of consensual behavior are manifested in
the decreasing propensity of the Justices to deliver joint opinions with their colleagues. If
this is due to norm-related behavior, identifying a changing consensual norm on the
Supreme Court first requires an identification of what constitutes a deviation from
consensus. Often called a measure of “dissensus,” this can be little more than a count of
occurrences of non-j oining behavior (Caldeira and Zorn 1998). Most times, this amounts
to a calculation of the proportion of dissenting or concurring Opinions per opinion written
(Haynie 1992; Hurwitz and Lanier 2004; Walker et al. 1988).74

Clearly, this method measures only that level of dissensus that can be seen in the
final opinion tally (Atkins and Green 1976). It acts as a proxy for the actual level of
disagreement existing among the members of the Court. This can be problematic for
researchers interested in determining the root cause of the measured dissensus, and its
relationship to some more general consensual norm. In short, an increasing incidence of

separate opinions does not necessarily mean that a norm of consensus that once existed

 

74 Caldeira and Zorn (1998) do not standardize the dissents and concurrences, but argue that the results
would have been similar if they had.

128

has now weakened (Epstein et al. 2001). If changes in agenda composition are the root
cause of increased dissensus, it would be reasonable to believe that a norm of consensus
never existed in the first place. In that case, the increasingly contentious cases may have
brought disagreement to a Court that experienced little real disagreement in the past and,
thus, had no need of a consensual norm to discourage the expression of individual
opinion.

Several researchers have provided qualitative documentation of an active culture of
consensual norms in the pre-1940 Supreme Court (Schaefer 1966; Wood et al. 1998b).
Epstein, Segal and Spaeth (2001) address this problem directly through an empirical
investigation of Chief Justice Waite’s docket books. They utilize an independent
measure of dissensus: the votes of the Justices during the Court’s private conferences.
They compare these private votes to the reported votes that, of course, provide the public
face of unanimity during the Waite Court (1874-1888). While they find evidence that the
nondiscretionary caseload muted disagreement all around, they also find that “a norm of
consensus did, in all likelihood, exist” (Epstein et al. 2001, 376).

Caldeira and Zorn (1998) test the consensual norm hypothesis from a different
angle. They begin from the theoretical stance that, if a norm of consensus is the
underlying cause of shifting levels of public disagreement, this norm should similarly
impact both dissent rates and rates of concurring opinions. They analyze rates of dissent
and concurrence between 1800 and 1991. They find a cointegrated relationship between
rates of concurrence and dissent, with changes in the former following changes in the
latter. They argue that this specific type of relationship demonstrates that “a common

element underlies levels of both dissents and concurrences, and thus provides more direct

129

evidence for the existence and influence of consensual norms” (Caldeira and Zorn 1998,
888)

In this view, concurrences and dissents are both types of nonconsensual behavior.
In the Supreme Court, though, dissents and concurrences serve very different purposes.
While dissents express more complete dissatisfaction with the judicial outcome, they also
fit with the policy-oriented predictions of the attitudinal and strategic models of decision-
making behavior (Maveety 2003). These models focus on the policy importance of the
case outcome, and argue that doctrinal rules contained in the Opinions “merely rationalize
decisions; they are not the causes of them” (Segal and Spaeth 1993, 66). For this reason,
as Maveety explains, “to concur is to agree but to disagree, yet to disagree about
matters which judicial behaviori sm discounted as superfluous: the legal content of the
majority opinion” (Maveety 2003, 176). Concurrences can result in plurality opinions (or
“judgments of the Court”), undercutting the precedential value of the opinion without
tangible policy-related benefits (Davis and Reynolds 1974).

In this way, the attitudinal and strategic models would predict different causes for
dissents as compared to concurrences. For each, dissenting behavior is a reflection of
dissatisfaction with the substantive outcome of the case. Strategic theorists have
predicted that concurrences will be used as a negotiating tool (Murphy 1964), and
attitudinalists have speculated that concurrences are essentially throw-away votes cast in
cases involving less-strongly-held policy preferences (Segal and Spaeth 1993). Indeed,
both camps seem skeptical of their own post hoc attempts to explain concurrences, and
instead seem to coalesce around a norm-based explanation (Maveety 2003). Caldeira and

Zorn (1998) argue—implicitly, at least—that rates of these two distinct types of

130

nonconsensual behavior would not trend together unless an underlying consensual norm
were responsible. Investigating the possibility of consensual norms on the High Court
can provide insight into the aggregate patterns of separate opinions and joining behavior.
This is critical for the present research project, beginning as it does with the assertion that
joining behavior in a seriatim institution is something unusual and deserving of
explanation.

Consensual Norms in the Australian Context

Investigations of the norm of consensus on the US. Supreme Court usually begin
with the tenure of Chief Justice John Marshall. While this is appropriate for the kinds of
questions asked by observers of the Supreme Court, it starts fi'om a premise that is
inappropriate for courts operating under the British tradition of seriatim opinion
production. Marshall “eliminated the practice of seriatim opinions that allowed each
Justice to voice his individual views on every case,” believing that this would best protect
the institutional legitimacy of the Court (Walker et al. 1988, 362). The norm of
consensus that would result from a wholesale repudiation of the seriatim tradition, then,
would be one in which the expression of both dissenting and concurring opinions would
be discouraged. It is this sort of norm that is identified in the work of Caldeira and Zorn
(1998)

Given the Australian High Court’s adherence to the seriatim opinion tradition,
however, such a norm would be quite unexpected. In Australia, judges conform to a
much different norm. By definition, the seriatim opinion tradition tolerates concurring
opinions. In the context of this tradition, a norm of consensus would not likely include
pressure to join an opinion for the Court. However, this does not preclude a different sort

of consensual norm—one that values individual voices that agree with the disposition of

131

the case. Such a norm may encourage consensus by suppressing dissenting votes, while
treating separately-written Opinions in support of the vote consensus as consensual
behavior. Of course, this is of less immediate concern in the context of a model of
joining behavior on the High Court. Ifthe writing of concurring opinions is truly
suppressed by a consensual norm, the joining behavior on the High Court can be
explained in the same way as it is on the Supreme Court. Ifnot, then an alternative
model of j oining behavior will be necessary to explain why unconstrained judges choose
to join.
The Norm of Concurring Opinions

The simplest definition Of a concurring opinion is a separate opinion in favor of the
majority vote coalition’s judgment in the case. Such an opinion advocates for the same
case outcome as the majority of the judges on the bench. In contrast, a dissenting opinion
is one that advocates a case outcome different from the outcome of the majority. While
this is a relatively straightforward classification system in the context of the United
States, these concepts “are more than a little difficult to apply due to the seriatim
tradition followed by the Australian courts” (Lynch 2002a, 471). Indeed, a concurring
opinion in Australia often has no single majority opinion with which to concur. In other
words, Australian cases are disposed of with what would be called a plurality opinion in
the United States: a collection of opinions in support of a particular outcome, but with no
single opinion signed by a majority of the judges.

The individual responsibility for one’s own reasons is seen to trump the need for
the institution to produce a unified Opinion of the court (Coper 2002c). This follows
more closely the opinion of Thomas Jefferson on the matter, who notes that the seri atim

tradition

132

shews whether every judge has taken the trouble of understanding the case, of
investigating it minutely, and of forming an opinion for himself, instead of pinning
it on another’s sleeve. (Jefferson 1822, 1)

The result of this tradition is that judges frequently author—and are expected to author—
separate opinions, regardless of the degree to which they agree with the majority on the
disposition of the case.

In the United States, this practice is frowned upon. Researchers describe the task
of sorting out the mess of seriatim-style opinions as Herculean (Gerber and Park 1997).
They lament the lack of clarity in cases without a unified majority opinion, as well as the
danger they pose to the legitimacy of the Court as an institution (Davis and Reynolds
1974). In the High Court of Australia, however, this lack of clarity is not seen to be a
critical problem. While some legal practitioners continue to call for a more united front
(McMillan 2003 ), the practice is generally accepted. In Australia, “the law is not the
function of a majority opinion but instead of the highest common factor among the
opinions supporting the judgment” (Rosenberg and Williams 1997, 443).

Ever since the addition of the independent-minded Higgins and Isaacs to the High
Court in 1906, the decision-making process on the Court has been characterized by this
“rugged individualism” (Smyth and Narayan 2004, 404). Indeed, each judge is seen to
have a duty under the judicial oath to produce reasons for which he can be held
individually accountable (COper 2002c). While some Justices have occasionally
expressed an interest in creating more joint opinions where the judges are already
inclined to agree with one another (Gleeson 2004b; Mason 2003a), “there does not seem
to be even an informal policy against separate judgments” (Bagaric and McConvill 2005,

5).

133

Of the little empirical research there is on decision making in the High Court, much
of it concerns voting behavior as opposed to opinion-writing behavior (e. g., Blackshield
1972; Schubert 1969a). There is, though, some empirical evidence supporting the claim
that the institutional norms allow the writing of separate concurring Opinions. In a study
of joining behavior on the Latham Court (1935-1950), Smyth (2002b) reports that nearly
60% of cases reported were delivered in the seriatim style. Narayan and Smyth (2005)
explain that the seriatim tradition creates what they call a “low consensus equilibrium,” in
which there is a lack of consistent and significant promotion of j oint opinion production.

There are several ways to assess the level of support for this norm of concurring
opinions. One way, in keeping with the measurements used by Narayan and Smyth
(2005), is to look at concurring opinions as a percentage of all written opinions over time.
They do not count cases, but instead count opinions—individual statements of reasons, be
they in support of the majority or not. In the data analyzed in the present study, from

1965 to 2002 an average of 53% of all opinions written were separate opinions in support

of the majority disposition of the case.75 As Figure 6 illustrates, this number has
decreased slightly over time, with a distinct dip between 1981 and 1996. In contrast,
joint opinions in support of the majority represent 29% of all opinions written.

This measure, however, fails to account for the fewer opportunities for joint
majority opinions as compared to concurring opinions. Iffour Justices decide to join, the
resulting opinion is only counted as one joint opinion. Ifthe same judges decide to write
separately, this is counted as four separate opinions. This significantly undercounts joint

opinions—a problem which will be addressed at some length in the next section. Ifthe

 

5 Judgments written in cases drsposed of by a Single judge are removed from thrs analysrs, as they provrde
no opportunity for joining, concurring or dissenting behavior on the part of the judge.

134

Figure 6 - Concurring and Joint Opinions in Majority per 100 Reported, 1965-2002

100

 

804

 

 

 

604
U1
C
0
IE
0.
9 40«
E
“6 .
(D ’f “ ,"“.;“ ’1----
2 20“ _
g Concurrences
U .-..
g 0 .Jornt Oplnlons

 

19555710 '7'1 '73 '7'5 '7'7 is '8'1 '8'3 '0'5 '87 '59 '91 '93 '95 '57 '99 '01

frequency of concurring and joint opinions in the majority by case is substituted for the
opinion-based measure, the joint opinions are no longer undercounted. Figure 7 shows
the percentage of cases per year with at least one concurring opinion. On average, 70.1%
of cases have at least one concuning opinion filed. Additionally, Figure 7 shOws the
percentage of cases with a joint opinion in the majority. A joint opinion in support of the
majority can be penned by a coalition of as few as two Justices. A case can easily have a
mixture of j oint opinions and concurrences in support of the majority.

Taken together, these two accounts of consensual behavior indicate that, for the
period investigated, concurring opinions are not successfully discouraged on the High
Court. These opinions account for over half of the opinions filed, and are present in the
majority of cases heard by the Court. That is not to say, however, that joint opinions are
effectively discouraged, either. Clearly, the incidence of j oint opinions in the majority

has remained high through this period, and especially so since the early 19805.

135

Figure 7 - Percentage of Cases with Concurrences and Joint Opinions in Majority, 1965-2002

 

.....

100 .. 7 .

 

 

U)

8

8 __

.5 At Least One
in Concurrence
,8 20« _ _ - .

8 At least One

:5” Joint Opinion
0. 0

 

 

1955 '57'759 '71 '7th '75 '77 '79 '51 '53 '55 '57 '55 '51 '53 '55 '57 '55 '51

Of course, the evidence above is descriptive in nature, and represents only an
account of the observable incidence of such behavior. Determining whether this stems
from a norm is much more difficult. As the research on the US. Supreme Court
indicates, finding empirical evidence of an underlying norm accounting for observable
behavior is rarely a simple matter. The cause can be advanced by identifying external
confirmation of such a norm, or by failing to find evidence of independent causes for the
behavior. The task of identifying dynamism over time, however, will be reserved for
later in the chapter.

The self-conscious maintenance of the seriatim tradition is, in itself, evidence that
there is an institutional norm at work. Chief Justice Gleeson argues that the practice of
encouraging joint majority opinions “involves a formal discipline that has never been
accepted in the High Court” (Gleeson 2004b). Chief Justice Mason agrees, arguing that
shoehoming divergent legal perspectives into a single opinion risks compromising

judicial integrity (Mason 2003a). Justice Kirby attributes a similar sentiment to Justice

136

Kitto, who argued that, “on balance, the writing of individual judgments tends to produce
the better work” (Kitto, as quoted in Kirby 1998c, 29).

Still, the fact remains that several Justices, especially those in leadership positions,
have sought to facilitate the joint majority opinion. Chief Justice Griffith’s very early
leadership has been described as unity seeking. in the mold of John Marshall (McGinley
1987). More recently, Chief Justices have attempted to facilitate the production of j oint
opinions. Gleeson has instituted regular conferences that, while informal, “provide for
the exchange of views [and] the reaching of consensual opinions and judgments”
(Callinan 1999). Chief Justice Mason also made “a more concerted effort to achieve that
result, particularly by inviting one of our number to write an initial judgment or draft”
(Mason 2003a).

Even those Justices who would like to see more joint opinions are quick to defend
the seriatim tradition. Chief Justice Gleeson explains,

I welcome joint judgments when we can get them and I’m anxious to promote the

writing of joint judgments but not at the expense of people feeling obliged to

compromise or to join in what is not really an expression of their own opinion.
(Gleeson, CJ, quoted in Clark 2003, 10)

Indeed, observers of the Court have seen these attempts at encouraging joint opinions as
lackluster at best. According to Kildea, Australian courts “have not adopted, in any
methodical way, a practice that encourages joint judgments” (Kildea 2003, 61). It seems,
then, that a separate opinion norm may indeed exist on the High Court. Despite the
rhetoric encouraging joint opinions, the Justices themselves seem reluctant to
compromise individual ”accountability to facilitate them. Joint opinions seem welcome
when they happen to emerge, but a culture of encouraging negotiation and compromise

appears to be absent on the Court. While separate concurring opinions may not be openly

137

encouraged, they seem clearly not to be actively discouraged. As Justice Callinan noted
early in his High Court career, “no justice seeks to exert any pressures on the others”
(Callinan 1999).

In short, the seriatim tradition should result in particular patterns in concuning
behavior over time. Namely, we should expect that concurrence is frequent and that this
rate is relatively stable over time. Ifthis is the case, it provides support for the claim that
the seriatim tradition is a functioning norm on the Court that at least accommodates (and
at most encourages) the production of concuning opinions. The previous section
provides some confirmation that concurring opinions are produced at a relatively high
rate over time (see Figure 6 and Figure 7). The following section will address this in
more detail. First, however, it is necessary to consider the meaning of the phrase
“consensual norm” as we would expect it to apply to a court with an active seriatim
tradition. By evaluating the various existing approaches to the study of consensual norms
in the US. Supreme Court and in the High Court of Australia, an alternative measure of
concurrence and dissent emerges. By using this insight, it is possible to develop a
comprehensive analysis of patterns of concurrence and dissent that applies to the unique
institutional configuration of the High Court.

Dissent and the Consensual Norm

Identifying an underlying norm of consensus requires more than looking at the
patterns of opinion writing in support of the majority. Many of the approaches used by
American researchers to identify an underlying norm have utilized information about
both dissenting and concurring opinion production (e. g., Epstein et al. 2001; Haynie
1992). This certainly makes sense, as the most obvious breach of a consensual norm is to

dissent. In the context of the US. Supreme Court, the incidence of concurring and

138

dissenting opinions are both used as proxy measures for a failure to reach consensus. In
essence, “dissenting and concurring opinions and votes represent modes of conflict on the
Supreme Court” (Caldeira and Zorn 1998, 877).

In the seriatim opinion tradition, though, concurring opinions are the norm.
Certainly, some Chief Justices of the High Court have been inspired to pursue joint
majority opinions, usually in response to pressure from the legal community (Clark
2003). Unlike the Supreme Court, however, the High Court does not view concurring
opinions as an expression of conflict. The dissenting opinion in the High Court, though,
does represent a distinct break from the majority on the part of the individual dissenter.
For this reason, if we are to expect any sort of consensual norm in a court with a seriatim
opinion tradition, we would expect this norm to extend to dissenting opinions and not
concurrences.

Researchers have attempted to discern whether there has been a norm of consensus
in Australia. Mirroring the approach of Caldeira and Zorn (1998), Smyth and colleagues
have attempted to find evidence of an underlying consensual norm guiding the High
Court’s behavior over time. In an early version of the research, Smyth (2002c) applies a
cointegration analysis to aggregate-level data on High Court opinion production rates
from 1903-1975. Smyth looks for a cointegrated relationship between concurring
opinions and dissenting opinions on the High Court, but fails to find one. As such,
analysis fails to yield support for the hypothesis that the same underlying process drives
both concurring and dissenting behavior in Australia. Instead, Smyth concludes that “a
single consensual norm does not underlie decision making in the High Court” (Smyth

2002c,255)

139

Given what we know about the role of concurring opinions in Australia, however,
this is not terribly surprising. The seriatim opinion tradition is itself a norm, guiding the
High Court’s institutional view of concurring opinion production. Ifthis seriatim
tradition fails to discourage concurring opinions, it can do so in one of two ways. First,
the tradition may lump separate concurring opinions together with dissenting opinions,
classifying both as institutionally acceptable outlets for judicial expression. In this way,
the tradition would be tolerant of judicial dissensus. In such a situation, a norm of
consensus would clearly not exist.

Alternatively, the seri atim tradition may distinguish between concurring opinions
and dissenting opinions, accepting the former but discouraging the latter. Such a norm
can be considered a norm of consensus, but must be differentiated from the American
version by the way consensus is defined. In the Supreme Court since Marshall, the
concurring opinion has been seen as an expression of dissensus. This Australian version
of the consensual norm, a seriatim consensual norm, could include concurring opinions as
expressions of consensus.

In a more recent revisiting of the consensual norm question, Narayan and Smyth
(2005) find some support for a more understated version of the American-style
consensual norm in the High Court. In an analysis spanning 1904-2001, they reinterpret
the meaning of the consensual norm, this time looking for “some norm on the issue of
dissent in operation on the Court at most times,” with this norm prescribing “either high
or low dissent” (Narayan and Smyth 2005, 165). They are not looking for a seriatim
consensual norm as defined above, but instead seek to find a common norm underlying

the changes in both concurring and dissenting behavior.

140

They find some evidence for a cointegrated relationship between dissenting and
concurring opinions on the High Court. In other words, they find a long-memoried
relationship between changes in dissent rates and changes in rates of concurrence.
Because of the institutional differences between the High Court and the Supreme Court,
Narayan and Smyth (2005) use slightly different measures from those in Caldeira and
Zorn (1998). First, Narayan and Smyth decide to “treat the majority joint opinion in any
given case as the de facto opinion of the Court” (2005, 155). By joint majority opinion,
they mean any joint opinion written in support of the majority vote coalition. As such,
what they call a “majority joint opinion” is not usually signed by a majority of the
Justices. They concede that joint majority opinions in the High Court have a
substantially different connotation from the Supreme Court counterpart, but argue that
this “comes closest to representing the opinion of the Court with a system of seriatim
opinion writing” (2005, 155).

There is no added precedential weight given to joint majority opinions in the High
Court (Gleeson 2004b). Additionally, a substantial percentage of cases will have no joint
majority opinion with which a judge could concur, if he should so desire. What this
means, in a sense, is that the opinions classified as “concurring” are actually part of the
majority. Andrew Lynch gives the following example of this difficulty:

It has certainly not been uncommon for all seven members of the Court to deliver

their own individual opinions in firll. In such instances it is, to an extent, quite

meaningless to regard concurrences and dissents as being of the same ilk—
judgments which together stand ‘separate’ from a clearly identified and concerted

majority. Rather the collection of concurrences amounts to a majority and it is the
dissents alone which are on the outer. (Lynch 2003a)

This is not to say that the decision to measure concurrence this way is incorrect. Instead,

this definition of the concurring decision as a distinctly nonconsensual behavior may be

141

misleading in that it includes opinions written in situations similar to the one outlined
above.

The second place where the Narayan and Smyth (2005) piece diverges from
Caldeira and Zorn’s (1998) methodology is essentially a consequence of the first, and has
to do with the way opinions are counted. Previous research on consensual norms in the
Supreme Court typically used weighted concurrence and dissent rates, or the number of
opinions per 100 majority opinions (Haynie 1992; Walker et al. 1988). In other words,
the data consisted of the number of separate opinions per 100 decided cases. Caldeira
and Zorn (1998) use the raw number of opinions with concurrences and dissents. They
do not standardize their data by the number of cases per year, but they do consider the
concurrence and dissent rates on the case level.

Perhaps as a way to simplify the data collection process in light of the seriatim
opinion tradition in the High Court, Narayan and Smyth (2005) analyze the number of
concurring and dissenting opinions per 100 opinions. This differs from the
aforementioned studies because it does not consider the concurrence and dissent rates on
the case level. This works to overemphasize the variation in concurrence rates, weighting
the cases delivered seriatim more heavily than those delivered in joint opinion style.

Consider two types of unanimous cases in the High Court: the first type is
delivered seriatim, with seven different opinions all supporting the majority, the second
type is disposed of with an opinion signed by all five judges and two separate concurring
opinions. Next, imagine that there were fifty instances of each type in one year. The
Narayan and Smyth methodology weighs the seriatim cases two and one-third times more

heavily than those delivered in joint opinion style. The fifty seriatim cases yield seven

142

opinions each, for 350 total opinions. For this half of the agenda, one hundred percent of
opinions are concuning opinions. The fifty joint opinion style cases, on the other hand,
yield only 150 opinions total: 100 concurring opinions and 50 joint opinions. For this
half of the agenda, two-thirds of the opinions are concurring. Taken as a whole, the
Smyth and Narayan methodology would determine that 90% of opinions from this
hypothetical year are concurring opinions.

While this is indeed true, it fails to capture the phenomenon more relevant to the
study of consensual norms: the number of judge-level opinion-writing opportunities
leading to concurring Opinions. Using the above example, weihave seven judges with
one hundred opinion-writing opportunities each, for 700 total. In the fifty seriatim cases,
all opinion-writing opportunities resulted in concurring opinions, or 350 in all. In the
joint opinion cases, 250 of the opinion-writing opportunities resulted in joint opinions.
The remaining 100 led to concurring opinions. Total, this makes for 450 concurring
opinions out of 700 opinion-writing opportunities, or 64.3%.

The difference between 90% and 64.3% is substantial. Using the opinion-writing
opportunities standardization allows for a more precisely targeted measure. This
operationalization also helps to account for differences in panel sizes both case-by-case
and year-by-year. It is true that, when the panel system is working correctly, there will
be more opinion-writing opportunities in constitutional cases and other cases deemed
important by the Chief Justice (Brennan 2002a). This is because these important cases
will be heard by a firll bench, while other cases will be heard by a smaller panel. What is

of interest here, though, is the individual-level judge behavior. This behavior is more

143

accurately captured by investigating the choices judges make for each decisional
opportunity.

For this reason, the investigation of the consensual norm in the High Court will
focus on the opinion-writing opportunity, although the Narayan and Smyth (2005)
measure will be presented for purposes of comparison. This investigation is intended to
accomplish two goals. First, it will revisit the relationship between concurrence and
dissent in the High Court. Specifically, it will provide a replication of Smyth (2002c) and
Narayan and Smyth (2005). Second, it will investigate the longitudinal patterns of
concurrence and joining behavior. Taken together, these analyses will provide insight
into whether joining behavior is something that occurs in the High Court for reasons
other than the “consensual norm” that has been described in the literature.

An Underlying Norm of Consensus in Australia?

As the previous section explains, past research has attempted to find evidence of a
norm of consensus on the High Court (Narayan and Smyth 2005; Smyth 2002c). This
norm of consensus, if it exists, would have significant consequences for our evaluation of
the seriatim tradition in the Court. Indeed, such a finding would implicate a waning
tradition of separate opinion writing. Additionally, the presence of such a norm would
suggest that joining behavior is not something particulariy unusual given the institutional
context of the Court, as joining in the US. Supreme Court is not particularly unusual.

Caldeira and Zorn (1998) suggest a very specific operational definition of the consensual
norm in the judicial context. Because this consensual norm is something driving both
dissents and concurrences, they argue that finding a cointegrated relationship between
rates of concurrences and dissents over time “is good circumstantial evidence” of the

existence of the consensual norm (Caldeira and Zorn 1998, 881). As such, the opinion-

144

writing data from the High Court of Australia from 1965-2002 will be evaluated for such
evidence.

Stationarity and cointegration

A cointegrating relationship exists between two series of order 1(1) if the difference
between them yields an 1(0) series. In other words, two non-means stationary series are
cointegrated if a linear, means stationary combination of these variables exists. In this
case, the series share a linear attractor, and tend to trend together. If a shock occurs in
one series, the other exerts a gravitational pressure of sorts on it, pulling it back in line.
This provides an avenue for a very intuitive indicator of theoretical constructs like
institutional norms, as such a norm could be the origin of the trend in each series. Surely,
this is the application intended by those applying cointegration techniques in search of
consensual norms in judicial institutions (i.e., Caldeira and Zorn 1998; Smyth 2002c;
Narayan and Smyth 2005; 2007)}.

No pair of series can be cointegrated if either series is stationary—in other words,
both series must trend in order for them to be trending together. For a time series to have
second-order stationarity which is the customary assumption for normally distributed

data, the process must have a constant mean and variance over time (Maddala and Kim

1998).76 In mathematical notation for series y,, this can be represented as

EO’z) = 1507-1): -~ = EO’l-n) = 71

Vol) = Vtm) = = Vein) = oz

 

76 . . . . . . . . . . .
Thrs rs contrasted wrth first-order statronarrty, which requrres that the distribution of the series would be

unaffected by an arbitrary shift in time. This is not the customary assumption because it is “defined in
terms of the distribution” (Maddala and Kim 1998, 10).

145

COVO’t) = cor/m.» = = cor/em) = 72

Variance stationarity is fairly straightforward to diagnose and correct. While there are
tests for variance stationarity, simply looking at the data is a reasonable approach to
diagnosis (Granger and Newbold 1977). Ignoring it can lead to overparameterized
models (Milionis 2004), but correcting for variance nonstationarity can be as simple as
computing a log transformation for the series (Box and Cox 1964). Indeed, this is the
approach used by Narayan and Smith (2005).

Once the variance is stationary, the stationarity of the means can be assessed. A
means-stationary process is one in which the data exhibit no trend. Diagnosing means
stationarity is important in order to satisfy the assumptions of many analytical models,
and also as a first step in cointegration analysis (Zivot and Wang 2006). There are many
ways to accomplish this, the most famous being the Augmented Dickey-Fuller (ADF)
(Said and Dickey 1984) test (Elder and Kennedy 2001). Like other unit root tests (e. g.,
Phillips and Perron 1988), the null hypothesis of the ADF test is non-stationarity, or first-
order integration. More specifically, the ADF test is a test for random walk, random walk

with drift, or random walk with deterministic trend,
k
Ay, = A + [17+ 704-1 + Z 7,- AyH + 8,
i=1
where ,60 is a constant (representing drift), ,6 is the constant on time trend 1 (representing

deterministic trend), and y is the test statistic with a non-standard t distribution. This

augmented version of the original Dickey-Fuller (1979) test contains a lagged

146

endogenous variable to help ensure that there is no information left in the residuals of the
test equation.

The Elliot, Rothenberg and Stock (ERS) test (1996) is a modified version of this
type of autoregressive unit root test, improving on the difficulty the ADF test and other
unit root tests have in detecting alternative, non-unit root hypotheses (Zivot and Wang
2006). This test, also called the DF-GLS test, essentially uses a process of GLS
detrending to derive the test statistic more efficiently. The error term is less constrained,
and can be specified using different formulation. This allows for the removal of the
deterministic terms in the ADF test, giving it more power to distinguish between
persistent 1(0) series and those containing a unit root.

Stationarity tests, including the popular KPSS test (Kwiatkowski et al. 1992), begin
with a null hypothesis of stationarity. The KPSS test is a test of parameter constancy,
with the null hypothesis being that the parameter for trend is equal to zero. More
precisely, the KPSS is actually two different tests: a test for stationarity in level form
(Hu) and a test for stationarity with deterministic trend (Ht). This test is often used in
conjunction with unit root tests as a confirmatory step, as in Narayan and Smyth (2005).
Like the ERS test, the KPSS test was designed to more effi ciently distinguish the
persistent I(0) series from the 1(1) series.77

Identifying a unit root in a series is important, as analysis of two means non-
stationary series is likely to yield evidence of relationships that are actually spurious
(Granger and Newbold 1977; Yule 1926). The standard treatment of such a series is to

render them stationary through the process of first differencing. While this process can

 

77 . . . . . . .
For a drscussron of issues surroundrng unrt root dragnosrs, see Campbell and Perron (1991).

147

help us avoid the problems of spurious regression results, it eliminates information
contained in these series pertaining to their possible long-run relationship. Engle and
Granger (1987) identify an alternative which involves the identification and examination
of possible cointegration between the series.

There are several ways to identify cointegration between a pair of series. Engle and
Granger (1987) present a two-step method. This method involves regressing one series
on the other, then examining the residuals of this cointegrating regression for stationarity.
Ifthe residuals are stationary, the null hypothesis of no cointegration between the series
can be rejected. The Johansen and Juselius (1990) procedure is a pair of maximum
likelihood tests evaluating two different measures of cointegration.

Evaluating the High Court data for unit roots

Of course, a cointegrated relationship necessarily requires that both series have a
long run pattern in the first place. If either series is 1(0), there is no cointegrated
relationship by definition. For this reason, the first step in any such analysis is to identify
the order of integration for the relevant series. Before testing the means stationarity of
the two series, it is critical to identify and correct for possible variance nonstationarity.
Figure 7Error! Reference source not found. demonstrates that the series of concurring
opinions is likely to be variance nonstationary. To correct for this, it is customary to
apply a natural log transformation to the data (Tabachnick and Fidell 2001 ).

A series of unit root tests, as described above, can now be run on the transformed
series. To be sure, the results for the KPSS stationarity tests for each series are not
entirely consistent with the results of the various unit root tests (see Table 6). This is not
a terribly unusual result. As one researcher notes, “much of the problem with the unit

root literature arises from the belief that estimates or test statistics would provide the

148

answer in themselves” (Smith 2000, 200-201). There is, however, a good deal of

evidence that the log transformation of the concurring opinion series contains a unit root.

Figure 8 - Concurring and Dissenting Opinions per Opinion-Writing Opportunity

100

 

80-

 

 

 

60«
m .
g 40‘ ‘ . .
O .
”<5 -
81
.9 20‘ _
E, Concurrences
U
ff 0 Joint Opinions

 

V V”
U

1965 '57 '5'9 '7'1 '7'3 '7'5 '77 '79 '91 '93 '9'5 '97 'u '91 '93 '95 737 99 '01
The dissenting opinion series, however, does not appear to have a unit root.

The results for the unit root tests of the log-transformed Narayan and Smyth (2005)
measure yield similar results (see Table 7). Narayan and Smyth calculate the number of
concurring and dissenting opinions per one hundred written opinions. As was discussed
earlier in the chapter, this measure tends to weight separate opinions much more heavily.
The present analysis finds that, like the measure weighted by opinion-writing
opportunity, the measure of concurring opinions contains a unit root. This is also
consistent with the findings in Narayan and Smyth’s analysis (2005). The present results
differ, however, when it comes to the evidence of the means stationarity of the measure
of dissenting opinions as displayed in Table 7.

There are several possible explanations for this difference. Certainly, the present

study uses a smaller sample than the Narayan and Smyth (2005) piece, which traces

149

separate opinions from 1904-2001. The earlier piece by Smyth (2002c) analyzed

separate opinions from 1903-1975 and, while it failed to find a cointegrated relationship,

Table 6 - Unit Root Tests on Log-Transformed Separate Opinions per OWO

i)l\'.\t‘lll.\ ('tiiit‘iil‘icrrt‘os ('i‘iiit'nl \riliit' l/' ("1

 

Dickey-Fuller Tests . ..
Constant, No Trend -3.90* -1.94 -2.94

Constant, Trend -3.92* -1.96 -3.53
Augmented Dickey-Fuller (ADF) Tests

No Trend -3.68* -1.47 -2.94
With Trend -3.59* -1.46 -3.53
(# of Lags) 0 0

Elliot, Rothenberg and Stock (ERS) Tests

DF-GLS No Trend -3.96* (0) -1.60 (0) -1.95
DF-GLS With Trend -4.03* (0) -2.18 (0) -3. 19
Kwiatkowski et al. (KPSS) Tests

Lags =0 0.17 0.38 0.15
Lags=2 0.11* 0.17 0.15
Lags=4 0.10* 0.13* 0.15
Lags = 8 009* 0.13* 0.15
Mean -2. 17 -1.22

Standard Deviation 0.19 0.39
N 38 38

 

* Indicates rejection of the null hypothesis. DF, ADF and ERS critical values are based on
MacKinnon (1991). Optimal lags for ADF selected on the basis of Schwartz’s Information Criterion.
Optimal lags for ERS based on Ng-Perron (1995) sequential t-test. KPSS results are eta(,u) statistics.

 

 

 

it did find evidence that both the concuning and dissenting opinion series contained a
unit root. The sample in the present analysis is 38 years, which means that there are
fewer time points analyzed in the unit root tests. Shorter time spans have been
demonstrated to result in decreased power in traditional unit root tests, making it harder

for them to identify correctly a stationary series (Campbell and Perron 1991). The

150

Table 7 - Unit Root Tests of Log-Transformed Separate Opinions per 100 Opinions

 

l)l\\t‘lll\ ('rint'in'it'nt't's (‘i‘itltxil \ LllllL‘ l/’ 1'51

Dickey-Fuller Tests

Constant, No Trend -3. 17* -2. 13 -2.94
Constant, Trend -4.43* -2.35 -3.53
Augmented Dickey-Fuller (ADF) Tests

No Trend -2.15* -1.71 -2.94
With Trend -2.33* -1.80 -3.53
(# of Lags) 3 0

Elliot, Rothenberg and Stock (ERS) Tests

DF-GLS No Trend -1.76 (2) -1.80 (0) -1.95
DF-GLS With Trend —4.55* (1) -2.54 (0) -3.19
Kwiatkowski et al. (KPSS) Tests

Lags = O 0.17 034* 0.15
Lags=2 011* 017* 0.15
Lags=4 010* 013* 0.15
Lags = 8 009* 013* 0.15
Mean -1.65 -0.70

Standard Deviation 0.22 0.23

N 38 38

 

* Indicates rejection of the null hypothesis. DF, ADF and ERS critical values are based on
MacKinnon (1991). Optimal lags for ADF selected on the basis of Schwartz’s Information Criterion.
Optimal lags for ERS based on Ng-Perron (1995) sequential t-test. KPSS results are etan) statistics.

 

 

 

decreased power leads to a lessened ability to correctly identify persistent non unit root
series (Elliot et al. 1996).

In the present situation, however, the question is whether the series representing
dissent has been mistakenly diagnosed as a non-unit root series. Whatever the reason, the
failure to find evidence of a unit root in the dissenting opinion series does not mark the
end of the search for a cointegrated relationship. Instead, the series can be tested for a
fiactionally cointegrated relationship by loosening the assumption that the series must be

1(0) of 1(1).

151

Testing for fractional cointegration

A means-stationary process is one in which the data exhibit no trend. Diagnosing
means stationarity is important in order to satisfy the assumptions of many analytical
models, and also as a first step in cointegration analysis (Zivot and Wang 2003). Many
researchers argue, however, that series compiled through the aggregation of data created
through disparate individual-level data generating processes are likely to be fractionally
integrated (Box-Steffensmeier and Tomlinson 2000; Granger 1980; Lebo et al. 2000).
Instead of the series being characterized by the traditional “knife-edged” classification of
1(0) or 1(1), the order of integration is allowed to vary. The fractionally integrated series
is said to be I(d), such that 0<d<1. Fractionally integrated series are long-memoried
processes, as they “exhibit statistically significant correlations across time lags” (Clarke
and Lebo 2003b, 288).

Traditional cointegration tests assume that the series in question both contain unit
roots, and that error correction term derived from the cointegrating regression is means-
stationary (J ohansen and Juselius 1990). These restrictions, though, are not required
under the Granger representation theorem (Bollerslev and Baillie 1994). Just as the
concept of integration can be generalized to fractional integration, “[e]xtending the
methodology of cointegration to include fractional cases enables one to generalize the
concept of an error correction mechanism” (Clarke and Lebo 2003b, 295).

In other words, the requirement that the error correction mechanism be 1(0) can also
be relaxed, allowing for an error correction term of order I(d). Ifthe error correction term
is significant in the cointegrating regression and has an order of integration less than

either of the original series, the original series may be appropriately described as

152

fractionally cointegrated. In this system, “the effect of a shock declines at a slower rate”
than the classical cointegration paradigm allows (Bollerslev and Baillie 1994, 742).

The presence of a fractionally cointegrated relationship, which is a generalization
of the cointegration procedure used by Caldeira and Zorn (1998), requires that some
preconditions be met. First, each of the individual series must be at least fractionally

integrated (i.e., I(d) such that 0 < d <1). The rates of concurrence and dissent rates for

each of the four courts, then, must be evaluated for means-stationarity.78 Then, if both
series are means-nonstationary, a cointegrating regression can be estimated to determine
whether or not the series are cointegrated. Ifthe series are found to be cointegrated, an
error correction model can be specified to estimate the long- and short-run dynamics of
the system.

As one researcher notes, “much of the problem with the unit root literature arises
from the belief that estimates or test statistics would provide the answer in themselves”
(Smith 2000, 200-201). One solution to this problem is to estimate the value of d
directly. In other words, “instead of running multiple tests and looking for patterns
suggesting stationarity, one can instead rely upon the point estimates of d and the
associated t-ratios” (Box-Steffensmeier and Tomlinson 2000, 66). These parameters are
estimated using Robinson’s (1995) semi-parametric estimator, and they are reported in

the bottom half of Table 8. The results show evidence that, while dissent rates are not

 

78 . . . . . . .

In order for a time series to have second-order stationarity which is the customary assumption for
normally distributed data, the process must have a constant mean and variance over time (Maddala and
Kim 1998). Ignoring variance nonstationarity can lead to overparameterized models (Milionis 2004), but
correcting for it can be as simple as computing a log transformation for the series (Box and Cox 1964). As
such, each of the series here are log—transformed.

153

Table 8 — Unit Root Tests and Robinson's Gaussian Semiparametric Estimates of d (OWO Data)

 

 

 

 

 

 

Concurrences Dissents
KPSS (1992)
l = O 0.41 * 0,14
l = 2 0.18 * 0.11
1 = 4 0.13 0.10
I = 8 0.11 0.10
ERS (1996)
Trend -1.46(3) * -2.08(5)
No Trend -2.02(0) -3.92(]) *
Robinson’s d. (1995)1
Est. 61+ 1 0.73 0.31
5.13. (0.12) (0.12)
Ho: d= O -2.23 * -5.69
H0: d= 1 6.02 * 2.56 *
* Indicates rejection of the null hypothesis at p < .05 level. ERS critical values are based on
MacKinnon (1991) and optimal lags for ERS (indicated in parentheses) are based on Ng-Perron
(Ng and Perron 1995) sequential t-test. KPSS results are eta(u) statistics.
'1' Due to the fact that the parameter space of d is constrained such that -l .5<d<0.5, the parameter
estimates were calculated on the first-differenced series. As such, the results reported for the
estimate of d have been adjusted by adding one. The t-ratios associated with the null hypotheses
reflect the probability that the d of the differenced series is equal to 0 or 1.

 

 

1(1), each series is significantly different both from 1(0) and 1(1). In other words,
concurrence and dissent rates are not stationary, but instead are fractionally integrated.

The next step is to test concurrence and dissent rates for cointegration. The classic
method, as employed by both Caldeira and Zorn (1998) and Narayan and Smyth (2005) is
to use cointegration tests such as Engle-Granger (1987) or Johansen-Juselius (1990).
Because we have determined that the series of interest are all significantly different from
1(1), however, a fractional cointegration method is more appropriate. Demonstrating the
presence of a fractionally cointegrating relationship is as simple as demonstrating that
“the residuals produced by regressing one fractionally cointegrated series on another one
are of a lower order of integration than are either of the fractionally integrated series”
(Clarke and Lebo 2003a, 294).

The results for the cointegrating regression of dissent rates on rates of concurrence

in Australia are presented in Table 9. Here, the estimate of d for the residuals is 0.72,

154

which makes the order of integration higher than that of the rate of dissent, which is
estimated to be an I(0.31) series (see Table 8). As such, concurrences and dissents are
not fractionally cointegrated. Through these analyses, no evidence has been found to
support the hypothesis that rates of dissent and concurrence share a common linear

attractor.

Table 9 - Cointegrating Regression of Dissents on Concurrences

 

Independent \ :Irinlrles (‘nel'l'ieient Std. l'ir'r'or'

Constant -0.386 (0.171) *
Percent of Cases with a Dissent 0.035 (0.185)
Log-transformed

 

Rob. (l995)r for the Error Term

Est. (1+ 1 0.72 Ho: d= O -2.31 *

SE. (.012) Ho: d= l 5.94 *

‘1' Due to the fact that the parameter space of d is constrained such that -1.5<d<0.5, the parameter

estimates were calculated on the first-differenced series. As such, the results reported for the estimate

of d have been adjusted by adding one. The t-ratios associated with the null hypotheses reflect the
robability that the d of the differenced series is equal to 0 or 1.

 

 

 

 

This result, while contrary to the findings of Narayan and Smyth (2005), is more in
keeping with the expectations we would have given the High Court’s explicit adherence
to the seriatim opinion tradition. Narayan and Smyth were left to describe this
relationship as evidence of a “low consensus equilibrium,” presumably different from the
high consensual norm of the US. Supreme Court only inasmuch as it exists in a court
with a seriatim opinion tradition (Narayan and Smyth 2005, 148). As they explain,

The low consensus equilibrium in the High Court is a manifestation of the

prevalence of seriatim opinion writing and general absence of institutions designed

to build consensus, while the high consensus equilibrium on the US Supreme

Court, at least up to the 19403, reflected institutional practices such as the opinion

of the Court and formal conferencing arrangements. (Narayan and Smyth 2005,
165)

While they argue that this “makes sense from a comparative perspective,” they do

not explain the mechanism by which we would expect concurrences and dissents in a

155

seriatim court to have a common underlying cause. It makes more sense, surely, to
expect a consensual norm to impact only dissents. A seriatim consensual norm, as the
term is used earlier in this chapter, refers to an institutional norm of consensus that
considers concurrence an acceptable expression of consensus. Unlike the way the
consensual norm is conceptualized in the US. Supreme Court research (and the research
by Narayan and Smyth), the seriatim consensual norm would, by definition, have
different mechanisms driving concurring and dissenting behavior.

The analysis in this chapter does not prove the existence of a seriatim consensual
norm. As difficult as the traditional consensual norm is to identify, a seriatim consensual
norm is even harder to identify empirically. Because this norm would treat concurrences
and dissents differently, there is no obvious factor with which the series of concurring
Opinions would be cointegrated.

This analysis does, however, provide evidence that the seriatim tradition is not
malfunctioning in Australia. While there is fluctuation in concuning and dissenting
opinion behavior over time, there is no evidence in this analysis to suggest that they are
driven by a single common factor. In itself, this is an important finding. While the critics
of the seriatim tradition in Australia may lament it, it appears that the High Court has not
yet abandoned the rugged individualism in opinion writing that has characterized the
institution for most of its history. That being the case, the phenomenon of the joint

opinion on the High Court remains to be explained.

156

CHAPTER SDC THE PRODUCTION OF JOINT OPINIONS

Courts of final appeal have diflerent methods for the preparation and production of
opinions. It is an issue of interest to the profession, law teachers and students, and
judges of other courts.

—Chief Justice Murray Gleeson (2004a)

As the previous chapter demonstrates, there is little evidence that an underlying
consensual norm influences rates of dissenting and separately written concurring opinions
on the High Court Of Australia. Indeed, this finding is in keeping with the High Court’s
institutional tradition of using a seriatim Opinion-writing style. The seriatim tradition,
when functioning properly, creates a norm of separate Opinion writing. In the previous
chapter, no evidence is found which might suggest rates Of concurrences and dissents are
driven by a common factor. What the previous chapter cannot tell us is why, in a court
with a seriatim Opinion tradition, judges would sacrifice the ideological flexibility
afforded by such a tradition in order to coauthor joint opinions.

This chapter develops a set Of conceptual hypotheses about joining behavior on the
High Court. As Chapter 4 illustrates, theories to explain joining behavior in the US.
Supreme Court cannot be transplanted wholesale to the High Court. The hypotheses
presented here are derived from many relevant judicial behavior theories, but adapted for
the unique traditions Of this institution.

Joining Behavior and Case-Level Characteristics

By joining behavior at the case level, many important factors can be isolated.
Among these are characteristics of the decision making environment, characteristics of
the case itself, the role Of other government actors, the litigation history, and

characteristics of the case outcome. Each of these is addressed in turn below.

157

Characteristics of the Decision Making Environment

Several characteristics Of the decision making environment may influence joining
behavior on the Court. Small group theory suggests that increased workload pressure
may influence a judge’s decision to join in Opinion (Brenner 2003). Another important
small group explanation Of j oining behavior is the leadership Of the Chief Justice (Ulmer
1971). Finally, attitudinal theory suggests that judges will take actions that are in line
with their policy preferences (Segal and Spaeth 1993). As such, judges are expected to
join only when their policy preferences are reflected in the Opinion. These theories, while
developed in the institutional context of the Supreme Court, can be used fruitfully to
establish expectations Of joining behavior in the High Court.

Judicial Workload

Among the challenges of being a judge is the pressure of a heavy workload.
Certainly, Justices on the Supreme Court and the Australian High Court alike have
registered complaints about excessive workload burdens (Baum 1997; Craske and Gibbs
2002). Justice Michael McHugh has noted that “the workload Of the High Court Justices
borders on oppressive” (McHugh 2005, 3). The heavy workload of the US. Supreme
Court has led researchers to question the ability of the Justices to adequately address the
growing number Of petitions it must dispose Of (Jucewicz and Baum 1990; O'Brien
1985)

While the High Court of Australia deals with a smaller number of agenda-setting
decisions, jurisdictional restrictions put in place by the Australian Parliament force
judicial review cases to bypass the Federal court, “thereby imposing an oppressive and
inappropriate burden on the High Court” (Mason 2003a, 7). In addition, the High Court

Justices handle the agenda-setting decisions on their own (Horton 2002). This is in stark

158

contrast to greater degree of discretionary power delegated to the US. Supreme clerks
(Duxbury 2001).

One of the most significant drains on the High Court Justices is the hearing and
disposal of the increasing number Of special leave petitions (Gleeson 2004b). The
addition Of the video link system in 1987 allowed the Justices to begin hearing
nationwide petitions without the need tO travel (Jackson 1996). While this made the
process more efficient, these petitions are still a significant drain on the Court’s time.
The Court deals with fewer agenda-setting decisions than the US. Supreme Court does

(Mason 2003 a), but the High Court has a tradition of hearing oral arguments in virtually

all Of these special leave petitions.79

The consideration and disposition of applications for special leave to appeal is a
duty that the Justices do not take lightly (McHugh 2005). Typically, special leave
petitions are heard by a panel of three Justices. This is in stark contrast to the American
tradition of dealing with petitions for certiorari through written submissions. Special
leave hearings allow the judge to deal directly with the litigants, “without the gO-between
of the associate, without the risk Of documents not reaching the Judge, of being
summarised in some erroneous way, or of crucial information simply being buried in the
briefs or missing entirely from the submissions” (Horton 2002, 120).

It has sometimes been argued that a heavy workload encourages individual judges
to join the Opinions of their colleagues, thus producing a higher incidence of j Oint

Opinions (Brenner 2003). As the hypothesis goes, judges can save time by assenting to

 

7 . . . . . .

9 Thrs tradition has been changed by a 2005 revrsron of the High Court Rules. The new rules allow the
Justices to consider certain types of special leave petitions “on the papers” instead Of in oral arguments.
See (Gleeson 2005).

159

the opinion of a colleague instead of preparing a separate Opinion (Posner 1995). There
is, however, at least as much reason to expect that an increase in workload might produce
the opposite results. Justices on the US. Supreme Court have noted that the production
of separate opinions can be a natural result of time pressures (Kirman 1995). While the
time taken tO actually research and produce an individual Opinion may be substantial, it
may pale in comparison to what is necessary to negotiate an Opinion acceptable to
multiple judges. As Baum notes, “workload concerns might actually encourage the
writing Of separate opinions and move the Court fiirther from unanimity, because it is
time-consuming to seek consensus (1997, 47).

From the perspective Of the Opinion author, “circulating a new draft in order to
accommodate one’s colleagues requires an investment Of time and energy” (W ahlbeck et
al. 1998, 301). The pressures of a heavy workload can stifle the maturing of collective
thought on a particular case (Davis and Reynolds 1974). Australian legal scholars seem
to agree that workload pressures tend to decrease the number of j oint opinions. High
Court researcher Michael Coper contrasts the Opinion-writing behavior on two very
important cases, one written in a short time and the other over the span of neariy a year:

The seven individual (and long) judgments in the Tasmanian Dam Case (1983)

appear to have been the product of a shortage of time to explore the potential for

agreement. By contrast, in Cole v. Whitfield (1988), the eleven months between
argument and decision no doubt facilitated the production of a single, unanimous,

joint judgment. (COper 2002c, 368)

Because of the time necessary to amass collegial support for joint Opinions - especially in
the absence of a majority opinion tradition — increased workload pressures is likely to

hinder the production Of j oint opinions on the High Court of Australia.

H mothesis 1.' As workload pressures increase, joint opinion production will
decrease.

160

The most obvious workload-related pressure is the actual amount of time remaining
in the term. Researchers have often used the end of the term as a deadline of large
importance to appellate justices (Maltzman and Wahlbeck 2005). As the end Of the term
approaches, the judges may likely feel significant pressure to dispose of the remaining
cases (W ahlbeck et al. 1998). Research on US. Supreme Court opinion assignment finds
that organizational and efficiency concerns are increasingly important as the end of the
term approaches (Maltzman and Wahlbeck 2004). This suggests:

Hypothesis 1 a: When a case is heard near the end of the term, joining behavior will
be less likely.

Additionally, the end of a natural court can have the same effect. When a High
Court Justice leaves the Bench, it is customary that any undelivered Opinions remain
undelivered (Williams 2002). The 1977 amendment to section 72 Of the Australian
Constitution, which provides for a mandatory retirement age of 70, makes the departure
of a member of the Bench slightly more predictable. The first Justice to retire at the
mandatory retirement age was Justice McHugh in 2005. Even still, the impending
retirement of a Justice is generally known by his or her colleagues.

Hypothesis 1b: When there is a retirement date approaching, joining behavior will
be less likely.

The time pressures that may hinder the production of joint Opinions are a result not
only of the number Of months between hearing and Opinion, but also Of the amount work
to be accomplished during that time. On the High Court of Australia, the vast majority of
judicial work consists of the disposition Of cases and petitions for special leave to appeal
(Craske and Gibbs 2002). Generally, increases in workload demands result in pressure
on the judges themselves, as the Justices remain reluctant to rely on their associates to

assist in opinion-writing duties (Leigh 2000).

161

H wothesis Ic: When the number of cases heard in a term is higher, joining
behavior will be less likely.

Leadership of the Chief

Another important characteristic of the decision-making environment is the
leadership of the Chief Justice. Variations in the ability of a Chief Justice to lead the way
in forging consensus is an important factor in determining levels of consensus behavior
on the US. Supreme Court (Haynie 1992). While there is some disagreement on the
issue (Best 2000), it is clear that many judges and scholars believe that the leadership
abilities of a Chief Justice are important to understanding the character of the institution
during that judge’s tenure. As Supreme Court Justice Charles Evans Hughes once noted,
“in a small body of able men with equal authority in the making of decisions, it is evident
that his actual influence will depend upon the strength of his character and the
demonstration of his ability in the intimate relations of the judges” (Hughes 1928, 57).
The differences in the success of various Chief Justices on the High Court in the first half
of the twentieth century have also been shown to correlate with levels of dissensus
behavior (Smyth 2003b). Thus,

ijothesis 2: Joining behavior will be related to the leadership of the Chief
Justice.

One of the recurring themes circling the question Of leadership styles is the practice
of conferencing. The US. Supreme Court famously conducts formal, private conferences
on a regular basis. The Justices use these conferences to discuss cases with each other,
usually for the first time (Powell 2004). It is in this atmosphere that ideas are honed and
tentative opinion coalitions are forged. Because of the majority Opinion tradition on the
US. Supreme Court, the need for conferences is acute (W ahlbeck et al. 1998). Around

the turn of the century, conferences were supplemented with the circulation of opinions

162

after conference and “the system of laborious negotiation over substance and language”
needed to facilitate this collaboration (Kirman 1995, 2102). Despite the fact that there is
no institutional requirement for these conferences, the judges maintain an extraordinarily
consistent meeting schedule (Rehnquist 2004).

The extent to which such an institution exists in Australia is unclear in the literature
(F ricke 1986). The New South Wales Court of Appeals maintains a strategy for
encouraging joint opinions, and one of the dimensions of this strategy is a system of
conferencing and equitable “primary d_r ” assignment (Kirby 2005c). Chief Justice
Murray Gleeson has made some effort to encourage joint opinions on the High Court, and
he sees his role as being an important variable in their proliferation (Clark 2003).
According to Justice Kirby, though, formal techniques for reducing excessive separate
opinions “have never been features of the High Court of Australia” (Kirby 20050, 3).
Indeed, there is nothing in the Australian High Court akin to the systematic conferencing
process of the US. Supreme Court. That said, several Chief Justices have attempted to
institute some sort of informal conference (Smyth 2003b). At the beginning of his tenure,
Chief Justice Gleeson had instituted a somewhat structured conference following oral
arguments. In a 1999 speech, Justice Callinan described these conferences:

The meetings are very useful because of the opportunity they provide for the

exchange of views, the reaching of consensual opinions and judgments, when

possible, and the identification, and usually, in consequence, the narrowing of
difference between the Justices. (Callinan 1999, 3)

Barwick tried a similar strategy, but was far less successful (Smyth 2003b).
The categorization of Chief Justices as “successful” or “unsuccessful” in this regard
will have a lot to do with their ability to promote, as Kirby describes, a “culture of mutual

respect, reinforced by happy social intercourse” (Kirby 2005c, 3). The production of

163

joint opinions “is affected by the extent to which the Chief Justice promotes
arrangements that will encourage the writing of j oint judgments” (Gleeson as quoted by
Clark 2003, 10). As Chapter 3 describes, the modern Chiefs have been very different
from each other in their ability to promote a harmonious working environment. As such,
we might expect joint opinions to flourish under some Chiefs, and to languish under
others. Because of the importance of the conferences as an opportunity for the
emergence of cooperative behavior,

H mathesis 2a: Joining behavior will be higher during the tenure of Chief Justices
who were able successfiilly to implement a system of regular conferences.

Being elevated to the position of Chief Justice is a great honor, but it also comes
with great administrative responsibility. Research on the US. Supreme Court indicates
that new Chief Justices have a distinct period of acclimation to their new jobs (Lanier and
Wood 2001). An analysis of the Canadian Supreme Court also finds evidence that new
Chiefs are less likely to author majority Opinions during the beginning of their tenure
(McCormick 1994; but see Wetstein and Ostberg 2005).

There is good reason to expect the opposite result in the High Court. While there
likely is a learning curve, many Chief Justices come to the position intending to facilitate
the production of j oint opinions (Clark 2003; Mason 2003c). For better or worse, many
of the Chief Justices abandon attempts to organize the formal conferences and regular
communication necessary to maintain a steady flow of j oint opinions (Smyth and
Narayan 2004; McMillan 2003). As such,

Hypothesis 2b: During the first year of a new C hief Justice ’5 tenure, joining
behavior will be more likely.

164

The ideological makeup of the panel

There are a couple of reasons to suspect that liberal and conservative Justices will
join opinions at different rates. First, we may hypothesize that liberal and conservative
Justices differ in their propensity to join. There is little systematic evidence to support
this assumption. The second reason, however, is a more tenable one. With the exception
of the Gleeson Court, the modern High Court has never had a majority of judges
appointed by the Labor Party. In other words, the appointees of Liberal-National
Coalition prime ministers have constituted the majority on the Court during the tenures of
Barwick, Gibbs, Mason and Brennan. The attitudinal model suggests that, in the absence
of a majority opinion tradition, judges will join only when their policy preferences are
adequately represented by a colleague’s opinion (Segal and Spaeth 2002). When there
are more ideologically compatible judges to choose from, joining behavior will be more
likely.

Hypothesis 3: The likelihood of joining behavior will depend on the ideological
makeup of the panel.

The logic described above would predict that, ceteris paribus, conservative panels
would be more likely than liberal panels to dispose of cases with a single Opinion (with
the exception of the Gleeson Court). The impact of panel ideology on other types of j oint
behavior is more difficult to predict. Ifliberal (or conservative) judges are more likely to
engage in joining behavior, liberal (or conservative) panels may produce more joint
opinions because of this tendency.

Hypothesis 3a: Single-opinion dispositions will be less likely on panels with higher
mean liberalism scores than on panels that are more conservative.

The ideological diversity Of the panel is an important consideration in predicting

joining behavior. Ideologically homogenous panels will be more likely to engage in both

165

types of j oining behavior. The heterogeneity of the majority coalition in the Supreme
Court has been linked to greater accommodation by the opinion writer (W ahlbeck et al.
1998), but it is unlikely to have this effect in the High Court. In the absence of a majority
opinion tradition, cases that would have required greater accommodation to win over a
majority of judges would simply be disposed of by several separate Opinions. On
homogenous panels, and individual judge is more likely to find his preferences supported
in the opinions of his colleagues. This is especially critical in the production of single-
opinion dispositions, especially in complex or salient cases.

H )pothesis 3b: Joining behavior will be less likely on panels with high diversity of
liberalism scores than on panels with less diversity of scores.

From an attitudinal perspective, the notion that ideological agreement would
predict joining behavior seems obvious. Certainly, the voting behavior Of judges on the
Supreme Court has been quite satisfactorily explained in this way (Segal and Spaeth
2002). Similar research has shown that attitudinal explanations of voting behavior can
help to explain decision-making behavior in the highest courts Of other countries,
including Australia (Ostberg et al. 2002; Songer and Johnson 2002; Schubert 1969a;
Smyth 2005b).

Much research into attitudinal explanations of judicial voting behavior has used
scales other dimensional analysis techniques to measure attitudes. The early jurimetrics
literature followed Glendon Schubert’s lead and used cumulative scaling or “scalogram”
analysis (Tate 1983). Criticism of this methodology points to serious problems of
properly ranking the cases and the difficulties imposed by seemingly inconsistent votes

(Tanenhaus 1966). The more sophisticated multidimensional scaling techniques have

166

bypassed many of these difficulties, and methods evolved to apply scaling techniques
even to courts that sat in panels (Mc Iver 1976).

One of the major limitations to using scaling techniques is the need to use the
available vote data in order to produce ideological scales. While the scales may prove
useful in helping to identify patterns of agreement, the measures cannot be used in a

model to explain the voting behavior from which they were derived. For this purpose, it

is necessary to collect information external to the data contained in the voting records. 80

Hypothesis 4: The relationship between a Justice ’s ideological position and the
position of the rest of the bench will affect the likelihood of his joining.

In the Supreme Court, majority opinion authors must increase their accommodation
efforts when the majority coalition is more diverse. Evidence from the US. Courts of
Appeals also suggests that a judge is more likely to author a separate opinion—either a
concurrence or a dissent—when he is ideologically distant from the majority opinion
author (Hettinger et al. 2003a). The ideological make-up of the majority coalition is
important in the context of the American courts because of the majority opinion tradition,
but the make-up of the panel as a whole will be more important in Australia. Because
there is no need to marshal a majority for the opinion, the ideological distance of the
individual judge to the mean Of the entire panel is the relevant factor. Research on the
High Court has suggested that when a Justice is an outsider, he is less likely to join
(Smyth 2001).

ijothesis 4a: When a Justice ’s ideological posi tion is far from the mean position
of the bench, he will be less likely to join.

 

80 . . . . .
One of the most accurate ways to do this rs to analyze the content of various sources of information

besides the judicial record (Segal and Cover 1989). As yet, no such measure has been created for members
of the High Court.

167

Case Characteristics

Certain characteristics Of the cases before the judges will likely be a factor in
determining whether a joint judgment will be produced. Specifically, joint judgment
production may depend on case complexity, salience, the influence of other government
actors, and litigation history.

Case complexity

Over time, the Court has seen the complexity of its cases increase steadily (Leigh
2000). Complex cases have many potential points of disagreement, and this can make
consensus on a joint judgment less likely (Spriggs et al. 1999). The production of joint
judgments is more difficult in complex cases, and the resulting judgments are less likely
to capture the real differences among the Justices (COper 20020). It is difficult to
pinpoint exactly what makes a case complex. As Smyth points out, “cases which are
complex might be complex precisely because they involve policy considerations” (Smyth
2003a, 8). In the Supreme Court, research has shown that it takes significantly more
drafts to marshal a majority opinion coalition in complex cases (W ahlbeck et al. 1998).

In Supreme Court research, case complexity has been measured using a count of
the number of issues raised in the case, the number of legal provisions addressed
(Johnson 2003), the number of Opinions published in a case, and various combinations of
these measures (Bailey et al. 2005; Spriggs et al. 1999; Wahlbeck et al. 1998). In terms
of joining behavior, case complexity is important because “the legal complexity
surrounding some cases may create additional uncertainty for the individual Justices as
they figure out their own position” (Hoekstra and Johnson 2003, 352). When there are

several issues or provisions that could be controlling in a case, judges are less able to

168

anticipate the direction of the final draft of the maj ority judgment and are less likely to
agree with its emphasis. As such,

Hypothesis 5: Joining behavior will be more likely in simple cases than in cases
that are more complex.

In the US. Supreme Court, oral arguments are generally limited to thirty minutes
per side. Oral argument is the only opportunity for “a direct interchange of ideas between
court and counsel” (Rehnquist 1983, 1021). A recent study Of oral argument in the
Supreme Court found evidence that “justices need more information in complex cases
and a pertinent source of such information is oral arguments” (Johnson 2004, 12]).
Justice Byron White notes that Supreme Court Justices treat lawyers in oral arguments as
resources for clarifying their own thoughts on a case (White 1982).

While the Supreme Court maintains abbreviated arguments, the High Court has
implemented no strict limits on oral arguments. As one prominent Australian legal
scholar notes, “the United States Supreme Court allows only one hour of oral argument
per case whereas the High Court in the Bank Nationalisation Case81 let the argument run
for 39 days” (COper 2003, 6). These days, most arguments in the High Court take place
in a single day. In complex cases, though, oral arguments will sometimes run longer
(Kirby 2006). Because the variability of oral argument length is related to the complexity
of the case, the former can serve as a proxy for the latter.

ijothesis 5a: Joining behavior will be less likely in cases with long oral
arguments than in cases where oral arguments last less than one day.

Another commonly used approximation of case complexity is the length of the

Opinion used to dispose of the case (Black and Spriggs 2007; Hettinger et al. 2004;

 

81 Bank of New South Wales v. Commonwealth (1948). 76 CLR 1.

169

Schuck and Elliot 1990). The rationale behind this is relatively simple; it takes more
pages to explain the reasons for judgment in complex cases. Additionally, the
consideration of more than one case adds a dimension of complexity to the case.

Hypothesis 5b: Joining behavior will be less likely in cases with longer judgments
than in cases with short judgments.

Hypothesis 5c: When the Court hears a number of different cases together, joining
behavior will be less likely.

When a court exercises discretionary jurisdiction, it can weed out uncontroversial
cases in favor of more important ones. The US. Supreme Court gained much more
agenda control through the Judiciary Act of 1925. The difference in the Court’s output
was marked by fewer cases decided on the merits, longer opinions, and more dissents
(Halpem and Vines 1977). The differences in agenda are also apparent in other national
high courts (Songer and Johnson 2007). As the earlier chapters have discussed, the High
Court’s control over its agenda has expanded over time. The nature of the Court’s work
changed significantly as a result (Jackson 1996), leaving the Court free to address more
salient cases (Pierce 2006).

Hypothesis 5e: When the Court is exercising control over its agenda, joining
behavior will be less likely.

Case salience

Research on opinion writing in the American courts has long pointed to case
salience as an important factor in a judge’s decision to write separately (e. g., Kluger
1976; Murphy 1964; Schubert 1959). Indeed, most judges are thought to be “more likely
to be concerned with the legal rules announced in important cases” (Wahlbeck et al.

1999, 496). This is thought to work independently from workload concerns, with Justices

likely less to rely on organizational concerns when writing Opinions in salient cases

170

(Maltzman and Wahlbeck 2004). Indeed, judges in non-salient cases may be more
willing to assent to the written judgment of a colleague than they might otherwise be
(Dorff and Brenner 1992; Spriggs et al. 1999).

Hypothesis 6: In salient cases, joining behavior will be less likely than in non-
salient cases.

One common measure Of case salience in the context of the US. Supreme Court is
the participation of amici. While their specific content may not influence the Court’s
final decision (Spriggs and Wahlbeck 1997), their presence and number serve as an
indicator that the importance of the case extends beyond the litigants (W ahlbeck et al.
1999). Indeed, this measure can capture a case’s “salience to the larger social and
political systems” (Hettinger et al. 2003b, 224).

In Australia, however, the participation of non-party interveners is more
complicated. Since the Judiciary Amendment Act of 1976, the attorneys-general of the
federal government and the states have had the right to intervene in constitutional cases
under section 78A(1). Non-governmental parties enjoy no such right. Indeed, the Court

has maintained, to varying degrees, the strict standard laid out by Chief Justice Dixon
more than fifty years ago (Williams 2000). In Australian Railways Union v. Victorian
Railways Commissioners (1930), 82 Dixon requires that intervening parties demonstrate
“some particular right, power or immunity in which they are concerned” in order for the
Court to grant leave to intervene (331).

The process of obtaining leave to intervene is shrouded in mystery. The Court only

sporadically gives reasons for denying leave to intervene (Williams 2000), making the

 

82 44 CLR 319.

171

process quite difficult to navigate for potential interveners. Indeed, the costs of

intervening are quite high. According to George Williams,
[t]he situation is made worse by the fact that the Court does not hold a directions
hearing before the case begins to indicate whether leave will be given. Instead,
potential interveners must prepare submissions for the actual day Of the hearing
both on whether they should be granted leave and also setting out the full argument
that they wish to make in the event that leave is granted. Counsel appearing for a
potential intervener must be prepared to present oral argument to the Court on each
issue that might be of interest to the Bench. Of course, if leave is refused, the

expense and time spent in preparing firll submissions and oral argument is wasted.
(Williams 2000, 384)

Perhaps because of these high costs, Williams finds that most interveners in the
court, even through the 19905, are attorneys-general intervening under section 78A(1)
(ibid.). As such, the number of interveners in any given High Court case is made up of
two main groups. The first and larger group consists of attorneys-general intervening in a
constitutional case. The second group is made up of other, non-governmental interveners
who have a demonstrable stake in the outcome and the ambition (and resources) to win
leave to intervene.

In either situation, the presence of non-party interveners in a High Court case is an
indication that the case is salient. In US. Supreme Court research, the use of amicus
participation has been criticized. As Collins points out, “amicus participation in a case
does not imply the case is salient to a particular jurist; instead it denotes that the case is
salient to those organized interests who filed the briefs” (2005, 3). Ifthis is true in
Australia, however, it is to a lesser extent. In the case of the non-governmental
interveners, the judges themselves play a role in vetting the submissions. Their assent to
intervener participation, then, is more directly related to their own view of the salience of

the case. As such,

172

Hypothesis 6a: Joining behavior will be less likely in cases with non-party
participation than in cases without non-party participation.

A Federal Court Judge notes that Australian courts “have not, until recently
favoured the participation as amicus curiae of public interest groups” (Kenny 2004, 8).
In some cases, interest groups may bypass this leave requirement and get standing to sue
as parties to the case. Just as in the United States, and interest group that fails in the
political branches may seek to “have its views imposed by means of a sweeping ruling
from the Court” (Galligan 1987, 23 5). In this way, public interest litigants represent
another proxy measure for case salience.

Hypothesis 6b: When a non-governmental organization is a party to the case,
joining behavior will be less likely.

Another important dimension of case salience has to do with the legal importance
of the case. The increase of important constitutional cases is often cited among reasons
for the increasing incidence of separate Opinions on the US. Supreme Court (Kirman
1995). On the High Court, the tradition of rugged individualism remains particularly
strong in certain issue areas. As Chief Justice Gleeson explains,

The cases in which we’re likely to write separate opinions are fairly predictable.

They are, by and large, constitutional cases and tort cases. . .Once you get outside

those two areas you actually find a very substantial number of joint judgments.
(Clark 2003 2003 #188, 10)

While Justices on the US. Supreme Court have been known to seek unanimity on
controversial cases of constitutional law (Scott et al. 2003), the approach is quite different
in the High Court. In constitutional cases, the Justices “feel a very strict obligation under
the constitution to express their individual Opinions” (Clark 2003,10). The central
importance of these cases encourages the Justices to provide their specific set of reasons.

Hypothesis 6c: Joining behavior will be less likely in constitutional cases than in
other cases.

173

It may be reasonably clear to American onlookers why constitutional cases might
elicit special treatment, but coupling these with tort cases will seem quite unusual. Cases
in contentious issue areas may well have a lower level of consensus than the norm. The
US. Supreme Court rarely fails to produce a majority opinion, but when they do, the
“vast majority of these cases concern civil rights and liberties” (Segal and Spaeth 2002,
388)

In Australia, though, tort law has been used to firrther civil liberties, including free
speech (Luntz 2002). One commentator has noted that, in Australia, “[t]ort law is
characterized by loose sets of relatively abstract principles which allow maximum
discretion to be exercised (in the main) by reference to ‘common sense’ values” (Bennett

1989, 216). The Court has recently indicated the possibility of tort law as a means to

protect against privacy invasions by the media (Lindsay 2002)83 in the absence of
constitutionally derived protections as are enjoyed in the United States (Butler 2005). In
a similar way, tort law is an avenue for protecting women against various types of harm,
including job discrimination (Hocking and Smith 1996). The Court has also indicated
that tort law may be the avenue for pursuing litigation regarding the subject of abortion
(Luntz and Harnbly 2006), perhaps providing remedies both for “wrongful birth” and for
“wrongful life” (Stretton 2005).

Australian tort law has traditionally been largely dependent on common law
precedent, and only recently has the Australian Parliament begun to take an active role in
reforming the law in this area (Gardiner and McGlone 1998). The dearth of legislation

means that the judiciary must take a more active role in creating a system of tort law than

 

83 ABC. v. Lenah Games Meats (2001) 185 ALR l

174

in other areas, and in addressing the calls for a more efficient and fair system of
balancing the interests of various groups of citizens (Cane 2003). This is especially
critical in situations dealing with the tort liability of public authorities, when the Court is
left to delineate the boundaries of individual and state power (Kneebone 1998). For this
reason,

Hypothesis 6d: 1n tort cases, joining behavior will be less likely than in other cases.
Government as a party to the case

When an agent of the federal government is a party to a High Court case, the Court
may approach its decision making differently. Rational choice research has been
developed to explain the relationship between the Court and other government actors in
the United States. The main point of this literature is a recognition that the Justices must
(or do) consider the power, position, and perspective of other players in the political
game. This neo-institutionalist approach emphasizes the importance of the judge’s
position in the context of the political system, seeing attitudes and motivations as only a
secondary influence on judicial behavior (Gillman and Clayton 1999).

Hypothesis 7: Joining behavior will be more likely when the Court is concerned
about the reaction of the other branches of government.

Because the executive branch controls the implementation of High Court decisions,
Justices wishing to pursue their policy goals will seek to avoid alienating the bureaucracy
wherever possible (Epstein and Walker 2001). While some research on the Supreme
Court has found evidence that executive influence is diminishing (Deen et al. 2001;
Pacelle and Marshall 2001), the threat of executive action may have a constraining effect

on the Court and its members (Johnson 2003).

175

In Australia, there is conflicting evidence as to the relative success of the federal
government in High Court litigation. Some evidence suggests that the federal
government fares better than any other subgroup in the High Court (Smyth 2000). Other
research finds only a modest net advantage for the federal government in public law cases
(Haynie et al. 2001). Even still, the High Court’s more explicit adoption of a lawmaking
role has caused tension between the Court and the other branches of government (Alvey
and Ryan 2005). As two Australian scholars have recently noted, “[w]hen the Court
speaks with many voices, its decisions are potentially more vulnerable to outside
criticism” (Smyth and Narayan 2004, 401). Ifmembers of the Court are interested in
creating a united front in the face of this influence, joint judgments may be the likely
result.

Hypothesis 7a: When the federal government is a party to a case, joining behavior
will be more likely.

Litigation history

Certain aspects of the case history may also be related to the production of joint
judgments. There may be systematic differences between cases appealed from different
types of courts.84 One type of distinction to be made is between courts of general
appellate jurisdiction and courts of specialized jurisdiction. A second distinction is
between state supreme courts and federal appellate courts.

Hypothesis 8: The source of the case will have an impact on the likelihood of
joining behavior.

 

84 Spaeth and Segal imply as much when they suggest that investigating the Supreme Court’s treatment of

different courts’ decisions “may help in the formulation and implementation of litigation strategies” (Segal
and Spaeth 2002, 229).

176

In the United States, there are few specialized judicial bodies from which cases
may be appealed to the Supreme Court. In Australia, there are several separate

administrative tribunals and other specialized tribunals that have been recognized by the

High Court as substantively different from the general jurisdiction courts.85 As one
researcher observes, the Court sees such tribunals as an extension of the executive power,
“accountable to the courts in the same way as other executive decision-makers” (Pearson
2007, 5). These tribunals are “a significant provider of adjudication services” in
Australia, and are “responsible for developing law affecting more Australians than the
courts” (Creyke 2001, 426). These tribunals have been saddled with tremendous
responsibility in issues ranging from the provision of unemployment benefits (Bacon
2002) to the protection of native title claims (McIntyre et a1. 1999).

In the American context, numerous studies have shown administrative agencies to
be among the most successfirl litigants before the Supreme Court (Cannon and Giles
1972; Pritchett 1948; Sheehan 1990, 1992), although less so before federal appellate
courts (Songer and Humphries 1999). In Australia, the right to judicial review of
administrative action, at least at the federal level, is enshrined in the constitution.
Innovations in administrative law over the years have resulted in the creation of
numerous administrative tribunals, including the creation of a general appeals tribunal,
the Administrative Appeals Tribunal (AAT) in 1975 (Griffiths 1985). While some
scholars have doubted the ideological importance of administrative law cases in the
United States (Wald 1999). Some prominent legal scholars have been critical of the High

Court’s general tendency to favor executive discretion, especially in matters of

 

85 Craig v. South Australia (1995) 185 CLR 163

177

immigration and other human rights cases (e. g., McMillan 2002). Research on the
impact of US. Supreme Court rulings on administrative law have shown that compliance
is enhanced when there is a clear statement of policy delivered by larger majority
coalitions (Baum 1981; Johnson 1987; Wasby 1970; but see Spriggs 1996). Additionally,
institutional and jurisprudential concerns may lead Justices to promote a more united
front in an administrative law cases (Cohen and Spitzer 1994). As such,

Hypothesis 8a: Cases appealed fiom administrative tribunals or specialized courts
will be more likely to result in joining behavior than cases fiom other sources.

In addition to differences in subject matter jurisdiction, the geographic jurisdiction
of the lower appellate court may be systematically related to appellate decision making.
Like the United States, Australia has a dual system of courts consisting of state courts and
federal courts. Because a “natural antagonism exists between the Supreme Court and
state courts—particularly the states’ highest appellate courts”—we may expect a
systematic relationship between high court decision making and the geographic
jurisdiction of the lower appellate court (Cannon 1974, 109). The “long tradition of
competition between courts” has also been acknowledged by members of the Australian
judiciary (Spigelman 2004).86

Hypothesis 8b: Joining behavior will be more likely in cases appealed fiom federal
appellate courts than fiom other courts.

 

Another consideration here is the difference in jurisdiction between the state and federal courts. While
there are differences, they are much less obvious than in the American system (Crawford 1993). With the
passage of the Jurisdiction of Courts (Cross-vesting) Act 1987 (Cth), the line between state and federal
jurisdiction became even more blurry. In Re Wakim (1999) 198 CLR 511, the Court struck down this
scheme, much to the disappointment of many observers (Spigelman 2004).

178

The Disposition of the Case

Certain characteristics of the outcome of High Court cases may also be related to
joining behavior. Important explanatory disposition characteristics include a unanimous
vote, the ideological direction of the outcome, and whether the Court overturns the
decision of the court below.

Hypothesis 9: Characteristics of the outcome of the case will have an impact on the
likelihood of joining behavior.

Unanimity

Joint judgments are more likely for an obvious reason: when all Justices agree on
the outcome, there are more opportunities to collaborate. Certainly, a unanimous vote
“does not necessarily indicate that there is agreement over how certain issues should be
treated” (Atkins and Green 1976, 740). Indeed, truly consensual decisions appear to have
made up less than three percent of the Supreme Court’s unanimous decisions on the
Warren and Burger Courts (Spaeth 1989). On the other hand, the absence of dissenting
votes presents a larger pool of colleagues supporting the same case disposition. For this
reason,

Hypothesis 9a: Joining behavior will be more likely in unanimous cases than in
nonunanimous cases.

The ideological direction of the outcome

In the Supreme Court, some evidence suggests that Justices react differently to the
progeny of liberal cases as compared to conservative cases (Spaeth and Segal 2001; but
see Songer and Lindquist 1996). Even so, there is little reason to expect that the
ideological direction of a case outcome would directly influence the rate of j oint

judgment production on the Court (Hurwitz and Lanier 2004). Instead, the degree of

179

congruence between the Court’s baseline ideology and the outcome of an individual case
may be related to joint judgment production.

Hypothesis 9b: Case-level joining behavior will be less likely in cases with a
liberal outcome.

Dissent is “usually conceptualized as a disagreement with the results reached by the
majority on a multi-judge court” (Peterson 1981, 412). Many studies involving joining
behavior in the United States involve case-level analyses of majority coalition size
(Rohde 1972a) and fluidity (Brenner and Spaeth 1988) as well as accommodation rates
(W ahlbeck et al. 1998). In these analyses, dissenting behavior is empirically equivalent
to any other behavior in which the Justice does not join the majority coalition. In other
words, the joining behavior of interest is joining the majority judgment. Other types of
joining behavior are less interesting in the Supreme Court, partly because they are quite
rare (Segal and Spaeth 2002). If we consider all types of joining behavior on the
Supreme Court, joint judgments are far more likely on the winning side of the vote than
on the losing side. This, of course, is a function of the majority opinion tradition.

In judge-level analyses of joining behavior, the vote status of the individual judge
can be controlled for in the analysis. One such study uses dissenting vote status as part of
the dependent variable, predicting the type of separate opinion a Justice will author or
join (W ahlbeck et al. 1999). This piece, however, does not attempt to separate joining
behavior from the act of authoring a separate Opinion. Segal and Spaeth (2002) evaluate
joining behavior in non-maj ority opinions, but do not separate these special opinions into
concurrences and dissents. They explain this decision in a footnote, arguing that “no
theoretical reasons support one rather than the other as a vehicle for the expression of

personal policy preferences” (Segal and Spaeth 2002, 395, n.82).

180

Without a majority judgment tradition, the relative rarity of joint judgments in the
majority and the minority is an empirical question. Certainly, there is at least one more
potential joining partner for Justices in the majority than for those in dissent. This is not
the case in the Supreme Court, where most members of the majority vote coalition feel
pressure to join the majority opinion coalition (Maveety 2003). In courts with a majority
opinion tradition, concurrences often “belong to the space of dissent” (Johnson and
Belleau 2006, 2). In the seriatim tradition, as Chapter 5 demonstrates, concurrences and
dissents are different in kind. Because “concurring” opinions on the High Court are
essentially part of the majority judgment (Lynch 2002a), it is the dissenting opinions
alone which “can be uncongenial and burdensome” (Kirby 2005a).

Hypothesis 9c: The decision to join will be less likely when a Justice dissents from
the ideological outcome of the case.

Allowing an appeal

Unlike intermediate appellate courts, final courts of appeal tend to overturn lower
court decisions more often than they let them stand. This is especially true in courts with
discretionary jurisdiction. Recent data from the Supreme Court show the rate hovering
above 75% (Segal et al. 2007). Because the Court may choose its cases, most of its
reversals represent its exercise of error correction (Glick 1983; but see Snyder 195 9).
There is evidence that Supreme Court Justices are more likely to join the majority
opinion in these error correction situations (Brenner et al. 1990). Similarly, High Court
Justices may be more willing to entertain joining opportunities in cases where the Court
allows the appeal, overturning the decision of the lower court.

Hypothesis 9d: Joining behavior will be less likely in cases where the decision of
the court below is overturned

181

Joining Behavior and Judge-Level Characteristics

In addition to the case-level factors addressed in the previous section, an
understanding of j oining behavior will require the incorporation of judicial characteristics
in the model. The relationship between joining behavior and judicial career and other
social background characteristics are explored. Additionally, the disparity between the
judge’s ideology and that of his colleagues is connected to joining behavior.

Approach to the Working Environment

Judges differ in the way they approach the decisions of law and fact they must
make as the centerpiece of their judicial duty (Wells 1991). While the attractiveness of a
judge as a collaborator might be related to things like expertise, the judge might instead
be something of an expert at collaboration. As two prominent High Court scholars have
noted, maybe “some Justices have a greater like of, and aptitude for, co-Operative work”
(Lynch and Williams 2005, 12). The development of collaboration skills, then, might
logically place a judge nearer to the top of the list of potential coauthors.

Hypothesis 10: A Justice ’3 decision to join will be influenced by his approach to
the task of judging.

Career history

As Ulmer notes, “judges, like other men, are born, develop, mature, and become
socialized” (1970, 580). Professional socialization, in particular, is an important
component of judicial decision making (Tate and Sittiwong 1989). In civil law systems,
judges and lawyers part ways early in the career paths; judges are trained to be judges and
lawyers to be lawyers (Gleeson 2003a). This is not the casein systems sharing the
English common law tradition. As in the United States, Justices on the High Court of

Australia come frOm advanced positions in different professional backgrounds. Most of

182

these judges come from one or more of three main groups: attorneys in private practice,
lower court judges and government attorneys (Hulme 1994). While the focus of each of
these professions is the law, the barrister’s approach to the working environment differs
greatly from that of the judge and the government attorney (Goldsmith 2006). As Chief
Justice Gleeson notes, “no one ever professed to believe that all good barristers would
make good judges” (Gleeson 2003a). In other words, the requirements needed to excel in
private practice are to some extend different from those needed on the bench.

While the life of a barrister may not be nasty, brutish or short, it is often described
as rather solitary. A twenty-five year veteran of the profession warns that prospective
barristers should be aware of the “long anti-social hours” (Etherington 2005). Early work
on the social backgrounds of judges has theorized that particular kinds of professional
experience that can alter the judge’s orientation toward cooperative behavior. Ulmer
finds some evidence that previous experience as a non-political lawyer can impact one’s
disposition for dissenting behavior (1970). His conception of dissenting behavior is
relevant in the present context because he describes the behavior as “a form of
compensation for those who perform better in the dissenter’s role than in the role of the
majority judge — a position often calling for an ability to compromise” (Ulmer 1970,
588)

As compared to the work of the barrister, the work of a political attorney is more
collaborative in nature. Australian attorneys-general are uniquely political as compared
to their counterparts in the United Kingdom (McCarthy 2004). Their responsibilities
include the formulation of legal advice for the relevant government officials, as well as

significant ministerial duties (Carney 1997). Indeed, the significant decision-making

183

duties of the attorneys-general make them more similar to judges than to self-employed
barristers (Comall 2007). While their private-sector counterparts work alone
(Etherington 2005), government attorneys work in collegial departments. This
experience provides them with the ability to “[let] everyone have their say within a
sensible structure and time frame,” which fosters the ability to collaborate (Goldsmith
2006, 8). As such,

Hypothesis 10a: Justices with previous experience as government attorney will be
more likely to join than will other Justices.

In addition to experience as a government attorney, judicial experience likely
contributes to the development of skills and temperament needed to produce joint
opinions. Previous research has found judicial experience to contribute to higher rates of
liberal decisions in Canada (Tate and Sittiwong 1989) and the United States (Tate
1981b). It has been associated with assertion of judicial power (Wood et al. 1998a) and a
willingness to abandon previous court precedents (Schmidhauser 1962). A recent meta-
analysis found that “of the twenty-two studies located that investigate this linkage, nearly
70% found some sort of a relationship between career experience and judicial choices”
(Epstein et al. 2003, 954).

There is reason to believe that previous judicial experience may socialize judges to
the practice of writing jointly. Intermediate appellate courts in America are generally
characterized by their high levels of consensus (Brace and Hall 2005; Hettinger et al.
2003b). Justice Kirby’s (2007b) account of opinion writing on the Supreme Court of
New South Wales indicates that joint opinions are the standard on that court.

Hypothesis 10b: Justices with prior judicial experience will be more likely to join
than those without previous judicial experience.

184

Acclimation to the job and colleagues

Much of the American research suggests that opinion assignors will assign fewer
majority opinions to freshman judges and most attribute this pattern to the organizational
concerns of the assignor (Brenner and Hagle 1996; Howard 1968; Maltzman and
Wahlbeck 1996a). Others attribute the pattern to the freshman’s inability to marshal
strong majorities in support of their majority opinions (Lax and Cameron 2007). Because
experienced authors take less time to construct well-crafted opinions, they “need not

compromise so heavily toward the ideological preferences of the median Justice” (Lax

and Cameron 2007, 294).87 An analysis of the High Court found evidence that about a
third of the Justices experienced acclimation effects, making them more likely to join at
the beginning of their tenure (Smyth 2002a).

Hypothesis 100: Justices will be more likely to join during the beginning of their
career on the High Court bench.

Another influence on the decision to join may be a predisposition for co-authorship.
Indeed, “it may be that some Justices have a greater like of, and aptitude for, co-operative
work” (Lynch and Williams 2005, 12). Ifa Justice has established a pattern of joining
behavior, he is likely to continue that pattern in subsequent cases.

Hypothesis 10d: When a Justice has established a pattern of joining, he will be
more likely to continue to join.

Social Background Characteristics

Researchers have long used the personal attributes of judges to explain their voting

behavior, essentially capturing relevant judicial attitudes indirectly (Goldman and Sarat

 

87 While evidence from the Supreme Court Of Canada show no freshman effect on voting or opinion-
authoring patterns, the researchers attribute this finding to that courts greater interest in “fostering and
maintaining a collegial environment and reaching unanimous rulings” (Ostberg et al. 2003, 710).

185

1978; Schmidhauser 1962; Tate 1981b). Certainly, this is the approach that has been
taken by most of the jurimetrics scholars in the investigation of judicial decision-making
behavior on the High Court of Australia (Blackshield 1972; Schubert 1969a). In his
study of the Dixon Court, Schubert included more than a dozen personal attribute
variables (Schubert 1969a). Indeed, even some of the recent quantitative work on the
High Court has centered on the use of the personal attribute model (Smyth 2003b).

While most of the personal attribute research seeks to predict voting behavior, it may also
be relevant to an assessment of joining behavior.

Hypothesis 1]: Certain personal background characteristics of the judges will
influence their decision to join.

Religious background

Religious background is another time-tested indicator of judicial behavior (Ulmer
1973). In particular, Catholicism has been associated with judicial support for liberal
outcomes in the United States (Schubert 1969a) and Canada (Songer and Johnson 2007;
Tate and Sittiwong 1989). It has also been linked to higher rates of dissent behavior in
the United States (Ulmer 1970) and Australia (Smyth 2003b).

Hypothesis 1 1 a: A Justice’s religious background will influence her decision to
join.

State-based variations

Previous research indicates that where a person grows up can influence his
behavior as a judge later in life (Tate 1981b). Research on the Canadian Supreme Court
indicates that regionalism—specifically, whether a judge is from Quebec—has an effect
on judicial liberalism rates (Songer and Johnson 2007; Tate and Sittiwong 1989). Region
has also been linked to perceptions of judicial ability (Walker and Hulbary 1978) and the

tendency to dissent (Schmidhauser 1962).

186

Having practiced law at the New South Wales Bar (N SWB) is thought to be a
unique experience that may shape a Justice’s attitudes and approach (McMillan 2003).
The NSWB has a tradition of very high standards, and has been the preeminent legal
association since before the Australian founding (Bennett 1969). The Court itself has
been accused of being distinctly Sydney-centric (Allen 2008; Hudson and Crabb 2002),
as has the preeminent legal journals in Australia (Lindgren 2007).

Hypothesis 1 1 b: Being fiom New South Wales will aflect the likelihood that a
judge will join.

Party of the Appointing Prime Minister

In the United States, the selection of Supreme Court nominees is overtly political.
Nominees are selected according to the President’s perception of the nominee’s political
ideology and judicial philosophy (Maltese 1998). Indeed, political party affiliation is a
long-used and successful explanatory variable for judicial decisions on the merits in the
United States (Pinello 1999). Outside of the United States, though, high court judges are
generally perceived to be “substantially outside the systems containing those who make
and administer the laws” (Dawson and Ward 1987, 254).

The current Chief Justice has taken pains to defend the idea that the party of the
appointing prime minister has no relationship to the behavior of the appointees.
According to a piece in the Sydney Morning Herald, “Gleeson has even gone to the
trouble of preparing a statistical analysis which, he says, show the High Court is
untouched by political bias” (Ackland 2003, 5). Ifthe party of the appointing prime
minister in Australia operates in the way the party of the appointing president is in the
United States, then it is important to include this variable in any model of judicial

behavior in the High Court.

187

Hypothesis 1 Tc: Justices appointed by Australian Labor Party Prime Ministers
will be less likely to join than will those appointed by Liberal-National Coalition
Prime Ministers.

These eleven hypotheses have been adapted from the existing literature, which is
heavily weighted toward explaining joining behavior on the US. Supreme Court. The
defining characteristic in the Supreme Court—at least as far as joining behavior is
concerned—is the majority opinion tradition. The High Court of Australia does not share
this majority opinion tradition. As the previous chapter demonstrates, there is little
evidence that American-style consensual norms drive opinion-writing behavior on the
High Court. As such, the hypotheses in this chapter have been adjusted to account for
these differences. In the next chapter, these hypotheses will be operationalized using data

from the High Courts Judicial Database (Haynie et al. 2007).

188

CHAPTER SEVEN: DATA AND MEASURES

I am sure that these descriptions will strike a responsive chord with those of us who
increasingly shake our heads in despair at the typical joint majority opinions of the
present High Court - pretentious displays of erudition that yield no intelligible
principle, and contribute nothing to any discernible pattern of legal reasoning that
justifies the result.

—Professor Anthony Blackshield (2004)

While some observers, such as Professor Blackshield, have been disappointed with
the shape of the Court’s recent joint opinions, the Court has been making use of this type
of opinion at an increasing rate (Coper 20020). This chapter sets out to explain the
presence of j oint opinions in an institutional context that is seemingly hostile to
cooperative opinion writing. Operational hypotheses are presented below, following the
structure of the previous chapter. The data that will be used to test these hypotheses are
described, as are some of the methodological issues these data may present.

Joining Behavior on the High Court

In the seriatim tradition, the judges each typically pen their own opinion and are
identified as authors in the reports. Because of this tradition, the articulation of
individual opinions is relatively high on the High Court. One way to illustrate this
relative frequency is through a calculation of the ratio of opinions to votes (Schubert
1969b). Over time, the opinion-to-vote ratio has tended to decrease substantially (see
Figure 9). From the beginning of the Barwick Court in 1964 through the retirement of
Justice Gaudron in 2003, this ratio is .621. This is much lower than the .917 ratio
reported in Schubert’s analysis of the High Court from 1951 to 1961 (Schubert 1969b).
This indicates that the rate of j oint opinions on the High Court has increased since the

1950s

189

Figure 9 - Opinion to Vote Ratio on the High Court over Time

9

 

81

6.

Sq

44

 

Opinion to Vote Ratio

 

 

U U V U V V I V V U ‘U V V U V V

1965 '67 '69 '71 '73 '75 '77 '79 '81 '83 '85 '87 'I '91 '93 '95 '97 '99 '01

Schubert defined joining behavior as “social deference,” and created a scale of
inter-agreement between judges as its measure (Schubert 1969b). The deference measure
is impractical, though, in large part due to the increasing inability to distinguish opinion
writers from opinion joiners. The typical unanimous case in the seriatim tradition has

several separately written opinions. In the present analysis, 20% of the cases have one

opinion for each vote cast.88 For these opinions, the single author of each is clear. A
parallel could be drawn between this situation and a set of special concurrences in a
Supreme Court case, although there is no single majority opinion on the High Court with
which to concur.

This is not always the case, though. In some instances, the High Court will hand
down a single set of reasons, disposing of the case with a single opinion. In such a
situation, it is easy to identify the majority opinion. Because the judges’ names are listed

in order of seniority, however, it is nearly impossible to ascribe properly the authorship of

 

88 . . . .
This 18 426 cases of the 2147 cases With more than one judge on the panel.

190

such an opinion. In this way, the author of a single-opinion disposition on the High Court
can only be identified as “the Court.” In other instances, two or three members of the
Bench may join. Again, the authors of such a joint opinion will be listed in order of
seniority, with no indication as to the authorship of the opinion. Assigning writing
behavior to individual judges, then, is difficult.
Problems Distinguishing the Author fi'om the Joiner

Although joining behavior is the subject of this analysis, distinguishing joining
from authoring on an individual level is also problematic. On occasion, High Court

Justices have been known to join the opinion of a colleague with a single statement, like

the following from the opinion of Justice Kitto in Murphy v. Ramsay (1964):89

KI'I'I‘O, J .: I agree in the judgment of the Chief Justice and have nothing to add.

In this case, it is clear that Chief Justice Barwick has authored an Opinion and
Justice Kitto has joined it. For this reason, it is possible to call this type of opinion a
“joining statement.” In this analysis, authors of such joining statements are included as
joiners to the opinion. When a joining statement is present, joiners are easily
distinguished from authors.

Another Justice in Murphy v. Ramsay provides an example of a more complex form

of j oining behavior. The following is the beginning of Justice Taylor’s opinion in that

90
case:

TAYLOR, J. I agree that this appeal should be dismissed for the reasons given by
the Chief Justice and I wish to add only a brief observation with respect to the
appellants' argument .. . .

 

89 111 CLR 344 at page 351

90 Ibid.

191

After this, Justice Taylor proceeds to clarify his perspective for a paragraph. While there
is no tradition of concurring opinions on the High Court per se, this example is similar to
what would be called a simple concurrence on the Supreme Court. The trouble, of
course, is determining the point at which a joining statement becomes an opinion of its
own accord. For the purposes of the present analysis, instances in which the judge
includes substantive content or clarification (beyond what is contained in the joined
opinion) have been treated as individual opinions. This is the classification for the
statement by Justice Taylor excerpted above. Where the separate statement provides no
more than a restatement of the arguments and information of the joined opinion, the
statement is classified as a joining statement.

Sometimes, a judge will write a separate sentence adopting the reasons from a

jointly-written opinion. In Geyer v. Downs (1977),91 Justice Mason adds another layer of
complexity, agreeing with the reasons of two different opinions:

MASON, J. I would allow this appeal for the reasons given by Murphy and Aickin
J .J . in their joint judgment and for the reasons given by Stephen J.

The one thing that is clear from this opinion is that Mason is doing the joining. We can
also assume that Justice Stephen has authored one opinion with which Mason agrees, but
we can only speculate as to who authored the Murphy-Aickin opinion. Indeed, it may
well have been a truly collaborative effort. While this may provide a clear example of
joining behavior on the part of Justice Mason, it does not give us a way to identify a
single judge (nor a single opinion) with which Justice Mason agrees. In order to avoid
the perilous task of matching such an opinion with one or another of the joined opinions,

such cases have been classified as individual opinions. For the example above, this

 

91 138 CLR 91 at page 95.

192

opinion is classified as a single-authored opinion by Justice Mason. While the vote totals
will indicate his agreement with both of the joined opinions on the merits, this particular
type of statement does not allow for a replicable allocation of the joiner/author
combination.

These considerations, combined with the absence of data on conference votes,
make the present analysis distinct from investigations of j oining behavior on the Supreme
Court. The use of the term “joiner” in this context is perhaps inaccurate; because we
cannot systematically distinguish the three, “joiner” will refer to joiners, coauthors and
joinees. In the same way, the term “joining behavior” will refer to joining, coauthoring
and being joined by another Justice.

Operational Definitions of Joining Behavior

The unit of analysis for the first series of models is the individual case. This single-
opinion model will be applied first to all cases, and then to only cases with a unanimous
vote on the outcome. This second application will allow for the investigation of the
characteristics that make single-opinion cases different from other unanimous cases. The
dependent variable in each analysis is a dichotomous variable, where the presence of the
joint opinion is coded as 1 and the absence is 0.

The data for this analysis come from the High Courts Judicial Database (HCJD)
project (Haynie et al. 2007). This ambitious data proj ect—with over 17,000 judicial
decisions collected across ten national high courts—is a worthy answer to Schubert’s
attempt to “stimulat[e] the production of better quality data by demonstrating the effects
of their lack” (Schubert 1969b). The data file for the High Court of Australia consists of
the universe of cases reported in the Commonwealth Law Reports between the beginning

of the Barwick Court (April 27, 1964) through the retirement of Justice Gaudron

193

(February 10, 2003). Cases in which only one judge presided have been removed,
leaving 2145 cases and 873 unanimous cases.

For the period under analysis, 19.3% of all cases were disposed of with a single
opinion (see Table 10). In the subset of unanimous decisions, 32.1% were single-opinion
dispositions. Joint opinions involving two or more Justices were present in 80.7% of all
cases, and in 86.4% of unanimous cases. For both of these dependent variables,
extremely short statements by individual Justices indicating their agreement with the
reasons of a colleague are included in the tally of opinion joiners. Any statement longer
than a sentence or two, or any statement adding substantive observations to the opinion of
a colleague, is counted as an individual Opinion.

The second series of models evaluate the individual decision to join. This model
will be applied first to all decisions and second to criminal cases only. For each analysis,
the dependent variable is a dummy variable, which is coded as 1 if the Justice was

involved in a joint opinion in the case, and 0 otherwise. These analyses are conducted on

a judge-centered version of the HCJD project data (Brennan 1996b, 13).92 This
transformation leaves the data organized by opinion-writing opportunity. There are
10,855 opinion-writing opportunities in the full dataset, 2,425 of which are criminal
cases. In the fill] judge-centered dataset, 56.7% of opinion-writing opportunities resulted
in a decision to join. The rate for criminal cases is slightly higher, at 57.8%. As in the
first series of models, joining behavior includes short statements of assent, but excludes

longer statements.

 

.,
9" The data transformation was completed using STATA code detailed by Collins (2006).

194

Predicting the Single Opinion for the Court
Few activities of the members of the United States Supreme Court are more
important than the crafting of a majority opinion. It is the legal rule articulated in the
majority opinion, not just the disposition of a particular case, that serves as a guide for
other branches of government, the lower courts, and even the general public (Wahlbeck
et al. 1998, 311). In most existing judicial decision making research, unanimous cases
disposed of with a single opinion are often seen as the least interesting (e. g, Schubert
1965; Tate 1981b). When the focus is on unexpected patterns of agreement (as opposed
to unexpected pattems of disagreement), these cases become far more interesting.
Specifically, these cases are examples of the most obvious abrogation of the seriatim
norm. By analyzing these cases, we can determine why some unanimous cases are
disposed of using a single opinion instead of the multiple opinions we would expect.
In their investigation of decision making on the Supreme Court of Canada, Tate
and Sittiwong explain their decision to exclude unanimous decisions from the data:
Nonunanimous decisions are used because they mark the territory of dissensus
among the justices of a collegial court; they are the cases on which the justices are
unable to reach agreement on a single policy outcome, despite institutional
pressures to do so, the cases in which presumably learned and well-intentioned
jurists can see at least two alternative interpretations of the law, at least two
competing policy options. (Tate and Sittiwong 1989, 902)
Here, a strong case is made for excluding unanimous decisions from an analysis of the
conditions in which judges disagree with one another as to the outcome of a case. For a
case to be helpfirl in this regard, it must have some interesting variation among the votes

of the participants. Ifwe are interested in the decision to join, however, we are looking

for disagreement of a different kind.

195

It is certainly feasible for judges agreeing on the outcome of a case to disagree on
the wording of the reasons. In the context of the US. Supreme Court, this takes the form
of the concurring opinion. This possibility increases the alternatives available to the
judges

In arriving at their decisions, the individual justices are free actors in two separate

senses. First, they may vote as they see fit. Second, except for the opinion of

the Court, the justices may write such opinions as they desire to explain their
individual votes. As a consequence, a justice may be a member of the voting

coalition, but not a member of the opinion coalition that supports that vote. (Segal
and Spaeth 2002, 383)

Of course, this situation is the norm in the seriatim tradition. There is no institutional
norm encouraging the creation of a majority opinion for purposes of conferring
precedential weight. In other words, the second dimension of this freedom of action is
much more pronounced in the High Court. Not only are judges free to vote as they wish,
but they are also free of institutional pressure to adopt the reasons of their peers.

In a court with a seriatim tradition, then, we would expect most unanimous
decisions to consist of an explanation of reasons from each of the participating judges. In
the case of the High Court, a significant percentage of unanimous cases are resolved by a

single opinion. In the 873 unanimous decisions of the High Court in the nearly four

decades included in this analysis, 32% of these were single-opinion dispositions.93 As
Figure 10 shows, the early 19805 are associated with a large increase in the percentage of
unanimous cases disposed of with a single opinion. Beginning in the early 19905, the

single-opinion disposition rate begins its return to the earlier pattern.

 

93 As noted earlier, some of these single judgments take the form of a single statement of reasons
accompanied by several single-sentence statements of agreement from the other judges on the panel.

196

Figure 10 - Percentage of Unanimous Cases with Single-Opinion Dispositions

100

 

9m
80-
7o«
60‘
50-
40-
304
20-

10'
0 U U U U U U U U U U U T U U U U U U
1%4 '58 '68 '70 '72 '74 '78 '78 'I '82 '84 '85 'w '90 '92 '94 '% '98 '00 2002

 

 

 

Approaching the question from the case level is the most appropriate strategy for
analyzing the production of single-opinion case dispositions. The actions of each
individual Justice are important, but the only way a single opinion can be produced is if
all of the members of the panel choose to join. Using a binary logistic regression, we can
investigate the case-level hypotheses from the preceding chapter as they relate to the
production of this particular type of j oint opinion.

Predicting Other Joint Opinions

Of course, the singl e-opinion disposition is not the only joining behavior we see in
the High Court. Despite the lack of institutional incentive, the Justices have been
forming smaller opinion coalitions with increasing frequency (see Figure 11). Observers
have sometimes lauded (Kildea 2003; Moisidis 2002) and sometimes criticized
(Blackshield 2004) this development. None, however, has been able to pinpoint the

reason for this trend (Bagaric and McConvill 2005; Coper 2002c).

197

As in the case of single-opinion dispositions, an analysis of j oining coalitions does

not require the exclusion of unanimous cases. Instead, the influence of unanimity on the

production of these joint opinions can be evaluated empirically. Of the 2,145 cases

included in this analysis, 80% have at least one joint opinion. In nearly 57% of opinion-

writing opportunities, Justices engaged in joining behavior.

Figure 11- Percentage of Cases with Any Joint Opinion and Smaller Coalitions Only

 

100
l I / \ \ d
90. , \- — a \ / \ u
N f \//
80s . . . 1 "
Any JorntOpinion , ,\
704 \
fx/ \
604 x/ V Smaller Coalitions
\

501'” I \ / \\/

40..

301

20s

10¢
0

 

 

 

1984 '88 '88 '70

1

'72 '74 '78 ’78 '8] '82

'84

U

'8'% '93'92 '94'% '98 '[IJ2002

Operational Hypotheses

The challenge now is to explain these instances of joining behavior. The

conceptual hypotheses outlined in the previous chapter provide numerous plausible

explanations for joining behavior on the High Court. The concepts presented in Chapter

6 are operationalized, providing a set of propositions that can be testing using the HCJD

project data (Haynie et al. 2007), with additions and modifications as described below.

Characteristics of the Decision Making Environment

Hypothesis 1: As workload pressures decrease, joint opinions will be less likely.

198

As the previous chapter illustrates, workload pressures are expected to decrease the
joint opinion production rates on the High Court. For this institution, though, some
workload pressures are easier to isolate than others are. While the judges cite increasing

special leave hearings as an important drag on their time, there is not as of this writing a
readily available source of information about this topic.94 Drawing on previous research
addressed in Chapter 6, though, several measures of Court-level workload pressures can
be derived.
H )pothesis 1 a: When oral arguments take place less than three months before the
end of the term, single-opinion dispositions, joint opinions, and decisions to join
will be less likely.
Hypothesis 1b: When final opinions in a case are handed down within three months
of a known retirement single-opinion dispositions, joint opinions, and decisions to

join will be less likely.

Hypothesis 1 c: When the number of cases heard in a term is higher, single-opinion
dispositions, joint opinions, and decisions to join will be less likely.

The independent variables in the above hypotheses serve as operational proxies for
the workload pressure facing the Justices. The oral argument variable is a dummy
variable coded as 1 when the oral argument is held in the last three months of the term,
and 0 otherwise. The dates for oral argument are included with the official opinions in
the Commonwealth Law Reports. Because the Court’s term ends before the July recess,
this variable is coded as 1 when the oral argument falls between March 1 and June 31.
This is a similar strategy to that used in research on the US. Supreme Court (W ahlbeck et

al. 1998). For both unanimous and non-unanimous cases, oral arguments were heard

 

94 The Commonwealth Law Reports provide lists of special leave petitions that were denied, but these data
have not yet been aggregated into special leave rates over time. Additionally, these special leave listings
indicate which Justices were on the special leave panel. These data, once compiled, could allow for judge-
specific estimations of agenda-setting pressures.

199

during the last three months of the term about 40% of cases and opinion-writing
opportunities (see Table 10).

The retirement proximity variable is also a dummy variable. This variable is coded
1 if the final decision was handed down within three months of a known retirement, and 0
otherwise. The distinction is made between known retirements and other departures from

the bench (e. g., deaths) because the hypothesized effect has to do with the judges’

reparation for the departure.95 Around 20% of the cases in the dataset—and 23% of
opinion-writing opportunities—were decided within three months of a known retirement.
The final variable from this set of hypotheses is the number of cases decided during the
term. For the purposes of this analysis, the variable has been calculated as a count of
cases published in the Commonwealth Law Reports with dates falling within the relevant
term. The average number of cases per term is around 62 for the period under
investigation. This number does not include the number of special leave hearings, nor
does it include unreported decisions.

Hypothesis 2: Joint opinions will be more likely during the tenure of Chief Justices
with successfitl leadership styles.

As the previous chapters have indicated, the Chief Justices of the High Court have
enjoyed varying degrees of success in creating an atmosphere of cooperation on the
Bench (Smyth and Narayan 2004). Some of this has been traced to the inability of some
Chiefs to coordinate the activities of highly individualistic judges (Fricke 1986). To be

sure, Chief Justice Barwick showed little interest in maintaining collegiality as his term

 

95 Information about retirement dates was gathered from various sources, including the Oxford Companion
to the High Court of Australia (Coper et al. 2002) and the High Court of Australia’s official website
(http://wwwhcourtgovau).

200

Table 10 — Descriptive Statistics for Model Variablesl

( 'riminal

All l'nzmimou All

 

 

 

Variables (uses 8 (‘ascs ()\\‘()s ()“Os
n=2,145 n=873 n = 10,855 n = 2,425
Dependent Variables
Joint Opinion Flag 80.7 % 86.4 % 56.9 % 57.8 %
Single-Opinion Disposition 19.3 % 32.1 % - - - - - -
Independent Variables
Workload
Late Oral Argument 40.7 % 42.0 % 41.0 % 37.0 %
Retirement Near 21.4 % 19.3 % 23.0 % 25.0 %
Cases This Term 61.3 cases 61.9 cases 60.2 cases 57.8 cases
Chief Justice
Barwick 45.5 % 46.9 % 42.0 % 31.0 %
Gibbs 17.3 % 18.3 % 17.5 % 16.8 %
Mason 19.3 % 18.7 % 21.1% 27.9 %
Brennan 6.8 % 6.4 % 7.3 % 8.3 %
Gleeson 11.2 % 9.7 % 12.5 % 16.1%
New ChiefJustice 12.7 % 13.9% 13.2% 15.0%
Ideology Measures
Panel Liberalism Score .09 .08 - - - - - -
Panel Diversity . 13 . 1 3 0.09 0.09
Distance from Panel Mean - - - - - - 0.07 0.07
Agree with Outcome Direction - - - - - - 58.0 % 69.3 %
Complexity
Long Oral Argument 45.8 % 41.3 % 47% 40.0 %
Avg. Opinion Length 10.2 pages 10.0 pages 10.6 pages 9.7 pages
Number of Matters 1.2 matters 1.2 matters 1.2 matters 1.2 matters
No Leave as of Right 83.6 % 82.8 % 50.0 % - - -
Salience
Nonparty Participants 0.5 parties 0.4 parties 0.6 parties 0.4 parties
NGO as a Party 7.5 % 7.5 % 8.2 % 0.7 %
Constitutional Issue 18.0 % 16.0 % 22.0 % 22.0 %
Tort Cases 14.9% 13.5% 17.0% ---
Fed. Gov’t as a Party 38.6 % 39.2 % 37.0% 72.5 %
Litigation History
Admin/Specialized 7.5 % 8.2 % 7.6 % 1.6 %
Federal Jurisdiction 29.7 % 28.4 % 12.4 % 5.7 %
Case Outcome Characteristics
Unanimous Vote 62.4 % - - - - - - - - -
Liberal Outcome 42.3 % 45.0 % - - - - - -
Allow Appeal 48.0 % 47.1 % 47.0 % 50.0 %
Judge Characteristics
Gov’t Attorney Experience - - - - - - 39.2 % 38.5 %
Prior Judicial Experience - - - - - - 62.8 % 67.4 %
Freshman (1" 50 Cases) - - - - - - 5.4 % 5.1 %
Predisposition to Join - - - - - - 0.43 0.46
Catholic - - - - - - 45.0 % 50.8 %
New South Wales - - - - - - 57.8 % 57.2 %
Labor Appointing P.M. - - - - - - 30.0 % 34.0 %
Control Variables
Panel Size 5.1 Justices 5.0 Justices 5.4 Justices 5.5 Justices
Coalition Size - - - - - - 4.4 Justices 4.4 Justices
T For the dummy variables, the results given are the percentage of judge decisions where x=1. For the
other variables, the results are the means. Note that cases decided by a single Justice are excluded.

 

 

progressed, to the extent that he refused to speak at all to one of his colleagues (Hocking
1997)
Hypothesis 2a: Single-opinion dispositions, joint opinions, and decisions to join

will be more likely in the Mason and Gibbs Courts and less likely during the
Barwick Court.

H )pothesis 2b: Single-opinion dispositions, joint opinions, and decisions to join
will be more likely during the first year of a new Chief Justice ’s tenure.

The role of the Chief Justice in facilitating single-opinion dispositions is captured
in this analysis by the use of a nominal-level variable. For each Chief Justice in this
sample—Barwick, Gibbs, Mason, Brennan and Gleeson—this variable takes on a
different value. In the full model below, the Barwick Court is the reference category, and
the results for the other Chief Justices are in comparison to Justice Barwick. This is not
an arbitrary decision. The Barwick Court spans the longest period in these data,
stretching from 1964 through 1981. Nearly half of the decisions in this dataset were
made under Barwick’s tenure. Because panel sizes have increased over time, only 42%
of opinion-writing opportunities in these data come during Barwick’s tenure.

Chief Justice Barwick is also known for his unsuccessful attempts at instituting
consensus-building conferences (Blackshield 1980; Winterton 1998). Indeed, Barwick
himself has confessed that his suggestion that conferences be a regular part of the
calendar fell on deaf ears (Barwick 1995). The Mason Court, by contrast, is recognized
for its cooperative procedures and generally positive personal relationships (Groves and
Smyth 2004). Chief Justice Mason was able coordinate draft opinion exchanges and, in
this way he “fostered [the Court’s] collegiate spirit” (Brennan 2002b, 13).

The beginning of a new Chiefship may also affect the production of joint opinions.

Previous research on acclimation effects has measured the “freshman” period as the first

202

one (Brenner and Hagle 1996), two (W etstein and Ostberg 2005), or three years (Hagle
1993; Smyth 2002a). Most of the research uses the one-term time span, as Justices are
expected “to adapt quickly to the simple mechanics of the Court’s opinion writing
procedures and workload” (Brenner and Hagle 1996, 239). The new Chief Justice
variable is coded 1 when the case or opinion-writing opportunity comes in the first year
of a new Chiefship and 0 otherwise. About 13% of opinion-writing opportunities are
under the tenure of a new Chief Justice. This is 12.7% of all cases, and 13.9% of
unanimous cases.

Hypothesis 3: The likelihood of a joint opinion will depend on the ideological
makeup of the panel.

Research on the US. Supreme Court generally utilizes two liberal-conservative
dimensions—one for economics and one for civil rights and liberties—to capture the
meaningful ideological differences among the Justices (Mishler and Sheehan 1996;
Schubert 1965; Songer 1978; Tate 1981b). These two scales explain the bulk of the work
of the Supreme Court, which is “unambiguously classifiable into these two categories”
(Tate 1981b, 356). The work of the High Court, though, may not be so easy to classify.
In his very early jurimetrics work on the High Court, Schubert used the two-scale
strategy to model ideological outcomes (Schubert 1969a). The same year, Schubert used
a collectivism scale and an authoritarian scale to predict judicial interagreement (1969b).
Blackshield (1972) used six separate areas to predict judicial votes, none of them except
the criminal appeals category approximating the American civil rights and liberties
dimension.

For the purposes of predicting joining behavior across all issue areas, though, an

issue-specific set of ideological scales may be unnecessary. Instead, individual judicial

203

liberalism scores are computed using an average of the outcome direction across the
universe of each judge’s votes. The original HCJD outcome direction variable has been
recoded such that -1 is a vote supporting a traditionally conservative outcome, and 1
supporting a traditionally liberal outcome (see Appendix B, Section 3). Decisions with
no identifiable outcome directionality were recoded to 0. The resulting scores range from
-.14 to .37. This scoring demonstrates concurrent validity, as the mean score for Labor

appointees is significantly higher than the mean score for Liberal appointees (see Figure

12).96 Because the dependent variables are all related to joining behavior, using voting
behavior to derive the independent variable is acceptable.

Hypothesis 3a: Single-opinion dispositions and joint opinions will be more likely
on panels with high mean liberalism scores than on panels with low scores.

Hypothesis 3b: Single—opinion dispositions, joint opinions, and decisions to join
will be less likely on panels where there is more variation in the members’
liberalism scores than on more ideologically homogenous panels.

For the case-level analyses of joining behavior and single-opinion dispositions, the
mean panel liberalism score is a way to isolate the general ideological makeup of the
panel. Because the High Court fiequently sits in rotating panels, the mean panel scores
vary from case to case as well as among natural courts. The metric remains the same as
the individual level, with higher numbers representing greater support for liberal
outcomes. The overall panel mean is .09, and falls to .08 when nonunanimous cases are
excluded. 1

For the case-level and judge-level analyses, a measure of ideological diversity on

the panel is an important addition to a model of j oining behavior. For each panel, the

 

Liberal appointees have a mean score of 0.16, while Labor appointees have a mean score of 0.03. An
independent samples t-test not assuming equal variances rejects the null hypothesis that these scores are not
difi’erent from each other (t=6l .73, p=.000).

204

Figure 12 - Liberalism Scores by Party of Appointing Prime Minister

 

 

 

 

it -)(- -)1- arses smarts it
44 4 4 44 <
PARTY
*LNC
i . _ . . «ALP
-.2 -.1 0.0 1 2 .3 4

Liberalism Score

standard deviation of panel members’ individual liberalism scores serves as the
ideological diversity measures. The average panel diversity score across all cases is .13,
and across all opinion-writing opportunities is .09.

Hypothesis 4: The ideological position of the Justice relative to the panel mean
will influence the decision to join.

H ypothesis 4a: If a Justice ’s liberalism score is further from the panel mean, she
will be less likely to join.

In the judge-level analyses, we leverage the additional information about the
individual Justice’s ideological position relative to the panel mean. For each opinion-
writing opportunity, the Justice’s individual position is subtracted from the panel mean,
yielding a measure of the relative distance from the panel mean. For these analyses, the
average distance from the panel mean is .07 units.

Case Characteristics

Hypothesis 5: Joint opinions will be less likely in complex cases than in simple
cases.

205

As is the case in the US. Supreme Court (Wahlbeck et al. 1998) complex cases
likely require more individual accommodation in order to reach a consensus on the
reasons for judgment. Case complexity has been measured in numerous ways in previous
research. Some of these, like the number of opinions published in a case (W ahlbeck et a1.
1998), are obviously inappropriate for the present purpose. Measures of the number of
issues raised in a case or the number or relevant legal provisions (Maltzman et al. 2000)
are not as readily available for the High Court as they are for the US. Supreme Court.
Instead, several alternative measures are employed to capture variation in case
complexity.

Hypothesis 5a: Single-opinion dispositions, joint opinions, and decisions to join
will be less likely in cases with oral argument spanning more than one day.

Hypothesis 5 b: Single-opinion dispositions, joint opinions, and decisions to join
will be less likely in cases with higher average opinion length than in cases with
shorter opinions.

Hypothesis 50: When the Court combines a number of diflerent cases to consider
together, single-opinion dispositions, joint opinions, and decisions to join will be
less likely.

H )pothesis 5d: When the Court hears a case by special leave, single-opinion
dispositions, joint opinions, and decisions to join will be less likely.

Because the US. Supreme Court has a set length of time allotted for oral argument,
there is little variation duration of oral arguments. In the High Court, though, there are
no strict limits placed on oral arguments on the merits. Oral argument in most cases is
completed within a single day. As Justice Kirby recently pointed out, “[o]nly in the most
complex appeals will oral argument be permitted to stretch into a second day or firrther”
(Kirby 2006, 20). This institutional difference can be leveraged into a proxy measure of
case complexity. As such, this complexity measure for Hypothesis 2a is a dummy

variable coded as 1 when the oral arguments for a case span more than a day and 0

206

otherwise. Oral argument information is derived from the dates listed in the
Commonwealth Law Reports. In the period under investigation, about 46% Of cases
were argued for more than one day. The rate for unanimous cases is lower, at about 41%.
About 47% of opinion-writing opportunities followed long oral arguments.

The average length Of the Court’s opinions is another important measure Of the
complexity Of cases on the High Court’s docket (Campbell 2003; Groves and Smyth
2004). The average opinion length is a measure of the number Of pages the average
Opinion in a case spans in the Commonwealth Law Reports. This is calculated by divided
the number Of pages dedicated to the case by the number Of Opinions filed in that case.
The data in this analysis show an average Opinion length Of around 10 pages. When
nonunanimous cases are removed, the average Opinion length remains virtually
unchanged.

The two remaining measures of complexity are the number Of matters dealt with
and the Court’s control of its agenda. The number of matters variable is simply a count
Of the number of separate cases disposed Of in each single entry in the Commonwealth
Law Reports. For most cases, this number is one. The average number Of matters per
case (and per opinion-writing Opportunity) is 1.2 matters. The variable representing the
Court’s control over its own agenda is a dummy variable. It is coded 1 in situations
where the Court would have heard the case at its own discretion through the special leave
process.

Most criminal matters have always required special leave. As such, the agenda
control variable is 1 for all criminal cases. Many other appeals were as of right until the

Judiciary Amendment Act 1976 (Cth) began the process of restricting access to the Court

207

(Kirby 2007a). By 1984, appeals as Of right were essentially eliminated (Jackson 2001).
As such, non-criminal cases decided in 1984 or later are also coded as 1. The control Of
agenda variable is flagged in 83.6% of cases, but in only 50% of Opinion-writing
Opportunities.

Hypothesis 6: Joint opinions will be less likely in salient cases than in non-salient
cases.

As the previous chapter indicates, cases dealing with important legal or political
issues are expected to elicit fewer examples of collaborative behavior. One popular
measure for case salience in the American literature is the use Of a list Of landmark cases
(Epstein et al. 1996). This type Of measure captures case salience only with the benefit Of
hindsight, though, and what is important in the decision-making context is to capture the
judge’s view Of the case at the time (see Epstein and Segal 2000). Instead, measures Of
non-party participation and issue area salience are included in the analysis.

Hypothesis 6a: Single-opinion dispositions, joint opinions, and decisions to join
will be less likely when there is more non-party participation.

Hypothesis 6b: Single-opinion dispositions, joint opinions, and decisions to join
will be less likely when a non-governmental organization is a party to the case.

Hypothesis 60: Single-opinion dispositions, joint opinions, and decisions to join
will be less likely in constitutional cases than in other cases.

Hypothesis 6d: Single-opinion dispositions, joint opinions, and decisions to join
will be less likely in tort cases than in other cases.

While the use Of non-party interveners has been criticized in the American
literature for failing to capture salience to the judges themselves (Collins 2005), the
previous chapter illustrates that the High Court Justices play an important role in
allowing this sort of participation. The calculation Of non-party participation here

includes both amici curiae and the more common interveners (see Chapter 2). The

208

resulting measure is a count of the number Of non-party interveners who are granted leave
to participate in the case on the merits. Non-party participation rates are derived from the
Commonwealth Law Reports, where all interveners are listed. On average, there are 0.45
non-party participants per case. In the subset Of unanimous decisions, this number drops

to an average Of 0.37 parties. The average number per opinion-writing Opportunity is 0.6

parties.

The HCJD project includes information about the nature of the litigants appearing
before the High Court. TO measure the participation Of non-governmental organizations,
the private/nonprofit organization codes were used. This category includes business and
trade associations, unions, charitable organizations, political parties, and other political
groups. If any Of these parties is present, the non-governmental organization variable is
coded 1. This is the case in 7.5% Of cases and 8.2% Of opinion-writing Opportunities.

The third measure of salience is a dummy variable indicating whether a
constitutional issue was raised in the case. More specifically, this variable is coded as 1
when the Court is asked to review the constitutionality of a government law or action,
and 0 otherwise. This measure is derived from the text of the cases themselves. For the
period of study, a constitutional issue is raised in 18 % of cases, but in only 16% of
unanimous cases. Additionally, 22% Of opinion-writing Opportunities involved a case
posing a constitutional question.

As the previous chapter explains, tort cases have a different connotation in the High
Court than in the US. Supreme Court. These cases are politically salient in Australia
(Cane 2003), and are sometimes a vehicle through which human rights claims can be

litigated (Kyriakakis 2005). The measure for tort cases is a dummy variable coded as 1

209

when the main issue in the case is a tort claim, and 0 otherwise. Here, around 15% of
cases and 17% Of opinion-writing opportunities deal with torts.

Hypothesis 7: Joint opinions will be less likely when the Court is concerned about
the reaction of the other branches of government.

H )pothesis 7a: Single-opinion dispositions, joint opinions, and decisions to join
will be less likely when the federal government is a party to the litigation.

An indicator Of government participation can be created the same set of party type
variables as the indicator for non-governmental organizations. When the federal
government appears as a party to the case, the variable is coded 1. The federal
government was a party in nearly 40% Of cases in this analysis.

ijothesis 8: The source of the case will have an impact on the likelihood of joint
opinions.

The willingness Of judges tO assent to a single Opinion may be related to the
litigation history associated with the case. As the previous chapter explains, cases
coming from administrative or special tribunals might be expected to produce more
joining behavior than other cases. Additionally, cases from the federal jurisdiction are
expected to present questions that are more divisive to the Court.

Hypothesis 8a: Single-opinion dispositions, joint opinions, and decisions to join

will be more likely in cases appealed fiom administrative tribunals & specialized

courts than cases fiom other sources.

Hypothesis 8b: Single-opinion dispositions, joint opinions, and decisions to join

will be less likely in cases originating in the federal courts than cases fiom other

sources.

The variable representing administrative tribunals and specialized courts is coded 1
if the case was appealed directly from a body of this type, and 0 otherwise.

Administrative agencies and specialized courts may be at the state or the federal level,

and include bodies specializing in family law, industrial relations, environmental issues,

210

and various administrative appeal 8.97 In these data, around 8% Of the cases heard by the
High Court were appealed from an administrative tribunal or a specialized court.

Between 1965 and 2003, there have been several different courts in Australia
exercising federal original jurisdiction. The Federal Magistrates Court of Australia was
established by the Federal Magistrates Act 1999 (Cth), and it is classified by Haynie et al.
(2007) as a minor trial court in the federal system. Before 1999, the federal original
jurisdiction was exclusive to the Federal Court of Australia, established by the Federal
Court Of Australia Act 1976 (Cth). The Federal Court, in turn, consolidated the
jurisdiction of a number of other federal courts, including the Australian Industrial Court
and the Federal Court of Bankruptcy. Additionally, the Federal Court assumed some Of
the original jurisdiction previously exercised by the High Court Of Australia.

In the analyses tO follow, a dummy variable is included flagging those cases
originating in the federal jurisdiction. Cases tried in the original jurisdiction of the
Federal Magistrates Court, the Federal Court Of Australia, the Australian Industrial Court,
or the Federal Court of Bankruptcy are coded 1. Nearly 30% Of the cases in this analysis
originate in the federal jurisdiction. Of opinion-writing Opportunities, only 12.4% come
from the federal jurisdiction.

Outcome of the Case

Hypothesis 9: Characteristics of the outcome of the case will have an impact on the
likelihood of joint opinions.

In addition to the existing features of the case, the likelihood of a joint Opinion may
be affected by characteristics Of the disposition itself. The most Obvious variable Of

interest when discussing the outcome of the case is whether the vote on the case is

 

97 For a full list Of the institutions included in this category, see Appendix B.

211

unanimous. Additionally, we can control for both the ideological outcome Of the case
and the disposition in relation to the decision of the lower court.

ijothesis 9a: Joint opinions will be more likely in unanimous decisions than in
nonunanimous decisions.

Hypothesis 9b: Single-opinion dispositions and joint opinions will be less likely in
cases where the outcome is “liberal.”

Hypothesis 9c: A Justice is more likely to join when he agrees with the ideological
direction of the case outcome

Hypothesis 9d: Single-opinion dispositions, joint opinions, and decisions to join
will be less likely in cases where the decision of the court below is overturned

The measure of unanimity here is a dummy variable flagging cases with a
unanimous vote as to the orders. Of all cases in this dataset, 62.4% were unanimous.
When investigating single-Opini on dispositions, this variable is omitted because only
unanimous cases can be disposed of by a single opinion for the Court.98

The classification of case outcomes as liberal or conservative is “by now routine in
studies of judicial decision making” (Tate 1981b, 356). Indeed, much of the early
judicial behavior research explained the differences among judges in terms of these
ideological differences (Maveety 2003). Generally, researchers classify case outcomes as
liberal or conservative using the coding scheme developed by Harold Spaeth (1997). The
HCJD uses a similar approach. The liberal outcome variable in this analysis is derived
directly from this measure. Ifthe case outcome supports the economic underdog, the
criminal defendant, the federal government’s power to regulate, or other liberal outcomes

(see Appendix B), this variable is coded 1. In all other cases, it is coded as 0. In these

 

98 The unanimity variable is also left out of the judge-centered analyses because these analyses include a
variable that controls for the size the judge’s coalition. This variable is detailed in the section on control
variables below.

212

data, 42% of all case outcomes were liberal. Of the unanimous cases, 45% had liberal
outcomes.

Another outcome variable of interest is the treatment of the decision in the court
below. For this analysis, the treatment of the lower court’s decision is measured with a
dummy variable. In cases where the appellant wins—outright or in part—this variable is
coded as 1. Otherwise, this variable is coded as zero. The Court overturned the lower
court in 48% Of the cases and 47% Of the Opinion-writing Opportunities in this dataset.

Judge Characteristics

A series ofjudge-specific variables is also included in the judge-centered analyses.
These variables capture the relevant variations in social background, politics, and
previous experience. Schubert’s attempts to amass this kind of information on the High
Court Justices were largely frustrated by a lack of information (Schubert 1969a). Since
that time, the availability of background information on the Justices has improved
markedly, largely thanks to the efforts of the editors of The Oxford Companion to the
High Court of Australia (COper et al. 2002). The Justices themselves have roundly
praised the effort (McHugh 2002; Gleeson 2002c), and several current and former

Justices contributed to it. Much Of the background information comes from this volume,

supplemented by biographical information available through the High Court’s website.99

Approach to the Working Environment

Hypothesis 10: Justices with a collaborative approach to the working environment
will be more likely to join.

Of course, it is impossible to devise a direct measure of a collaborative approach to

judging. As the previous chapter outlines, several other characteristics may serve as

 

99 http://www.hcomt. gov.au

213

proxy measures for this concept. The first of these is experience as a judge and/or a
government attorney.

H )pothesis 10a: Justices with previous experience as a government attorney will
be more likely to join.

ijothesis 10b: Justices with prior judicial experience will be more likely to join.

ijothesis 10c: Justices will be more likely to join during their first year on the
Bench.

Hypothesis 10d: Justices who have developed a pattern of joining in previous
cases will be more likely to join.

The variable representing experience as a government attorney is a dummy variable
coded as 1 when the Justice has previously served as a government attorney. Experience
as a federal or state attomey-general or solicitor-general is included in this measure, but
lower-level prosecutors are not. Seven of the Justices in the sample have experience as a
government attorney, and nearly 40% Of Opinion-writing decisions were made by Justices

with this experience. For prior judicial experience, the dummy variable is coded 1 if the

Justice has any judicial experience at any level.100 Of the 26 judges in this analysis, 16
had at least some judicial experience prior to their elevation to the High Court. Almost
63% of opinion-writing decisions were made by judges with prior judicial experience.

Acclimation effects for judges have been measured in a number of different ways.
Some have considered a true freshman effect, including a flag for the first calendar year
of service (Ostberg et al. 2003). Others have advocated the use of a two year (Hettinger
et al. 2003a), three year (Smyth 2002a), or five year acclimation period (Hagle 1993).

For this analysis, the acclimation period variable is a dummy variable coded 1 when the

 

100 . . . . .
Other measurement strategies were also tested, rncludmg a vanable counting the number of years of

experience (Tate and Sittiwong 1989) as well as ordinal and dummy variables representing various lengths
of judicial service (Tate 1981b). None of these improved on the explanatory power of the variable.

214

Opinion-writing decision is one of his first 50. This corresponds roughly to a single year

on the Bench.101 Approximately 5% Of Opinion-writing decisions were made by Justices
in their freshman period.

The final measure of judicial approach is the presence of a pattern Of j oining
behavior. If a Justice has developed a pattern of j oining behavior in previous cases, that
Justice is more likely to continue this joining behavior. This measure is the average Of
the joining behavior from the previous five Opinion-writing opportunities. The variable

ranges from zero (indicating no joining behavior in the previous five cases) and 1

(indicating that the Justice has joined in each of the previous five cases).102 Interestingly,

seven of the twenty-six Justices never managed to string together five joint Opinions in a

row during the course of their tenure. '03 The average value of this measure is 0.43. In
other words, the average opinion-writing decision is made by a Justice who has joined in
2.15 of the previous five cases.

Social Background Characteristics

H )pothesis 11: Certain social background characteristics will be related to the
decision to join.

 

01 . . . . .
Several other measures were tested rn the model, including various calendar year dummies and a

dummy for the first 100 and 150 cases. None of these varied substantially from the results of the chosen
method.

102 Brace and Hall (l997)use a similar strategy for measuring the predisposition Of state supreme court
justices to support a death penalty outcome. They argue that this avoids circularity by using prior behavior
patterns to control for present decision-making choices.

103 These are Justices Jacobs, McTieman, Menzies, Murphy, and Walsh. By contrast, three Justices had
fewer than five 0-for-5 stretches: Chief Justice Gleeson and Justices Gummow and Toohey.

215

In addition to the career-related characteristics, several other social background
characteristics have been tied to judicial behavior in the literature. Following the
structure of the previous chapter, several of these are presented below.

ijothesis 1 1 a: Catholic Justices will be less likely to join than Justices with other
religious backgrounds will.

ijothesis 11 b: Justices fiom New South Wales will be less likely to join than will
Justices from other places.

H )pothesis 11c: Justices appointed by Labor Party Prime Ministers will be less
likely to join than will Liberal Party appointees.

Each of the remaining social background measures is a dummy variable coded 1
when the characteristic is present and 0 otherwise. The information used to create these
variables comes almost exclusively from the various biographical entries in the Oxford
Companion to the High Court of Australia (COper et al. 2002). Nine of the Justices have
a Catholic background, and 45% of the Opinion-writing decisions were made by Catholic
Justices. Fourteen Justices hail from New South Wales, and they are responsible for 58%
of the opinion-writing decisions. Finally, eight Justices in this sample were appointed by
Prime Ministers affiliated with the Australian Labor Party (ALP). These appointees were
responsible for 30% of the opinion-writing decisions.

Controlling for Panel and Coalition Size

When the panel size is larger, there are more joining possibilities for the
participating Justices. Even when the panel is large, joining opportunities are limited by
the number of Justices who agree on the outcome Of the case. At the same time securing
a single Opinion disposition for a case becomes more difficult the larger the panel size.

The more judges involved, the less likely they will all agree to a single set of reasons for

216

disposition. For different reasons, then, it is important to control for panel size in
analyses of both kinds of joint opinions.

This measure is not without substantive importance. The panel size is not
independent of the characteristics of the case. As in the case of the English Court of
Appeal the size of the panel can be an “especially useful indicator to case importance”
(Atkins 1991, 891). In Australia, the Chief Justice has a degree of flexibility in
determining both the size and composition of the panels assembled to hear cases. As
Chapter 2 reveals, the Chief Justice has a degree of flexibility in terms of both the size
and the composition of the panels. When the Chief believes that the case is important, he
will often assign all available Justices to the panel (Popple 2003). Even if the Chief
selects a smaller panel for a case, a judge who is not assigned to the panel may Opt to sit
on the case if he believes it to be important (McHugh 2005). In other words, salient cases

will have larger panels than other cases. Because of this, the average panel size has

grown over time.104 The average panel size is smaller in earlier years, most notably
during the tenure of Chief Justice Barwick.

As such, it is reasonable to expect that the inclusion of the panel size variable in a
model of j oining behavior may introduce multicollinearity. Multicollinearity is suspected
if the overall model explains a statistically significant amount of variance in the
dependent variable, but many Of the individual independent variables are not statistically
significant. Another warning sign is when the removal Of one statistically insignificant

variable from the model leads to a notable decrease in the model’s overall explanatory

 

104 Over time, the number Of cases falling under the jurisdiction Of a single Justice under the Judiciary Act

1903 (Cth) has fallen significantly. In the analyses presented here, though, cases heard by a single Justice
are excluded.

217

power. This happens because multicollinearity artificially inflates the variances of the
individual parameter estimates.

Multicollinearity is caused by a strong correlation between independent variables.
In some cases, these strong correlations are bivariate in nature. In other words,
sometimes the correlations between pairs of variables will be high, leading to the
multicollinearity problem. To check for this, bivariate correlations (or associations for
nominal-level data) are presented in Table 11. Each of the suspect independent variables
is included here, and bivariate statistics are presented for both control variables. While
there are significant associations between many of these bivariate pairs, none of these
pairs reveals extremely high levels of association. The highest association is between the
judge-level measure panel size and the dummy variable flagging constitutional cases. At

.498, this association is moderate.

Table 11 - Correlation Coefficients between Control and Independent Variables

 

Inrlc rentlr-nt Variable ('ontrol Variables

 

 

Panel (Case) Panel (OWO) Coalition
(OWO)
Salience
Nonparty Participants '1' .391 *** .404 '1‘" .189 ***
Constitutional Issue .463 *** .498 *** .228 ***
NGO Participation .069 *** .045 *** .043 ***
Tort Case .029 -.012 -.032 ***
Chief Justice
Barwick -.340 *** -.037 *** -.l99 ***
Gibbs .009 -.034 *** .034 ***
Mason .198 *** .198 *** .102 ***
Brennan .100 *** .079 *** .046 ***
Gleeson .200 *** .192 *** .095 ***
Panel * Coalition .489 *** .489 ***
‘1' The statistic presented here is Pearson’s r, as the nonparty participant variable is a continuous
variable. The other statistics presented are Spearman’s p, which is appgrriate for nominal-level data.

 

 

Sometimes, though, several variables may be involved in interdependent relationships. In

this case, the problematic relationships may not be obvious in the bivariate statistics. In

218

order to detect these relationships, the entire system of independent variables can be
evaluated for signs of multicollinearity. Using an OLS regression of the independent

variables on the dependent variables, the tolerance can be calculated by subtracting the
squared multiple correlation from one—or, 1-R2 (Miles and Shelven 2001). A tolerance

Of zero means that a variable is perfectly predicted by the other variables in the system.
The variance inflation factor (VIF) can be calculated for each variable in the system by
taking the inverse of the tolerance, providing a measure of how inflated the variance for
each regression coefficient actually is (Cohen et a1. 2003). These statistics, along with
the associated tolerance measures, are reported in Table 12..

Some researchers suggest that VIFs higher than two are indicative of collinearity
problems (Blaikie 2003). Others set the threshold at five (de Vaus 2002) or ten (Myers
1990). If high VIFs are found, researchers Often suggest dropping independent variables
from the analysis, using ridge regression, or collapsing the problematic independent
variables into a single index. This single-minded focus on multicollinearity may be
unwise, and less drastic approach to increase the stability of models is desirable (O'Brien
2007). In the absence of pure multicollinearity, increasing the sample size can reduce the
detrimental effects of collinearity on the variance of the regression coefficients
(Goldberger 1991). O’Brien demonstrates that in some circumstances, a regression with
a VIF as high as 40 ought not be dismissed out of hand (O'Brien 2007).

As Table 12 shows, the highest VIFs are in the case-level models. The panel
liberalism variable has a VIF of 10, while the panel diversity variable has a VIP of a little

more than 14. By contrast, the judge-level panel diversity variable has a VIF of 2.

Certainly, the additional data affects the R2 for each variable in the diagnostic regression

219

Table 12 - Collinearity DiagnosticsT

 

 

Case-Level Analysis Judge-Level Analysis
Independent Variables
Tolerance VIF Tolerance VIF

Workload

Late Oral Argument .991 1.009 .984 1.016

Retirement Near .460 2.176 .437 2.290

Cases This Term .501 1.996 .463 2.158
Chief Justice

Barwick --- --- --- ---

Gibbs .510 1.961 .368 2.717

Mason .315 3.172 .148 6.743

Brennan .449 2.227 .255 3.916

Gleeson .134 7.451 .161 6.204

New Chief Justice - - .767 1.304
Ideology Measures

Panel Liberalism Score .100 10.050 - - - - - -

Panel Diversity .070 14.235 .468 2.137

Distance from Panel Mean - - - - - - .495 2.020

Agree with Outcome Direction - - - - - - .587 1.703
Complexity

Long Oral Argument .843 1.187 .817 1.224

Avg. Opinion Length .694 1.441 .677 1.477

Number of Matters .933 1.072 .931 1.074

No Leave as of Right .610 1.640 .144 6.921
Salience

Nonparty Participants .701 1.427 .701 1.427

NGOasaParty .864 1.157 .877 1.141

Constitutional Issue .625 1 .601 .625 1 .599

Tort Cases .846 1.182 .842 1.188
Fed. Gov’t as a Party .793 1.261 .750 1.333
Litigation History

Admin/Specialized .793 1.261 .851 1.175

Federal Jurisdiction .566 1.766 .830 1.205
Case Outcome Characteristics

Nonunanimous Vote .928 1.078 - - - - - -

Liberal Outcome .907 1.102 - - - - - -

Allow Appeal .933 1.072 .923 1.083
Judge Characteristics

Gov’t Attorney Experience - - - - - - .554 1.806

Prior Judicial Experience - - - - - - .436 2.295

Freshman (1“ 50 Cases) - - - - - - .874 1.144

Predisposition to Join - - - - - - .741 1.349

Catholic - - — - - - .482 2.076

From New South Wales - - - - - - .590 1.694

Labor Appointing P.M. - - - - - - .595 1.694
Control Variables

Panel Size .601 1.665 .438 2.281

Coalition Size — - - - - - .450 2.225
1 The statistics reported here are tolerance scores and variance inflation factors (VIFs) for the full
models. These are calculated using an OLS regression where the dependent variable is a dichotomous
flag indicating the presence of anyjoint Opinion.

 

 

220

 

of each independent variable on all the Others, allowing for less inflation of the variance
associated with the estimated parameters.

The most Obvious problem associated with the high VlFs is the inability to reject
the null hypothesis in the models (O'Brien 2007). Parameter estimates can also be
volatile because the regression equation has no mechanism to allocate the parameter
estimates among the collinear independent variables (Miles and Shelven 2001).
Essentially, post hoc assessment of multicollinearity is necessary, as the diagnostics
“need to be interpreted in the context of other factors that influence the stability of the
estimates” (O'Brien 2007), including the stability of parameter estimates across different
specifications and the associated size of the confidence intervals. As such, the
specification of the models will include the controls and the other suspect measures until

their effects on the model can be assessed.

221

CHAPTER EIGHT: ANALYZING JOINING BEHAVIOR ON THE HIGH COURT

If the real process by which judges reach decisions is unveiled the effect will be
that the courts will be more accountable for their choices.

— Justice Michael McHugh, as quoted in Galligan (1995)

The previous two chapters have provided conceptual and Operational hypotheses
aimed at explaining joining behavior on the High Court of Australia. This chapter
presents several analyses, using logistic regressions to test the eleven operational
hypotheses set out in Chapter 7. Separate analyses are presented for joint opinions and
the special case of the single-Opinion disposition. The decision to join is evaluated at the
case level as well as at the judge level. Finally, a model of joining behavior on the
Gleeson Court is assembled and tested, using a multidimensional scaling analysis to
provide a spatial representation of j oining behavior on that Court.

Methodology

The dependent variables in these analyses are the presence or absence of a single
Opinion for the Court, the presence of any joint opinion, or a judge’s decision to join an
opinion. Because all of these are dichotomous variables, ordinary least squares
regression (OLS) is inappropriate (Aldrich and Nelson 1984; Maddala 1983; but see
Hellevik 2007). The resulting linear probability model (LPM) typically suffers from non-
norrnal error terms, heteroscedasticity, and estimates of conditional probabilities outside
the necessary range of 0 and 1. While several strategies have been proposed for
circumventing this issue (see Aldrich and Nelson 1984), most of these are much more
difficult as compared with the use of the maximum likelihood estimation of a nonlinear

probability model (Harrell 2001).

222

A binary dependent variable yi is distributed as a Bernoulli distribution, where the
possible values are 0 or 1. The probability that yi is equal to one, given the vector of
independent variable(s) xi can be described well using any cumulative distribution
function because they are all bounded by 0 and 1. This approach works nicely for
determining the odds that a binary variable takes one or the other value, as “it seems
likely that the effect of a unit change in the independent variable on the predicted
probability would be smaller near the floor or the ceiling than near the middle” (Pampel
2000,5)

This link function allows for the transformation ofyi with a Bernoulli distribution

to the scale of the linear predictor, while maintaining the nonlinear relationship between
the likelihood of the dependent variable of interest and the independent variable. Binary

logistic regression uses the logit firnction as a link function. This links the log odds of

observed outcome yi to the vector Of covariates xi. Instead of using the binary variable as

yi, the dependent variable is the log odds that yi =1. The likelihood function is the

product of all of the individual likelihoods. Taking the natural log of each side of this
equation renders the model linear in its logarithms, or log-linear (Knoke and Burke
1980)

This yields a direct probability model, and makes no assumptions about the

characteristics of the distribution OfP(Vi=1|xi) (Harrell 2001). The logit model has

“grown to become the de facto standard means for estimating regression-like models of

binary outcomes” (Zorn 2005, 157). Because of the slightly s—shaped curve of the logit

223

Table 13 - Case-Level Logistic Regression Predicting Single-Opinion Dispositions

\lodel listimates All ('ases (l) l‘naninrous ('ases (2)

 

 

 

l'iull Reduced 15ull Reduced
onstant 1.161 (.667) 1.555 (.329) *** 1.050(.723) 1.079 (.350) **
orkload
Late Oral Argument -.048 (.122) -.109 (.135)
Retirement Near .047 (.217) -.010 (.247)
Cases This Term .00] (.007)
hief Justice'r 23.344 *** 44.073 1‘" 22.573 *** 42.602 ***
Gibbs .591 (.221) ** .541 (. 170) *** .432 (.245) .450 (.190) *
Mason .774 (.265) ** .716 (.177) "'l' .657 (.299) * .616 (.205) **
Brennan -.316 (.372) -.380 (.287) -.742 (.414) -.788 (.329) *
Gleeson -.176 (.510) -.473 (.255) -.657 (.584) -.759 (.291) **
New Chief .080 (.181) .061 (.198)
deology Measures
Panel Liberalism -l .662 (1.669) -.696 (1.839)
Panel Diversity .384 (1.852) .349 (2.077)
omplexity
Long Oral Argument -.975 (.141) *** -.928 (. 138) *** -.999 (.160) *** -.970 (.157)***
Avg. Opinion Length 0.98 (.011)*** .095 (.010) *** .125 (.014) *** .124 (.014) ***
Number of Cases -.395 (.136) ** -.405 (.134) ** -.395 (.153)** -.390 (.150) **
No Appeal as of Right .221(.218) .264 (.237)
Salience
Nonparty Participation -.058 (.061) -.055 (.065)
Constitutional Issue .208 (.216) .088 (.233)
Tort Cases -.285 (.186) -.188 (.205)
Federal Gov’t a Party .139 (.137) .313 (.154)* .365 (.137)"
NGO a Party .370 (.224) .458 (.208) * .321 (.252) *
itigation History
Admin/Specialized .298 (.244) .110 (.273)
Federal Courts .028 (.174) .115 (.194)
ase Outcome
Liberal Outcome .203 (.126) .047 (.142)
Allow Appeal -.304 (.124) ** -.280 (.121) * -.255 (.139)
ontrol Variables
Panel Size , -.647 (.073) *** -.657 (.066) '1'" -.514 (.079) *** -1.079 **
.350

 

   
 
           

 

 

Omnibus Test Of Coeff. 1 299.356 *** 287.026 *** 237.026 *** 227.462 ***
Cox & Snell R-Square .130 .125 .170 .163
Nagelkerke R-Square .209 .201 .238 .229
Hosmer & Lemeshow Test 1 4.976 10.135 5.120 5.317
Correctly Predicted 82.6 % 82.4 % 73.7 % 73.2 %
n= 2144 2144 1275 1275
5.05, ** p301, *** p500]

This is a categorical variable. The Wald statistic is provided. The indicator group is Barwick CJ.
'9
The numbers presented here are test statistics distributed as chi“.

 

224

distribution, this model has the additional benefit of weighing more heavily the effect of
inputs near the center of the distribution. In other words, “increasingly larger inputs are
needed to have the same impact on the outcome near the ceiling or floor” (Pampel 2000,
6).

Models of Single-Opinion Dispositions

In the first pair of analyses, the dependent variable is the presence of a single-
Opinion disposition. Here, the dependent variable is coded 1 when the entire panel has
agreed to a single set of reasons. The results are presented in. First, the model was
applied to all cases decided by more than one Justice. The second application is to the
subset of cases with a unanimous vote on the merits. As expected, the panel size control
has a significant negative relationship with the probability of a single-Opinion disposition
in both models. As the panel size increases, the probability that all of the Justices will
agree to a single set of reasons decreases substantially. This relationship is presented
graphically in Figure 13, with separate lines for each Chief’s tenure.

The other hypothesis garnering strong support in these analyses is Hypothesis 5.
Three of the four case complexity measures are significant in both applications of the
model. Longer oral arguments and more matters at issue both decrease the likelihood of
a single-opinion disposition. The average Opinion length, however, has a consistently
positive relationship with the likelihood of a single-opinion disposition.

The average Opinion length for this analysis is 10.13 pages, with a standard
deviation of 8.05 pages. Ifwe move from one standard deviation below the mean to one
standard deviation above the mean, the predicted probability of a single-opinion
disposition increases from 0.07 to .029. This relationship may be a function of

bargaining and accommodation behavior. The length of US. Supreme Court majority

225

Figure 13 - Predicted Probabilities of a Single-Opinion Disposition by Panel Size

 

 

+ Barwick + Gibbs + Mason
+ Brennan + Gleeson

 

 

 

1.0

0.8 k

k

0.4

2.0 3.0 4.0 5.0 6.0 7.0

 

 

 

 

 

Predicted Probability

 

 

 

4'-
1

 

 

i
J
Opinions has been tied to bargaining activity, even after controlling for other measures of
case complexity (Spriggs 2008). In order to accommodate the views of more Justices, the
opinion may well need to be longer. This, of course, assumes that Justices are willing to
engage in bargaining behavior in the absence of an institutional drive to do so.
Unfortunately, no comparable indicators of bargaining are currently available for the
High Court.

A simpler explanation may be that this relationship is due to the nature of the

single-Opinion itself. Before the Dixon Court, the Justices did not make a habit Of
circulating draft opinions during the opinion-writing process (Fricke 2002b). While

Barwick was less successful at assembling conferences or opinion coalitions, the habit of

226

draft circulation that began in the Dixon Court largely survived (Marr 1980). As such,
the Justices are rarely unaware of the content of their colleagues’ opinions (McHugh
2005). Ifone Justice has included a discussion of the factual background of the case, for
instance, the remaining Justices may Omit this lengthy discussion in their own opinions.
The longer opinion will be averaged in with several shorter opinions. If only one opinion
is handed down in a case, that opinion will have to review the background information.
The pages dedicated to the factual background will not be averaged across several
opinions.

The performance of the salience measures is also counter to expectations.
Hypothesis 6 suggests that salient cases will be less likely to have a single-opinion
disposition. These models provide no support for this hypothesis. In the model tested on
all cases, the presence of a non-governmental organization as a litigant is associated with
a greater likelihood of a single-opinion disposition. When all other variables are held at
their mean and all cases are considered, the presence of an NGO as a litigant increases
the predicted probability of a single-opinion disposition from .14 to .21.

Case-Level Models of Joining Behavior

The next set of analyses uses the case-level data to construct models Of j oining
behavior. Here, the dependent variable is the presence of any joint opinion. First, the
model is applied to all cases decided by more than one Justice. Then, the model is
applied to criminal cases only. The results of these models are presented in Table 14.
Unlike the single-opinion disposition, an expanded panel size increases the opportunity
for a joint Opinion. Here, the panel size control is positively related to the likelihood of a

joint Opinion. Figure 14 demonstrates this, with separate curves for each Chief Justice.

227

Table 14 - Case-Level Logistic Regression Predicting Joining Behavior

 

 

 

 

 

 

 

Model Estimates All (-‘ases (3) ('riminal (‘ases (4)
Full Reduced Full Reduced
onstant 1.064 (.929) .924 (.412) * 3.987 (1.639) * 2.460 (1.094)*
Workload
Late Oral Argument -.073 (.122) .018 (.302)
Retirement Near .395 (.231) 1.283 (.646) * 1.239 (.629) *
Cases This Term -.002 (.007) -.019 (.017)
lChiefJusticeT 22.622 in 30.260 *** 9.429 * 7.578
Gibbs -.693 (.590) -.345 (. 165) * .268 (.585) .277 (.411)
Mason -.429 (.605) .802 (.249) *** .152 (.799) .488 (.454)
Brennan .656 (.672) -.302 (.278) -l.733 (.933) -1.226 (.681)
Gleeson -.631 (.673) .515 (.446) .366 (1.578) .355 (1.356)
New Chief -.113 (.201) .565 (.517)
deology Measures
Panel Liberalism 3.640 (1.501)” 3.442 (1.405) ** 12.494 (4.615) ** 13.294 (4.399) **
Panel Diversity -4.202 (2.119)* -4.095 (2.038) * -l3. 140 (7.209) -12.757 (6.962)
omplexity
Long Oral Argument -.253 (.130) * -.259 (.128)* -.668 (.318) "' -.727 (.293)**
Avg. Opinion Length .039 (.013) ** .032 (.013) ** -.026 (.033)
Number of Cases -.095 (.072) - - -
NO Appeal as ofRight .391 (.188)* --- ---
Salience
Nonparty Participation -.088 (.056) .012 (.145)
NGO a Party .150 (.241) -.612 (1.479)
Constitutional Issue -.772 (.189) *** -.856 (.166) *** -.564 (.408) -.736 (.354)*
Tort Cases -.184 (.172) --- ---
ederal Gov’t a Party -.152 (.143) .249 (.341)
itigation History
Admin/Specialized .111 (.244) - - -
Federal Courts -.035 (. 186) - - -
ase Outcome
Liberal Outcome .006 (.126) -.121 (.362)
Allow Appeal -.205 (.123) -.144 (.376)
Nonunanimous Vote -1.144 (.128) *** -1.151 (.126) *** -.890 (.298) ** -.888 (.292) **
ontrol Variables
Panel Size .329 (.329) *** .327 (.069) *** .039 (.187) .051 (.177)
Criminal Outcomes
Allow Jury Challenge .650 (.591)
Exclude Confession 4.981 (11.542)
Omnibus Test of Coeff. 1 328.007 *** 308.920 *** 91.004 *** 84.415 ***
Cox & Snell R-Square .142 .137 .179 .167
Nagelkerke R-Square .227 .214 .300 .280
Hosmer & Lemeshow Test 1: 2.280 5.808 3.239 11.724
Correctly Predicted 82.1 % 81.9% 85.7 % 85.2 %
n = 2148 2148 461 461
l" p505, ** p501, *** pS.001
This is a categorical variable. The Wald statistic is provided. The indicator group is Barwick CJ.
The numbers presented here are test statistics distributed as chiz.

 

 

228

Like the single-Opinion disposition model, Hypothesis 1 regarding workload
pressures is not supported. While joint Opinions are less likely under Chief Justice Gibbs,
the predicted probability for joint opinions increases from .82 to .95 in the Mason Court.
The complexity measures also mirror the singl e-opinion disposition model. While the
number of matters considered is not significant here, joint Opinions are less likely when

the Court entertains a long oral argument.

Figure 14 - Predicted Probabilities of Joint Opinions by Panel Size

 

 

+ Barwick —l-— Gibbs + Mason i
+ Brennan + Gleeson

 

 

 

 

 

 

 

 

 

 

 

 

 

0.4

 

Predicted Probability

0.2

 

 

 

0.0 i i . T
2.0 3.0 4.0 5.0 6.0 7.0

 

 

 

 

There are some notable differences between the single-opinion disposition model
and the joint Opinion model. First, joint opinions are less likely in nonunanimous cases.
This variable was omitted from the single-Opinion disposition models, as this kind of

disposition is not possible except in unanimous cases. Here, the probability of a joint

229

opinion is .91 in unanimous cases, but only .75 when there is at least one dissenting vote.
In addition, constitutional cases are less likely to garner a joint Opinion. The probability
drops from .88 to .75 when a constitutional question is raised.

Perhaps the most interesting difference between the single-opinion and joint
opinion models is the role of the ideology measures. The panel liberalism score is
positively related to joint opinion likelihood. In other words, as the panel mean
liberalism is raised, joint Opinions become more likely. This is illustrated in Figure 15,
with separate curves given for constitutional and non-constitutional cases. This gives

support to Hypothesis 3a.

Figure 15 - Predicted Probabilities of a Joint Opinion by Panel Liberalism

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

+ Constitutional —-I— Non-Constitutional
1.0
0.8 ///
g
25
3 0.6 ’7—
9
a.
E
g 0.4
Q. :
I
l
0.2 i
0.0 T 1 I l T— l
-0.1 0.0 0.1 0.2 0.3 0.4 0.5

 

 

 

230

At the same time, higher levels of ideological diversity make joint Opinions less
likely. As Figure 16 demonstrates, this is especially pronounced in constitutional cases.
In keeping with the prediction in Hypothesis 3b, the Justices become less likely to find a
partner with whom they agree with enough to author a joint opinion when ideological
diversity increases. Therefore, while panel liberalism and diversity do not significantly
affect the likelihood of single-opinion dispositions, they do influence the production of

joint opinions made up of smaller coalitions.

Figure 16 - Predicted Probabilities of a Joint Opinion by Panel Diversity

 

 

+ Constitutional + Non-Constitutional

 

 

 

1.0

 

0.8

 

0.6

\\ \-
o.4 \

0.2

 

 

Predicted Probability

 

 

 

0.0 I T l I T l
-0.1 0.0 0.1 0.2 0.3 0.4 0.5

 

 

 

 

The joint Opinion model was also applied to the subset of criminal cases. The
effect of Chief Justice on the likelihood of joint opinions is insignificant in criminal

cases. The Justices are more likely to author joint Opinions when a retirement date is

231

approaching. The probability that a joint Opinion be produced in a case increases from
.84 to .95 if the case is handed down within three months of a scheduled retirement. This
provides support for Hypothesis 1b. Additionally, criminal cases with long oral
arguments are less likely to produce a joint opinion. When oral arguments last more than

a single day, the predicted probability of a joint opinion drops from .91 to .83.

Figure 17 - Predicted Probabilities of a Joint Opinion in Criminal Cases by Panel Liberalism

 

+ Constitutional + Non-Constitutional

 

 

 

 

1.0 I

 

 

0.8

 

0.6

/

0.2 /

 

 

Predicted Probability

 

 

 

 

0.0 . _T l l l l l
-0.1 0.0 0.1 0.2 0.3 0.4 0.5

 

 

 

While the panel diversity measure is insignificant in the model, liberal panels are
still more likely to produce a joint Opinion. This effect is more pronounced in

constitutional cases, as Figure 17 illustrates. On conservative panels, the difference in

232

Table 15 — Judge-Level Logistic Regression Predicting Joining Behavior
All 0“ ()s (5)

\lodel Estimates (‘riminal Cases (6)

 

lirill Reduced liull Reduced

Constant -.185 (.256) -.364 (.145) ** -.897 (.725) -.839 (.347)*

Workload
Late Oral Argument .017 (.045) .029 (.103)

Retirement Near .044 (.081) .048 (.194)
Cases This Term -.003 (.003) -.003 (.007)

Chief Justicel 29.369 *** 35.677 *** 14.189 ** 17.666 ***
Gibbs .181 (.192) * .122 (.083) .144 (.521) .113 (.203)
Mason .501 (.139) *** .459 (.127) *** .276 (.317) .155 (.282)
Brennan .108 (.168) .078 (.149) -.547 (.381) -.641 (.316) *
Gleeson .010 (.171) .020 (.138) -.171 (.391) -.088 (.296)
New Chief -.014 (.074) .193 (.161)

IIdeology Measures
Panel Diversity .830 (.857) -1.567 (1.277)

Panel Mean Distance -2.419 (.555) *** -2.l49 (.482) *** -l.567 (1.277) -l.847 (.621)**

 

Agree with Direction .202 (.043) *** .197 (041) *** .508 (.105) *** .503 (.104) ***
Complexity
Long Oral Argument -.360 (.048) *** -.367 (.048) *** -.612 (. l 14) *** -.632 (.110)***
Avg. Opinion Length .053 (.005) *** .058 (.005) *** .045 (.012) *** .040 (.011)***
Number of Cases -.097 (.028) *** -.099 (.028) *** -.153 (.082)
Agenda Control .265 (.113)* .268 (.104) ** - - - - - -
Salience
Nonparty Participation -.049 (.017) *** -.047 (.017) ** -.038 (.046)
Constitutional Issue -.400 (.066) *** -.390 (.066) *** -. 163 (.155) -.281 (.140) *
Tort Cases -.l96 (.063) *"‘ -.168 (.060) ** - - -
NGO a Party .174 (.086) * .158 (.082) * -.418 (.620)
[ederal Gov’t a Party -.105 (.075) .376 (.123)** .276 (.1 17) *
itigation History
Admin/Specialized -.024 (.088) -.153 (.411)
Federal Courts -.105 (.075) -.068 (.218)
Case Outcome
Allow Appeal -.102 (.045) * -.109 (.044) ** .214 (.111)*
llustice Characteristics
Catholic -.030 (.063) -.246 (.137)
Labor Appointing PM. -.201 (.063) *** -.195 (.059) *** -.196 (.128)
From New South Wales .057 (.057) -.324 (.128) -.353 (.124) **
Gov’t Attorney Exp. .494 (.060) *** .516 (.053) *** .706 (.134) *** .653 (.128) ***
Prior Judicial Exp. .251 (.066) *** .274 (.057) *** .487 (. 161 ) ** .455 (. 133) ***
Predisposition to Join .651 (.095) *** .666 (.094) *** .854 (.209) *** .892 (.203) ***
Freshman (1't 50 Cases) -.078 (.103) -.173 (.219)
Control Variables
Panel Size -.483 (.032) *** -.480 (.031)*** -.447 (.077) *** -.458 (.073)***
Coalition Size .530 (.023) *** .531 (.022) *** .449 (.056) **"' .447 (.055) ***
Criminal Outcomes
Allow Jury Challenge -.476 (.161)** -.307 (.146) *
Exclude Confession -.780 (.339) * -.645 (.325) *

 

 

 

233

Table 15 (eont’d)

 

 

\lodel Summary All Cases (5) Criminal Cases (6)
Full Reduced liull Reduced
Omnibus Test Of Coeff. I 2403.582 *** 2394.600 *** 684.900 *** 666.699 ***
Cox & Snell R-Square .199 .198 .246 .240
Nagelkerke R-Square .267 .266 .331 .323
Hosmer & Lemeshow Test I 3.225 5.731 7.486 5.635
Correctly Predicted 69.8 % 69.8% 73.7 % 73.6 %

n = 10855 10855 2425 2425

*ps.05, ** p501, *** p500]

 

This is a categorical variable. The Wald statistic is provided. The indicator group is Barwick CJ.
The numbers presented here are test statistics distributed as chiz.

 

predicted probability between constitutional and non-constitutional cases is pronounced.
As the panel liberalism score approaches its maximum Of .49, the difference in the
predicted probability Of a joint opinion between constitutional and non-constitutional
cases shrinks as they both draw near 1.00.

J udge-Level Models of Joining Behavior

The third set of analyses applies the model of j Oining behavior—with the addition
Of several judge-centered control variables—on the judge-level data. The first models
each opinion-writing opportunity in the dataset, and the second isolates criminal cases.
The results are reported in Table 15. In addition to the panel size control variable, as the
size of the individual Justice’s coalition is also included. As the coalition size get larger,
the Justice is more likely to join. Increasing a coalition from two to four
raises the probability that a Justice will join from .28 to .53. Adding three more members
to the coalition boosts the probability to .85. The coalition size variable essentially
controls for the increased opportunity to join in larger vote coalitions. Because of this,
the panel size variable now has a negative sign. Once the size of the coalition is

controlled for, the panel size variable likely reflects the relationship between panel size

234

and case importance. As the panel size increases, the predicted probability of a decision
to join decreases (see Figure 18).

Perhaps the most Obvious difference between these models and the case-level
models is the number of null hypotheses that have been rejected. Despite the similarities
in overall model statistics, the increased volume of data allowed this model to overcome
the problem of micronumerocity (Goldberger 1991). When additional data are added in
an environment of incomplete multicollinearity, the detrimental effects of lower-level
multicollinearity on the parameter variances are lessened (O'Brien 2007). In this way, the
regression model is better able to assign appropriately-signed parameter estimates,

yielding more power in hypothesis testing (Miles and Shelven 2001). This is reflected in

Figure 18 - Predicted Probabilities of a Decision to Join by Panel Size

 

 

+ Constitutional + Non-Constitutional

 

 

 

1.0

 

0.8

 

 

0.6

 

0.4

 

Predcted Probability

0.2

 

 

 

0.0

 

1.0 2.0 3.0 4.0 5.0 6.0 7.0

 

 

 

235

Table 12, where the judge-level VIFs tend to be much smaller than the same measures in
the case-level data.

Even in this environment of increased hypothesis-testing power, Hypothesis 1 fares
badly. Neither model provides support for the argument that workload pressures
discourage (or encourage) joining behavior on the High Court. In addition, the
suggestion offered in Hypothesis 8—that the litigation history is important—remains
unsupported. In the Mason Court, the predicted probability of a decision to join is .90, as
compared to .87 during the tenure of the other Chiefs. With this exception, the Chief
Justice variables are insignificant here. This may be a result of the inclusion of the
Justice characteristics. Because the membership of the Court is generally stable over the
course of a Chiefs tenure, the relative distribution of particular Justice characteristics
may explain much of the variation that the Chief Justice variables would have
accumulated in the case-level analyses.

The ideology measures perform quite strongly in this analysis. The panel diversity
measure is insignificant in these models, but distance Of the individual Justice from the
panel mean is significant and large in the expected direction. As the panel mean moves
farther from the Justice’s liberalism score, she is less likely to join. Figure 19 illustrates
that this relationship holds for constitutional and non-constitutional cases alike, although
Justices at any distance from the panel mean less likely to join when a constitutional issue
is raised. In much the same way, Justices who agree with the ideological direction Of the

decision are more likely to join than those who are not.

236

Figure 19 - Predicted Probabilities of a Decision to Join by Ideological Distance

 

 

+ Constitutional + Non-Constitutional

 

 

 

 

1.0

 

.0
co

 

 

i

 

 

 

 

Predicted Probability
.0
A

 

0.2

 

 

 

0.0 . r . i r
0.0 0.1 0.1 0.2 0.2 0.3 0.3

 

 

All of the measures of case complexity are statistically significant, but they do not
universally support Hypothesis 5. Justices are less likely to join when the oral arguments
lasted more than one day, with the predicted probability falling from .63 to .54.
Similarly, cases dealing with more than one matter have less joining behavior. A case
with a single matter at issue has a predicted probability of .60, whereas a case with five
matters has a predicted probability of .50. At ten cases—which is less than the sample
maximum of 15—the probability falls to .38.

Like the case-level models, the average opinion length has a positive relationship
with the probability of j oining behavior. As the earlier discussion suggests, this result

may be a result of the additional space per Opinion taken up by case facts and background

237

information. The parameter estimate for the agenda control variable also has a positive
sign. When the Court is exercising its discretionary jurisdiction, Justices are more likely
to join. This is counter to the prediction of Hypothesis 5d. This hypothesis suggests that
agenda control will yield a greater proportion of complex cases, and that Justices will be
less likely to join in complex cases. It is difficult to say which portion of this argument is

incorrect.

Figure 20 - Predicted Probabilities of Joining Behavior by Nonparty Participation Rate

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

+NGO Party +No NGO Party

1.0

0.8
5 ;
25 0 6 ¥ - g
3 I
2
O.
E 0 4
o u
E
O.

0.2

0.0 a T . . . T, . I

0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 I

 

Hypothesis 6 predicts that Justices will be less likely to join in salient cases; this
hypothesis is strongly supported in the judge-level analysis. Constitutional cases, tort
cases, and cases with nonparty participation are all associated with lower joining rates.

Constitutional cases have a predicted probability of joining behavior of .51 as compared

238

with .61 non-constitutional cases. This relationship is demonstrated in Figure 18 and
Figure 19. The difference between tort cases and non-tort cases is less pronounced, with
the predicted probabilities at .56 and .60, respectively. Higher rates of nonparty
participation are also associated with lower rates of j oining behavior. At the same time,
the presence of a nongovernmental organization as a litigant is associated with an
increased probability of j oining behavior. Figure 20 illustrates this relationship.

In addition to the improved success of these measures, several of the Justice
characteristics are also related to the decision to join. Hypothesis 10 predicts that a
Justice’s approach to the working environment will be related to higher joining rates.

Previous experience as a government attorney or a judge is associated with a greater

Figure 21 - Predicted Probabilities of Decision to Join across Approach to Working Environment

 

7

 

...... GoV‘t Atty. +Jud. Exp. —*—Joining Patte

 

 

 

 

1.0

 

 

 

Predcted Probability

 

 

 

 

0.0 T 1 . r r
1.0 2.0 3.0 4.0 5.0 6.0 7.0

 

 

 

 

239

likelihood that a Justice will join. After a Justice has developed a pattern of j oining, she
is more likely to continue that pattern. These three factors affect the predicted probability
of a decision to join in very similar ways, as Figure 21 shows. Of the other judge
characteristics, only the party of the appointing prime minister is significant in the model.
The direction of this relationship is rather perplexing. When all other variables are held
at their means, Liberal National Coalition (LNC) appointees have a predicted probability
of .60, while Australian Labor Party (ALP) appointees have a predicted probability of
.56. This is not a remarkable difference, but it is statistically significant. In the case-
level models, liberal panels are more likely to yield a joint opinion. Additionally, the
discussion of Operational Hypothesis 3 demonstrates the relationship between liberalism
and ALP appointee status (see Figure 12).

This model was also applied to the subset opinion-writing opportunities in criminal
cases. The results are much the same as the full model, with a few notable exceptions.
Joining behavior in criminal cases is not more likely in the Mason Court. Labor Party
appointee status is insignificant in this model, but Justices from New South Wales are
less likely than other Justices to join in criminal cases.

When the federal government is a party to the case, the Justices are less likely to
join. Hypothesis 7a predicted the opposite result, but there is reason to suspect that
criminal cases may be different. The theory behind Hypothesis 7 is that the Court will be
more interested in presenting a united front when it is concerned about the reaction of
other branches of government. In criminal cases, the federal government is generally a

party only as the prosecutor of serious crimes. In these situations, the federal party to the

240

case is usually the Queen,103 the cases involve questions of law enforcement as opposed
to questions of intergovernmental balances of power.

Joining Behavior on the Gleeson Court

The final application of the model is a case study of joining behavior on the
Gleeson Court through the retirement of Justice Mary Gaudron in 2003. The results of
this model are reported in Table 16, and they reflect many of the same patterns as the
earlier models. The panel and coalition size controls work the same way as they do in the
other judge-centered models. A larger coalition size is associated with higher predicted

probabilities, while a larger panel size is associated with less joining behavior. The

workload hypothesis is still wholly unsupported here.106 The ideological hypotheses are
also supported. As Figure 22 shows, a Justice’s distance from the panel mean is a
powerfirl predictor of his decision to join.

Experience as a government attorney and a predisposition to join are both
associated with higher predicted probabilities of joining behavior. Because Justice
Gaudron is the only member of the G1 eeson Court with this experience, this coefficient
will capture all her idiosyncrasies. As before, appointment by a Labor Party prime
minister is associated with lower predicted probabilities. A freshman effect is evident in
this model, as well. Because the only freshmen in this sample are Chief Justice Gleeson
and Justice Callinan, this variable may be capturing other similarities between these

Justices.

 

105 . . . . . .
The Queen as a corporate entity rs the nominal prosecutor of serious cnmes. The actual prosecutor rs

usually the Commonwealth Attorney-General. Criminal cases not prosecuted by the Queen are generally
less serious, and the cases are usually handled by the State Director of Public Prosecutions or the State
Attomey-General.

O6 . . . . . . . .
The retrrement flag has been omitted from thrs analysrs, as thrs vanable rs a constant for the period.

241

Table 16 - Judge-Level Logistic Regression Predicting Joining Behavior on the Gleeson Court

 

 

 

 

 

 

 

\lodel Estimates J udge-l .cvcl (7)
1‘ul1 Reduced
onstant -.406 (.979) .678 (.532)
orkload
Late Oral Argument —.037 (.140)
Cases This Term .002 (.011)
Ideology Measures
Panel Diversity 7.879 (7.470)
Panel Mean Distance -24.126 (3.486) *** -23.657(2.76l) ***
Agree with Direction .427 (.134) *** .491 (.124) ***
omplexity
Long Oral Argument -.158 (.169)
Avg. Opinion Length .048 (.013) *** .027 (.008) **
Number of Cases -.078 (.103)
Salience
Nonparty Participation -.069 (.033) *
NGO a Party .277 (.324)
Constitutional Issue —.287 (.206) -.474 (.181) **
Tort Cases -.907 (.215) *** -.948 (.198) ***
ederal Gov’t a Party -.240 (.162) -.377 (.152) **
itigation History
Admin/Specialized -.308 (.288)
Federal Courts .027 (.170)
ase Outcome Characteristics
Allow Appeal .242 (.141) .298 (.133) *
hief Justice
New Chief Justice .112 (.188)
Unstice Characteristics
Gov’t Attorney Exp. 1.987 (.338) *** 1.824 (.260) ***
Prior Judicial Exp. .186 (.294)
Freshman (1't 50 Cases) .722 (.395) .767 (.358) *
Predisposition to Join .720 (.299) * .735 (.283) **
Catholic -.365 (.220)
From New South Wales .336 (.348)
Labor Prime Minister -.663 (.253) ** '-.357 (.145) **
ontrol Variables
Panel Size -.224 (.106) * -.221 (.089) **
Coalition Size .383 (.069) *** .322 (.062) ***
Omnibus Test of Coeff. I 373.867 *** 382.253 ***
Cox & Snell R-Square .253 .243
Nagelkerke R-Square .345 .330
Hosmer & Lemeshow Test 1 7.783 5.111
Correctly Predicted 74.1 % 73.7 %
n = 1371 1371
* p505, ** p501, *** ps.001
liThe numbers presented here are test statistics distributed as chiz.

 

 

242

Figure 22 - Predicted Probabilities of Decision to Join on Gleeson Court by Coalition Size

 

 

 

 

 

 

 

 

 

 

 

 

 

+At Panel Mean 7
+Awrage Distance from Mean
1— Maxumum Distance from Mean
1.0
0.8
g d/
z:
.8 0.6
2
o.
8
‘6 0.4 /#
‘8 ./"
at
0.2 //
0.0 1 1 1 l l
1.0 2.0 3.0 4.0 5.0 6.0 7.0

 

 

 

The measure of opinion length functions as it has across all models: longer average
opinions are linked with higher predicted probabilities of j oining behavior. With this
exception, the complexity hypothesis is unsupported in the Gleeson Court. The salience
hypothesis fares much better, though. The predicted probability of j oining behavior in a
constitutional case is .56, as compared with .67 for non-constitutional cases. The
relationship between joining behavior and tort cases is the same, with tort cases yielding
a predicted probability of .56 and non-tort cases a predicted probability of .67.

In order to create a more individualized picture of j oining behavior on the Gleeson
Court, the model has been adjusted and applied to the joining decisions of each individual

Justice in the Gleeson Court. The Justice characteristics are removed from the model,

243

and the results for each model are presented in Table 17. Certainly, these models do not
perform as well on the individual judges as they do on the full dataset. The explanatory
power is lower, and many of the hypotheses that were supported above are not supported
here. The earlier discussion of multicollinearity explains much of this. The division of
the data into seven pieces leads to micronumerosity—or too few data points—for each
individual model. This can artificially inflate the variance for individual parameter
estimates, leaving models with reasonable r-square analogs without hypothesis-testing
power.

In an attempt to alleviate some of these effects, a reduced model is presented in
Table 18. This model was chosen by including only those variables that showed promise
in the earlier models, while avoiding too many measures of the same concept. The
models perform better for Gleeson, Callinan, and Gaudron than for the rest of the
Justices. All three of these Justices are less likely to join in tort cases. Gleeson, Callinan,
and McHugh are all more likely to join when their coalition size is larger, but these
effects do not reach the p S .05 level of statistical significance. The panel size
coefficients are in the expected direction across the board, although they approach but do
not consistently reach statistical significance.

There are a few more insights here, though, that bear mentioning. First, Justices
Gaudron and Gummow are significantly more likely to join when they are in the majority
in a decision with a discemibly ideological outcome. In addition, along with Justice
Hayne their presence in an opinion coalition is associated with longer average opinion
lengths. Finally, Justice Kirby is significantly less likely to join when the case involves a

constitutional question.

244

 

 

 

 

 
 

 

  

  

   

92 SN 3 8a 2: 42 02 a
82: $6: 8w.» 2? wood 82.2 ~32 ase-5m
82. SN. 8.. man. 3%. m2. gm. £8.82
7.85% 108.3 253 33.8 ...: 8&3 ...-23.8 rig: geese
333. 22.3.4 :32. 239%. SS; £333. 152. 82.80
GS :- ame on 83 on- ...33 8.- 88 8.- 33 $.- 83 $.- .88
29280
$38. 8&3. Elmo 332.- 21%. 33%. 83mm. 8322
080330
338.. cioon- 838.. 633. 83%. aims.- 833- icoeom
£3 3“.- 95 2...- ANE $.- 25 mo. :83 $2 .333 8.7 $289 3.7 toe
23:. 632.7 $38. 83%.. :38. €38.- ..am.vo~.7 .228
85 8.- $3 3. 5.: on 5.: 8. an: a: G3 3.- fi: Ne. oez
$38.- 232- £35. 83E.- 8337 8:8.- 238. €8.52
oo=0=dm
3&8.- QSS- 83:.- GEMM. 838.- 638.- aSon- ”use“:
2.238. A398,- $38. $38. $332. :38. $38. ems:
3&8. 3&8. 83%. 5.5. 83%.- aim? 83$.- 3384
5:33:80
33 mm. 83 S: $3 mm. *8: R. 2.83 3. $3 2. *8: ow. ceme-
tame—003
£3 8. 93 S.- E: 8.- $3 8.- E: 8.- S: 3. :89 8. $30
$38.. 332.- 9.4.5. $38.- :38. 633.. 933. $83
vac—#52,
3:87 A3387 sear: $.9on endg- Amodmwd 5.88.7 8338
—TI..:__.4_/. 7.3-E’— 9:>.:__ 33:22:: 575:2“: :::___:.v :31933

...—SD .5326 2: .5 hertz—om 9:53. .3 20.52 1523—...— - S 036,—-

 

 

245

 

 

 

......__:..:.

.122

25.2:

5:: :_ ....

taco 2835 2: .8 332—»: ”£53. be £252 352:. ...—.22....— gosuom - w— 93:.

:31: :3

:::_::.v

...}...z:

02 SN 02 mom 2: 42 oz a
o3; :2 $3— omow Se 34.: we} use a a:
ME. ME. at. N2. 08. 3a. 8N. 8:332
as.” omaNN .22.. barrom 3.3.. OOVMN as 9%.: .13.. $3.0m .12.. ovmdv as... :bdm BSEEO
*3: 3. G: 8. § .v 3. £3 3.. E .v 2;. ...-*6: S. t? .V 3. 82.80
:3 2.. GS :. fig 2.- 83 3.. GS 2.- ......GS we.- ttsme 8.- sea
20.9.50
63 mm- :3 we- 63 Q.- 83 8. :3 on $3 3- $3 mu.- {50 .Bm
83 3. Q3 on- $3 5.. ASC S .- tame E7 135 8.7 ...-...?3 E7 toe
GE 2. ...QE ”N7 33 mn- 83 5.- $3 on- $3 on- 9:; 8.- .acou
. cocozem
a3 8. $3 8.- 243 8. .289 8. ...?3 mo. 33 8. $3 8. Ems-J
bioEEoU
83 mm. :3 3. a3 8. 1.5:» :33 3. $3 8. 83 we. 8a?
awe—002
$0. a :. 8N: S7 so: an. QIV 8. GM: §- §: 0: an: 8a 8928

 

 

246

Another way to evaluate the joining behavior of the Gleeson Court Justices is to
represent their positions graphically. Each Justice has many opportunities to join with all
of his colleagues on the bench, and the goal of these analyses is to isolate the factors
contributing to the decision to join. Pairs of Justices who tend to join when they have the
chance can be conceived as being ‘close’ to each other in terms of some set of criteria
relevant to opinion writing. Those who tend not to join with each other are ‘far’ from
each other in the same sense. By observing the patterns of these interactions, a spatial
representation of the distances between them can be created.

Multidimensional scaling uses a matrix of dissimilarities in order to reverse-
engineer a spatial representation of the Justices. This matrix can be constructed using the
proportion of potential joining opportunities in which each pair of Justices does not join.
In a general sense, a pair of Justices will be closer together in joining behavior terms
when they vote together frequently. In other words, analysis of a matrix like the one
described above would conflate the effects of j oining decisions with voting decisions.
For this reason, it is appropriate to condition the joining decision on vote compatibility.
Iftwo Justices do not end up on the same side of the vote, their opportunity to join is
essentially lost for that case. For this reason, the entries in the matrix represent joining
distances conditioned on votes. Each element pi, represents the proportion of cases in
which Justice i and Justice j voted together but did not join. Because this matrix is
symmetrical—with the list of Justices forming both the row and column entries, it is

amenable to multidimensional scaling analysis (Davison 1983).

247

Figure 23 - Scree Plot for Gleeson Court Conditional Distance Matrix

 

 

 

 

 

5
4.
3.
2 .-
g
.6 1
.'>
C
(D
O)
[I o
1 2 3 4 5 a 7
Component Number

For the purposes of this analysis, the values of the dissimilarities matrix are
conceptualized as distances in Euclidean space. Calculations are made with the goal of
creating spatial representations of the relative positions for each of the Justices. In order
to determine which representation best represents the distances between the Justices
without adding unnecessary complexity, it is important to identify the smallest level of
dimensionality necessary to depict accurately the judges in Euclidean space. The most
common method is to evaluate a scree plot—a plot of eigenvalues against the dimensions
in order of extraction. Typically, dimensions with eigenvalues smaller than one
considered unnecessary. Figure 23 presents the scree plot for the Gleeson Court

conditional distance matrix, and two dimensions are above the cutoff eigenvalue of one.

248

The two-dimensional representation of the Gleeson Court Justices is presented in Figure

24.107 The orientation of these dimensions is based only on the distances in the data, and
not on any external metric (such as party affiliation, activism, etc). This arrangement
depicts the degree to which Justices join with one another, given that they agree on the
vote. When Justices agree on the disposition of the case, Justices who are close together
on the map are more likely to join with each other. If judges are far from each other on
the map, they are relatively unlikely to join with each other even when they vote together
on the merits.

The most obvious feature of Figure 24 is Kirby’s position to the far left. To find
Kirby this far away from the other Justices may not be much of a surprise, given his

reputation as an “outsider” (Banham 2003; Pelly 2008). Kirby’s dissent rate over this

Figure 24 - Derived Configuration of Joining Behavior Conditional on Votes

 

 

 

McHugh
I
Hayne
, ' Gaudron
Kliby Gummow
I
Gleeson
I
Callinan

I

 

 

 

 

107 The MDS calculations reported here were made using the ALSCAL procedure in SPSS 11.0, and very
similar results were obtained using SAS 8.2. For a review of the procedure, see Davison (1983). For a
defense of the two-dimensional representation of voting behavior on the High Court of Australia, see Wood
(2002)

249

period was the highest, followed by McHugh and then Callinan (Lynch 2003b). While
Kirby and Callinan appear distant from the others—and from each other—McHugh
appears in the core group in the top-left corner. Of course, this spatial representation
does not include information about dissent rates. As such, McHugh is more likely than
the other dissenters to join when he is in agreement with other Justices—even if this does
not happen very ofien.

Legal scholar Andrew Lynch concludes that “[t]he Court seems to have a solid core
led by Justice Gummow and comprising in rough order of influence Chief Justice
Gleeson, Justice Hayne and Gaudron” (Lynch 2003b). In this analysis, though, McHugh
appears to be closer to this group than Gleeson. The difference is that the scaling
procedure has controlled for the vote on the merits. Lynch’s analysis uses percentage
tables of opinion-writing rates, and conditioned joining behavior cannot be detected in
this way.

Patterns of voting behavior are important, but they are not the main subject of this
analysis. Certainly, there are patterns of agreement in terms of vote outcomes that
coincide with ideology—or at least with the party of appointing prime minister, which is
an accepted proxy measure of ideology. The liberalism score used in the above models
was derived from voting behavior, and not from joining behavior. The scores for the
Gleeson Court Justices are presented in Figure 25, separated by appointing prime
minister. Gleeson, Callinan, and Hayne are the Liberal appointees; the rest were
appointed by Labor prime ministers. Comparing this arrangement with Figure 24, it
becomes clear that ideology has at least some relationship with joining patterns on the

Court. Gleeson, Hayne, McHugh, and Gummow appear in a tight cluster in Figure 25,

250

Figure 25 - Liberalism Scores of Gleeson Court Justices by Party of Appointing Prime Minister108

 

 

 

 

Callinan Gleeson
o e
Hayne
McHugh Gummow Kirby
:- a A a
Gaudron
PARTY
oLNC
- - ' _ - ' _ . - ' . AALP
02 0.00 .02 .04 .06 .00 .10 .12 .14 .16 .18 .20 .22

Liberalism Score

indicating that they tend to favor the same ideological case outcomes. Each of these
Justices appears near the others in Figure 24 as well, with Gleeson slightly farther from
the rest. In other words, these Justices tend to favor similar ideological outcomes, and
they also tend to coauthor opinions on a more regular basis.

The ideological agreement also performs well in many of the logistic regression
models of j oining behavior. Table 19 summarizes the performance of the various
hypotheses in each of the seven reduced models. Ideology does not seem to play a part in
single-opinion dispositions, but it is an important explanatory factor in each of the other
models. Liberal panels produce more joint opinions. The Justices are more likely to join

when they agree with the ideological outcome of the case, and less likely to join when

 

108 The liberalism score for each Justice has been assembled from his entire tenure on the High Court.

Some Justices had served long before the beginning of the Gleeson Court. For example, Justice Gaudron
was appointed in 1987, and Justice McHugh in 1989. During this time, the Court made more liberal
decisions. This is how Gaudron can appear to the liberal side of Kirby, even though Kirby’s dissent rate is
legendary. Over the course of their respective careers, a greater proportion of Gaudron’s votes supported
liberal outcomes than did Kirby ’s.

251

Table 19 - Performance of Hypotheses in Seven Models of Joining Behavior
Model Number

 

 

 

 

Workload

la Late Oral Argument - - -

lb Retirement Near *

1c Cases This Term

Chief Justice *** 1"ka tilt *** *Itntr

2a Gibbs *** * -. - - -

2a Mason *** ** *** *** _ - _
Brennan . . - - -
Gleeson =- -. - - -

2b New Chief Justice - - -

Ideology Measures

3a Panel Liberalism u an: - - _ - - _ - - -

3b Panel Diversity *

4a DiStanoe from Mean - - - - - — - - - _ - - antral: first: 13*..-

4b Agree with Direction - - - - - - - - - - - - *M Hut: anus

Complexity

5a Long Oral Argument *** *** * an: us: an:

5b AVg.OpinionLength can our: on - can nor: an

5c Number of Matters ** ** ***

5d No Leave asofRight --- .... ---

Salience

6a Nonparty Participants **

6b NGO as a Party . Mar

6c Constitutional Issue *** * M * ***

6d Tort Cases - - - a sun-

7a Fed. Gov’t as a Party ** ..- u

Litigation History

8a Admin/Specialized - - -

8b Federal Jurisdiction - - -

Case Outcome

9a Nonunanimous Vote - - - - - - *** -- - - - - - - - - -

9b Liberal Outcome - - - - - - - - -

9c Allow Appeal ** H .-

Judge Characteristics

10a Gov’t Attorney Exp. - - - - - - - - - - - - ”at un: am

10b Prior Judicial Exp. - - - - - - - - - - - - *1" Hull

10c Freshman(1"50) --- --- --- --- ..-

10d Predisposition to Join - - - - - - - - - - - - Hut snu: an:

11a Catholic --- --- --- ---

11b FromNSW --- --- --- --- an:

11c LaborP.M. --- --- --- --- a... ..

Control Variables

C1 Panel Size Hut: ant: Hut: :tnlut: an" an:

C2 Coalition Size - - - - - - - - - - - - an ml: taut:

* p505, ** p501, *** p$.001; the character - is substituted when the relationship is in the opposite

direction proposed by the hypotheses. Variables omitted from equations are denoted by - - -.

 

252

they are ideologically distant from the panel’s mean. Figure 24 suggests that even when
voting agreement is controlled for, the Justices tend to join with their ideological kin.

Certainly, ideological compatibility alone does not provide a satisfactory
explanation of the joining patterns found in these analyses. For example, Kirby and
Gaudron have very similar liberalism scores, but they appear on opposite ends of the
spectrum in Figure 24. While the salience and complexity of the case seems to drive
joining behavior—as does the leadership of the Chief Justice—these are controlled for in
Model 7, presented in Table 16. Instead, judge-specific‘characteristics seem to play a
role in the Justices’ joining behavior.

The predisposition-to-join variable provides some insight here. The predisposition-
to-j oin variable is a running average of the last five cases in which each Justice has
joined. Figure 26 shows the average value of this variable for each Justice, multiplied by
five. This yields the average number of times the judge has joined over the previous five
cases. The horizontal reference line is at the average, which is 2.14 cases. Gaudron’s
average is far above the mean, while Kirby’s is far below. The predisposition-to-j oin
variable is related to several of the other Justice-specific background variables. The
predisposition variable is significantly higher among Justices with experience as a judge
or government attorney, Justices who are not Catholic or from New South Wales, and
Justices who are not freshmen on the bench.109 This may introduce multicollinearity into
the regression models, although several of the measures involved consistently performed

strongly.

 

109 Independent samples t-test not assuming equal variances rejects the null hypothesis that these scores are

not different from each other when divided by government attorney experience (t=lO.852, p=.000), judicial
experience (t=21.078, p=.000), Catholicism (t=-18.275, p=.000), being from New South Wales (t=4.445,
p=.000), and freshman status (t=12.181, p=.000).

253

Figure 26 - Average Number of Decisions to Join over Previous Five Cases, by Justice

 

 

      

DWindeyer

E: Menzies

CE McTieman
McHugh

 

n

3.5d

'
C!
m

25
2D
1.5
10

5

91591 01 peuior' 5999:) jo # SW

254

Overall, joining behavior on the High Court seems to be related to several factors
familiar to students of judicial behavior. The production of a single-opinion disposition
is related to the leadership of the Chief Justice and the complexity of the case. Justices
are less likely to join in cases that are salient or complex. Liberal panels are more likely
to produce joint opinions, and Justices who are closer to the panel mean are more likely
to be involved in a joint opinion. Justices are also more likely to join when they agree
with the ideological direction of the outcome.

In addition to these factors, the Justice’s approach to the working environment
seems to be critical to the production of joint opinions. In short, some Justices are simply
more comfortable with the collaborative process. It is likely that Justices on the US.
Supreme Court also differ in their affinity for the process of collegial opinion production.
These differences, though, are largely masked by the majority opinion tradition in that
court. In the absence of this institutional pressure to join, the differences among the

Justices in this regard are more obvious.

255

CHAPTER NINE: CONCLUSION AND FUTURE RESEARCH
Truth is still elusive when she wears mathematical robes.

— Anthony Blackshield (1972, 62)

In a 2005 speech, Justice Kirby echoed Professor Blackshield’s sentiment,
commenting that, “on their own, statistics tell little” (Kirby 2005a). These comments
acknowledge that quantitative methods must be supplemented with carefully considered
theories in order to gain explanatory traction on complex judicial behaviors. Blackshield
notes that international comparisons are particularly hard, “and if comparison is to be
attained realistically at this level, specificity is out of the question” (Blackshield 1972,
64). But statistical methods and data availability have improved dramatically since
Professor Blackshield made this comment. The distribution of information-rich datasets
like the High Courts Judicial Database (Haynie et al. 2007), coupled with advanced
methodological theory and computing power, allow for more targeted and fruitful
analyses. This dissertation is but one example.

While the results are preliminary, some important trends stand out. The main
driving force behind joining behavior on the Supreme Court—the precedential weight of
the majority opinion—is absent on the High Court. But, like the US. Supreme Court, the
leadership of the Australian Chief Justice is strongly related to joining behavior levels.
Chiefs who are able to maintain a cooperative atmosphere and regular interactions among
the Justices are likely to see joining behavior increase during their tenure. Additionally,
judge-specific characteristics are related to the propensity to participate in joint Opinions.
Most notably, Justices who have experience with a cooperative approach to the working
environment are more likely to join than are Justices with a more distinctly “barrister

outlook.” Joining behavior is less likely in complex cases, where there are aspects of the

256

case on which the Justices may disagree. Joining behavior is also less likely in salient
cases, where Justices are less willing to assent to sub-optimal expressions of the reasons
for judgment. Perhaps most importantly, judicial policy preferences do seem to matter.
Justices who are farther from the ideological mean of the panel are less likely to join in
judgment than those who are near the mean. In other words, a Justice is more likely to
join when there are more ideologically compatible colleagues from which to choose.

So, why do High Court Justices join the opinions of their colleagues when they
don’t have to? It is not because they want to save time, nor because there is a source of
institutional pressure to join. Instead, Justices join because 1) the Chief Justice facilitates
joining behavior, 2) they are able to find coauthors with similar policy views, 3) a case is
simple and not of great importance, and 4) because they are accustomed to the process of
collegial decision making.

Evaluating the Findings

In his assessment of empirical legal scholarship, Heise presents a helpful
framework to address a number of data collection and methodological problems in the
existing literature (Heise 2002). Heise notes that using only the universe of published
Opinions can be problematic in judicial behavior research. These published opinions
represent “only an exceedingly small percentage of cases filed,” (Heise 2002, 844) and
are likely to be significantly different in many ways from those cases not disposed of
with published opinions (Atkins 1992). In this way, making generalizations from
findings using reported decisions to all decisions of the court may be problematic (Tate
1983)

The focus in this analysis is on the joining behavior of judges. Case salience is

included as an explanatory factor, but the cases reported in the Commonwealth Law

257

Reports are likely to represent salient cases much more heavily. To isolate fully the
effect of case salience on joining behavior, then, it would be helpful to analyze a dataset
that includes all cases to identify the differential application of the model to published
and unpublished decisions.

Another limitation mentioned by Heise is the application of issue-specific
explanations to “series of unrelated cases in general areas of the law” (Heise 2002, 844).
Heise levels this criticism at research attempting to use social background characteristics
to explain judicial votes for and against certain litigants or outcomes. While social
background characteristics are used as explanatory variables in this analysis, though, they
are not employed to explain judicial votes. Instead, they are analyzed in relation to the
propensity of an individual judge to join with his or her colleagues.

A related cause for concern is the categorization of cases into broad issue areas.
This may be somewhat problematic in the data analyzed here. These categorizations are
only helpful if cases in an individual issue area similar enough that we can assume
judicial behavior in one such case “reflects a general judicial attitude rather than an
individualized resolution of the unique facts in each particular case” (Heise 2002, 845).
The coding scheme for issue areas is contained in Section 2 of Appendix B. Clearly,
each issue category contains a relatively broad range of cases. Generally, though the
subject matter classifications have been constructed following Spaeth’s (1997) lead,
grouping like with like to the highest extent possible in a comparative database.

In addition to this, drawing comparisons using a cross-sectional database requires
an assumption that cases in one point in time are comparable to similar-looking cases at

another point in time. Heise notes that many studies rely on the “assumption that

258

different cases over a period of years are sufficiently similar that the results are both
comparable and generalizable” (Heise 2002, 845). Certainly, institutional changes over
time have changed the nature of High Court cases. In this analysis, several dummy
variables have been included in an attempt to group like cases with like.

There are some avenues for further research, the results of which might may clarify
the findings reported here. Here, I will address the two most promising (and achievable).
First, the attitudinal dimension of j oining behavior might be better understood using an
analysis of judge dyads. This data arrangement would create a much larger sample size,
combating the effects of multicollinearity in the models. It would allow for the
identification of shared characteristics that the panel distance variable in the current
analysis measures indirectly. Ideological scores and other judge-level characteristics
could be compared with each other, and it would be possible to determine, for example, if
Justices who share the same religious background (or experience, or party affiliation,
etc.) are more likely to join in judgment. This approach would also allow for the
identification of issue specialists and stable opinion coalitions.

A second avenue for future research is to develop a more precise indicator of
judicial policy preferences. In this analysis, because the dependent variable was opinion
behavior and not ideological outcome, the attitudinal measures were derived from the
ideological directionality of the votes cast by the individual Justices. While this avoids
the tautology problem, it is not an ideal measure. Certainly, the directionality coding
strategy‘for the HCJD project (Haynie et al. 2007) is not as precise as the Spaeth’s (1997)
scheme, and the meaning of the various outcomes may not be equivalent across countries

(see Appendix B, Section 3). It is possible simply to limit the issue areas subject to

259

analysis, as many analyses of the Supreme Court do, but this leaves precious few data
points to include. A measure akin to the Segal and Cover (1989) scores of Supreme
Court Justices would be ideal and could be achieved more easily using content analysis
techniques on information available in digital format.

The Australian legal community has divided on the issue of joint opinions. Some
argue that they are critical for the coherence of law and the legitimacy of the institution
(Kildea 2003) while others argue that separate opinions preserve the integrity of the
decision-making process (Bagaric and McConvill 2005) and that joint opinions can
amount to an abdication of judicial duty (Blackshield 2004). The Court’s production of
joint opinions has increased regardless, despite the apparently lack of institutional
motivation to engage in the accommodation necessary to craft joint reasons. By creating
a generalization of existing research on joining behavior on the US. Supreme Court, this

analysis has identified some concrete factors to help explain why.

260

APPENDD( A: JUDGES OF THE HIGH COURT OF AUSTRALIA BY SEAT

 

.533 83:... 3 ago. am»? 3:590 8.62. 2 $92 ..0.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

538 ago 25. ea

.35. 83o 5&3 355; megs; gem

gamma ascem 3228 :25 w

5.5.8 >202 as: age. as;
83:22 as; 88.. a; a? 533 _. .

32.530 coma-2 :25 33...“. 9.36

 

 

mo. ‘8. _mm. To. Tm. _om.

 

:wEQL-o—z

 

mm. _oh _mm. .8. _mm. _om. We. Vow. mm. _om. TN.

0

d

C.

 

T, _2. T9:

 

261

APPENDIX B: CODING CLASSIFICATIONS

Section 1- Administrative Agencies & Specialized Tribunals

 

State Agencies

Appeals Costs Board

Building Tribunal

Civil and Administrative Tribunal
Commercial Tribunal

Community Services Appeals Tribunal
Consumer Claims Tribunal

Criminal Injuries Compensation Tribunal
Dust Diseases Tribunal

Equal Opportunity Tribunal

Fair Trading Tribunal

Independent Commission Against Corruption
Industrial Relations Commission

Lands and Mining Tribunal

Mentally Impaired Defendants Review Board
Parole Board

Residential Tenancy Tribunal
Retirement Villages Disputes Tribunal
Social Security Appeals Tribunal

State Administrative Decisions Tribunal
Strata Titles Referee

Supervised Release Review Board
Victims Compensation Tribunal

 

Federal Agencies

Administrative Appeals Tribunal

Australian Industrial Relations Commission
Human Rights and Equal Opportunity Commission
Immigration Review Tribunal

Native Title Tribunal

Refugee Review Tribunal

 

State Special
Courts

Children’s Court
Compensation Court

Drug Court

Environmental Resources and Development Court
Family Court

Industrial Appeal Court
Industrial Relations Court
Land and Environment Court
Liquor Licensing Court
Local Government Court
Youth Court

 

Federal Special
Courts

 

Family Court (as a trial court)
Industrial Relations Court (1994-1997)

 

262

 

Section 2 - Issue Area Classifications

 

Criminal Cases

murder

killing without intent (e.g., manslaughter, reckless homicide)

attempted murder or homicide

rape

other crimes of violence against persons (e.g., assault, child abuse, armed
robbery, kidnap, arson, sexual assault)

attempted crime of violence (including attempted rape)

property crimes - serious (theft, burglary, forgery, fraud, destruction of
property) . '

property crimes - minor

drug offenses

crimes against morality (liquor, obscenity, disorderly conduct, gambling)
religious crimes (including blasphemy, heresy)

business regulation & license violations

govermnent corruption or attempts to corrupt government officials
political crimes (rebellion, subversion, sedition, illegal political activities
or demonstrations, disloyalty, WAR CRIMES)

other (perjury, contempt of court, DUI)

not able to classify crime

 

Civil Rights
Cases

civil rights claims by prisoners (does not include challenges to their
sentence) (includes claims of cruel and unusual punishment while in
prison, habeas corpus for convicted prisoners)

equal treatment under the law (discrimination on account of gender, race,
religion, age, ethnic, language)

. voting rights

speech, press, assembly, right to petition

religion rights (both challenges to restrictions on religion and challenges
to government aid to religion that discriminate against other religions)
privacy (include abortion)

access to public information

civil rights of juveniles

rights of indigenous peoples

other civil rights — including legality of detention for the non-convicted

 

Public Law
Cases

 

government health & safety regulation

government environmental regulation and natural resources

other government regulation of business

government regulation of agriculture

land reform and government regulation of land use (includes eminent
domain and condemnation)

government regulation of races (e. g., apartheid)

government regulation of unions or associations or business/labor
relations

other government regulation

taxation, tariffs

government benefits (medical insurance, welfare, unemployment benefits,
old age pensions or social security)

 

263

 

 

Public Law
Cases
(continued)

worker's compensation

abuse of governmental authority (other than those in topics included in the
civil rights or tort categories)

public employment including disputes between employees about
promotions, etc.

immigration, deportation, citizenship

disputes over elections

disputes over appointments to office

disputes related to removal of government officials

disputes between different units of government (Federalism)
special laws dealing with indigenous peoples

religious or customary law

language or cultural issues

other public law (NOT government contract disputes)

 

Economic
Cases

creditor-debtor disputes (collection; bankruptcy; insolvency cases; etc).
insurance disputes

other contract disputes, breach of contract, franchise agreements,
including disputes over government contracts

land possession disputes

landlord - tenant disputes - agricultural

landlord - tenant disputes - commercial or residential (and default)
copyrights, patents, trademarks

corporate law - disputes over control or mismanagement of corporations;
stockholders and creditors rights; securities fraud

labor/ management disputes (in private business)

other private economic disputes

 

Tort Cases

motor vehicle accidents

injured workers

professional malpractice (legal, medical, etc.)
government tort liability

product liability (not shipping or service liability)
other personal injury

libel or defamation

other torts (including shipping)

 

Family Law
Cases

marriage, divorce, and family disputes
inheritance, probate, succession disputes
other family law

 

 

Other Cases

regulation of bar and judiciary
other issue
unable to classify

 

264

 

Section 3 - “Liberal” Directionality of Decisions and Votes

 

Criminal Cases

Civil Rights Cases

Freedom of Speech
& Religion Cases

Public Law Cases

Economic Cases

Tort Cases

Bar/Judiciary Cases

 

Outcome supports the position of the defendant.

Outcome supports the position of the person alleging that their civil rights
have been violated.

Outcome supports the expansion or protection of the right of free
expression or free exercise of religion.

Outcome supports benefit claims of public employees or other recipients of
government benefits.
Outcome supports the incumbent in electoral disputes.

Outcome supports the national government over the states in federal
disputes.

Outcome supports the right of the government to regulate labor relations.

Outcome otherwise supports the position of the government.

Outcome supports the position of the economic underdog.

Outcome supports the position of an individual over that of the government
in government contract disputes.

Outcome supports the claim of the person alleging infringement of
copyright, patent or trademark.

Outcome supports the position of the labor union over that of the
corporation.

Outcome supports the party alleging the injury.

Outcome supports the position of the attorney or judge accused of
wrongdoing.

 

265

 

BIBLIOGRAPHY

Ackland, Richard. 2003. "All Rise: Judges, Last of the Secular Gods." Sydney Morning
Herald, August 30, 1, 5-6.

 

. 2007. "Rudd has Two High Court Spots to be Sorted Out. " Sydney Morning
Herald, Nov. 30.

Albrechtsen, Janet. 2003. "Leave Policy to the Law-Makers. " The Australian,
October 15, 15.

 

. 2007. "May the Best Person Preside. " The Australian, Aug. 14.

Aldrich, J. H., and F. D. Nelson. 1984. Linear Probability, Logit and Probit Models.
Thousand Oaks, CA: Sage.

Aliotta, Jilda M. 1988. "Combining Judges' Attributes and Case Characteristics: An
Alternative Approach to Explaining Supreme Court Decision-Making."
Judicature 71 :277-81.

Allen, James. 2008. "When it Comes to Picking High Court Justices, Give the Little Guys
a Go." The Australian, May 9.

Alvey, John, and Neal Ryan. 2005. "The Separation of Powers in Australia: Implications
for the State of Queensland." Paper read at Australasian Political Studies
Association Conference, Sep 28-30, at Dunedin, NZ.

Atkins, Burton. 1991. "Party Capability Theory as an Explanation for Intervention
Behavior in the English Court of Appeal." American Journal of Political Science
35 (4):881-903.

 

. 1992. "Data Collection in Comparative Judicial Research: A Note on the Effects
of Case Publication upon Theory Building and Hypothesis Testing." Western
Political Quarterly 45 (3):?83-92.

Atkins, Burton, and Justin J. Green. 1976. "Consensus on the United States Courts of
Appeals: Illusion or Reality? " American Journal of Political Science 20 (4)2735-
48.

Atkinson, David N., and Dale A. Neuman. 1969. "Toward a Cost Theory of Judicial
Alignments: The Case of the Truman Bloc." Midwest Journal of Political Science
13 (2):271.

Australian Law Reform Commission. 2000. "The Judicial Power of the Commonwealth:
A Review of the Judiciary Act 1903 and Related Legislation." Sydney: Australian
Law Reform Commission.

266

Axelrod, Robert. 1984. The Evolution of Cooperation. New York: Basic Books.

 

. 1986. "An Evolutionary Approach to Norms." American Political Science
Review 80 (4): 1095-1 1 1.

Bacon, Rachel. 2002. "Rewriting the Social Contract? The SSAT, the AAT and the
Contracting Out of Employment Services." Federal Law Review 30 (1)239-68.

Bagaric, Mirko, and James McConvill. 2005. "The High Court and the Utility of Multiple
Judgments." High Court Quarterly Review 1 (1)21.

Bailey, Michael A., Brian Kamoie, and Forrest Maltzman. 2005. "Signals from the Tenth
Justices: The Political Role of the Solicitor General in Supreme Court Decision
Making." American Journal of Political Science 49 (1):72-85.

Baker, Belinda, and Stephen Gageler. 2002. "F. G. Brennan." In The Oxford Companion
to the High Court of Australia, ed. T. Blackshield, M. Coper and G. Williams.
South Melbourne: Oxford University Press.

Banham, Cynthia. 2003. "Kirby the High Court Outsider. " Sydney Morning Herald,
February 24.

Barwick, Sir Garfield. 1995. A Radical Tory: Garfield Barwick's Reflections and
Recollections. Sydney, NSW: The Federation Press.

Baum, Lawrence. 1981. "Comparing the Implementation of Legislative and Judicial
Policies." In Eflective Policy Implementation, ed. D. A. Mazmanian and P. A.
Sabatier. Lexington, MA: Lexington Books.

 

. 1997. The Puzzle of Judicial Behavior. Edited by J. Aldrich, B. Bueno de
Mesquita, R. W. J ackman and D. Rohde. Ann Arbor: University of Michigan
Press.

 

. 2003. "C. Herman Pritchett: Innovater with an Ambigious Legacy." In The
Pioneers of Judicial Behavior, ed. N. Maveety. Ann Arbor: University of
Michigan.

Bayne, Peter. 2002. "Herbert Vere Evatt." In The Oxford Companion to the High Court of
Australia, ed. T. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

Beaumont, Bryan. 2002. "Federal Court of Australia." In The Oxford Companion to the
High Court of Australia, ed. T. Blackshield, M. Coper and G. Williams. South
Melbourne: Oxford University Press.

Benesh, Sara C., and Malia Reddick. 2002. "Overruled: An Event History Analysis of
Lower Court Reaction to Supreme Court Alteration of Precedent." Journal of
Politics 64 (2):534-50.

267

Bennett, David. 2002. "Argument Before the Court." In The Oxford Companion to the
High Court of Australia, ed. A. Blackshield, M. Coper and G. Williams. South
Melbome: Oxford University Press.

Bennett, J. M, ed. 1969. A History of the New South Wales Bar. Sydney: The Law Book
Company.

Bennett, John Michael. 1980. Keystone of the Federal Arch: A Historical Memoir of the
High Court of Australia to 1980. Canberra: Australian Government Publishing
Service. '

Bennett, L. 1989. "Ideology in Australian Judicial Practice: A Non-Reductionist Account
of a Jurisdictional Issue in Labour Law." International Journal of the Sociology of
LAW 17:207.

Bergman, Matthew P. 1991. "Dissent in the Judicial Process: Discord in Service of
Harmony." Denver University Law Review 68:79.

Best, Bradley J. 2000. "Rethinking the "Chief Justice Effect" and the Role of Ideology in
the Decline of Consensual Norms on the United States Supreme Court." Speech
delivered to Midwest Political Science Association on, at Chicago.

Bhattacharya, Mita, and Russell Smyth. 20013. "Aging and Productivity among Judges:
Some Empirical Evidence from the High Court of Australia." Australian
Economic Papers 40 (2): 199-212.

 

. 2001b. "The Determinants of Judicial Prestige and Influence: Some Empirical
Evidence from the High Court of Australia. " Journal of Legal Studies 30 (1):223-
52.

Black, Ryan C., and James F. Spriggs, H. 2007. "An Empirical Analysis of the Trends,
Determinants, and Effects of the Length of Majority Opinions of the US.
Supreme Court." Paper read at Conference on Empirical Legal Studies, Nov. 9-
10, at New York.

Blackshield, Anthony. 1980. "The High Court: Change and Decay." Legal Service
Bulletin 5:107.

 

. 2001. "Free Speech Cases." In The Oxford Companion to the High Court of
Australia, ed. A. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

 

. 2002a. "The Murphy Affair." In The Oxford Companion to the High Court of
Australia, ed. A. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

 

. 2004. "Judicial Activism and Judicial Restraint: Where Does the Balance Lie? "
Speech delivered to Constitutional Law Conference on February 20, at Sydney.

268

Blackshield, Anthony, Michael Coper, Graham Fri cke, and Troy Simpson. 2001.
"Notable Counsel." In The Oxford Companion to the High Court of Australia, ed.
A. Blackshield, M. Coper and G. Williams. South Melbourne: Oxford University
Press.

Blackshield, Tony. 1972. "Quantitative Analysis: The High Court of Australia, 1964-
1969." Lawasia 3:1.

 

. 1978. "X/Y/Z/N Scales: The High Court of Australia." In Understanding
Lawyers, ed. R. Tomasic. Sydney: Law Foundation of New SouthWales.

 

. 2002b. "Australia Acts." In The Oxford Companion to the High Court of
Australia, ed. T. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

 

. 2002c. "The Dismissal of 1975." In The Oxford Companion to the High Court of
Australia, ed. T. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

 

. 2002d. "The Engineer's Case." In The Oxford Companion to the High Court of
Australia, ed. T. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

 

. 2002e. "Jurimetrics." In The Oxford Companion to the High Court of A ustralia,
ed. T. Blackshield, M. Coper and G. Williams. South Melbourne: Oxford
University Press.

 

. 2002f. "Puisne Justices." In The Oxford Companion to the High Court of
Australia, ed. T. Blackshield, M. C0per and G. Williams. South Melbourne:
Oxford University Press.

Blackshield, Tony, Michael Coper, and John Goldring. 2002. "The Judicial Committee of
the Privy Council." In The Oxford Companion to the High Court of A ustralia, ed.

T. Blackshield, M. Coper and G. Williams. South Melbourne: Oxford University
Press.

Blackshield, Tony, and Mark Mackrell. 2002. "Kenneth Sydney Jacobs." In The Oxford
Companion to the High Court of Australia, ed. T. Blackshield, M. Coper and G.
Williams. South Melbourne: Oxford University Press.

Blaikie, Norman W. H. 2003. Analyzing Quantitative Data. Thousand Oaks, CA: Sage.

Bollerslev, Tim, and Richard T. Baillie. 1994. "Cointegraton, Fractional Cointegration,
and Exchange rate Dynamics." Journal of Finance 49 (2):73 7-45.

Botsman, Peter. 2000. The Great Constitutional Swindle. Annandale, NSW: Pluto Press.

269

Box-Steffensmeier, Janet M., and Andrew R. Tomlinson. 2000. "Fractional Integration
Methods in Political Science." Electoral Studies 19 (1)263-76.

Box, George, and David R. Cox. 1964. "An Analysis of Transformations." Journal of the
Royal Statistical Society 26 (B):21 1-52.

Brace, Paul, and Melinda Gann Hall. 1993. "Integrated Models of Judicial Dissent."
Journal of Politics 55 (4):914-35.

 

. 1997. "The Interplay of Preferences, Case Facts, Context, and Rules in the
Politics of Judicial Choice." Journal of Politics 59 (4): 1206-31.

 

. 2005. "Is Judicial Federalism Essential to Democracy? State Courts in the
Federal System." In The Judicial Branch, ed. K. L. Hall and K. T. McGuire. New
York: Oxford University Press.

Bradford, Gillian. 2007. "Kiefel Appointment Makes High Court History." ABC News,
Aug. 14.

Branson, Catherine. 1995. "Equal Opportunity: The Next Twenty Years." Speech
delivered to Equal Opportunity Commision of South Australia Mitchell Oration
on October 6, at Adelaide.

. 1997 . "Running on the Edge." Speech delivered to Women Lawyers’ Association
of NSW Speaker Series on October 15, at Sydney.

 

Brennan, Gerard. 1996a. "A Tribute to Sir Anthony Mason." In Courts of F inal
Jurisdiction: The Mason Court in Australia, ed. C. Saunders. Annandale, NSW:
Federation Press.

 

. 1996b. "A Tribute to Sir Anthony Mason." In Courts of Final Jurisdiction: The
Mason Court in Australia, ed. C. Saunders. Sydney, NSW: The Federation Press.

 

. 2002a. "Composition of Bench." In The Oxford Companion to the High Court of
Australia, ed. A. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

 

. 2002b. "Decision-Making Process." In The Oxford Companion to the High
Court of Australia, ed. A. Blackshield, M. Coper and G. Williams. South
Melbourne: Oxford University Press.

Brenner, Saul. 1979. "Minimum Winning Coalitions on the US. Supreme Court: A
Comparison of the Original Vote on the Merits with the Opinion Vote." American
Politics Quarterly 72384-92.

. 2003. "David Rohde: Rational Choice Theorist." In The Pioneers of Judicial
Behavior, ed. N. Maveety. Ann Arbor: University of Michigan Press.

 

270

Brenner, Saul, and Theodore S. Arrington. 2002. "Measuring Salience on the Supreme
Court: A Research Note. " Jurimetrics 43 :99-1 13.

Brenner, Saul, and Timothy M. Hagle. 1996. "Opinion Writing and Acclimation Effects."
Political Behavior 18 (3)223 5-61 .

Brenner, Saul, Timothy M. Hagle, and Harold J. Spaeth. 1990. "Increasing the Size of
Minimum Winning Original Coalitions on the Warren Court." Polity 23 (2)2309-
18.

Brenner, Saul, and Harold J. Spaeth. 1986. "Issue Specialization in Majority Opinion
Assignment on the Burger Court." Western Political Quarterly 39 (3)2520.

 

. 1988. "Majority Opinion Assignments and the Maintenance of the Original
Coalition on the Warren Court." American Journal of Political Science 32 (1):72.

Burmester, Henry. 2002. "Interveners and Amici Curiae." In The Oxford Companion to
the High Court of Australia, ed. A. Blackshield, M. Coper and G. Williams. South
Melbourne: Oxford University Press.

Butler, Des. 2005. "A Tort of Invasion of Privacy in Australia?" Melbourne University
Law Review 29:339.

Caldeira, Gregory A., and Christopher J. W. Zorn. 1998. "Of Time and Consensual
Norms in the Supreme Court." American Journal of Political Science 42 (3):874-
902. ’

Callinan, Ian. 1999. "Courts: First and Final." Speech delivered to Speakers' Forum on
August 17, at Sydney

 

. 2000. "A Toast to the New Silks." Speech delivered to New Silks Ceremony on
January 31, at Canberra.

Campbell, Enid. 2003. "Reasons for Judgment: Some Consumer Perspectives."
Australian Law Journal 77:62.

Campbell, Enid Mona, and H. P. Lee. 2001. The Australian Judiciary. Cambridge, UK. ;
New York, NY: Cambridge University Press.

Campbell, J. Louis. 1983. "The Spirit of Dissent." Judicature 66:305.

Campbell, John Y., and Pierre Perron. 1991. "Pitfalls and Opportunities: What
Macroeconomists Should Know about Unit Roots. " NBER Macroeconomics
Annual 62141-201.

Cane, Peter. 2003. "Reforming Tort Law in Australia: A Personal Perspective."
Melbourne University Law Review 27:649.

271

Cannon, Bradley C. 1974. "Reactions of State Supreme Courts to a US. Supreme Court
Civil Liberties Decision." Law & Society Review 8: 109-35.

Cannon, Bradley C., and Michael Giles. 1972. "Recurring Litigants: Federal Agencies
before the Supreme Court." Western Political Quarterly 28: 183-91.

Came, Greg. 2002. "A Compelling Consideration of Law and Politics in the High Court
of Australia." Melbourne University Law Review 26 (1):224-35.

Carney, Gerard. 1997. "Comment - The Role of the Attorney-General." Bond Law
Review 9:1-10.

Carrigan, Frank. 2003. "A Blast From the Past: The Resurgence of Legal Formalism. "
Melbourne University Law Review 27 (1):163-85.

Castles, Alex C. 1971. An Introduction to Australian Legal History. Sydney: The Law
Book Company.

Childs, Kevin. 2001. "Open Process for High Court Judges, Urges Justice Kirby." Law
Institute of Victoria, October 14.

Clark, Andrew. 2003. "In the Court of Murray Gleeson." Australian Financial Review,
August 22, 10-2.

Clarke, Harold D., and Matthew Lebo. 2003a. "Fractional (Co)integration and Governing
Party Support in Britain." British Journal of Political Science 33 2283-301.

Clarke, Harold D., and Matthew J. Lebo. 2003b. "Fractional (Co)integration and
Governing Party Support in Britain." British Journal of Political Science 33 :283-
301. '

Cohen, Jacob, Patricial Cohen, Stephen G. West, and Leona S. Aiken. 2003. Applied
Multple Regression / Correlation Analysis for the Behavioral Sciences. 3rd ed.
Mahwah, NJ: Lawrence Erlbaum.

Cohen, Linda R., and MAtthew L. Spitzer. 1994. "Solving the "Chevron" Puzzle." Law
and Contemporary Problems 57 (2)265-110.

Collins, Paul M. 2005. "Mobilizing Dissensus on the US. Supreme Court." Speech
delivered to Annual Meeting of the New York State Political Science Association
on April 15-16, at Niagara, NY.

Collins, Paul M., Jr. 2006. "Transforming the Original US. Supreme Court Judicial
Database: An Alternative Approach For Use With Stata." Law & Courts 16
(1)222-4.

 

. 2007. "Amici Curiae and Dissensus on the US. Supreme Court." Journal of
Empirical Legal Studies 5 (1)1143-70.

272

Cook, Beverly Blair. 1995. "Justice Brennan and the Institutionalization of Dissent
Assignment." Judicature 79: 17.

Coper, Michael. 2002a. "Cole v. Whitfield." In The Oxford Companion to the High Court
of Australia, ed. T. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

 

. 2002b. "Intergovernmental Immunities." In The Oxford Companion to the High
Court of Australia, ed. T. Blackshield, M. Coper and G. Williams. South
Melbourne: Oxford University Press.

 

. 2002c. "Joint Judgments and Separate Judgments." In The Oxford Companion to
the High Court of A ustralia, ed. A. Blackshield, M. Coper and G. Williams. South
Melbourne: Oxford University Press.

 

. 2003. "The Seven Habits of a Highly Effective High Court." Alternative Law
Journal 25:59-64.

 

. 2004. "The High Court and the Parliament: Partners in Law-Making or Hostile
Combatants?" Speech delivered to The Distinctive Foundations of Australian
Democracy on, at Canberra.

Coper, Michael, George Williams, and Anthony Blackshield. 2002. The Oxford
Companion to the High Court of Australia. Melbourne: Oxford University Press.

Comall, Robert. 2007. "Redefining the Role of Government Lawyers in Today's Public
Service." Speech delivered to ACLA National Conference on November 2, at
Canberra, ACT.

Cowen, Zelman. 1955. "A Comparison of the Constitutions of Australia and the United
States." Buffalo Law Review 4: 155-80.

 

. 1967. Isaac Isaacs. Melbourne: Oxford University Press.

 

. 2002. "John Greig Latham." In The Oxford Companion to the High Court of
Australia, ed. T. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

Cowen, Zelman, and Leslie Zines. 2002. Cowen and Zines '5 Federal Jurisdiction in
Australia. 3rd ed. Annandale, N.S.W.: Federation Press.

Craske, Rebecca, and Harry Gibbs. 2002. "Business of Court." In The Oxford Companion
to the High Court of Australia, ed. A. Blackshield, M. Coper and G. Williams.
South Melbourne: Oxford University Press.

Craske, Rebecca, and Richard Haigh. 2002. "Judgment Production." In The Oxford
Companion to the High Court of Australia, ed. A. Blackshield, M. Coper and G.
Williams. South Melbourne: Oxford University Press.

273

Crawford, James. 1993. Australian Courts of Law. 3rd ed. Melbourne: Oxford University
Press.

Creyke, Robin. 2001. "Tribunals: Divergence and Loss." Federal Law Review 29
(3):403-26.

Danelski, David J. 1960. "The Influence of the Chief Justice in the Decisional Process."
In Courts, Judges, and Politics, ed. W. M. Murphy and C. H. Pritchett. New
York: Random House.

Davis, Jessica Milner, and Troy Simpson. 2002. "Humour." In The Oxford Companion to
the High Court of Australia, ed. A. Blackshield, M. Coper and G. Williams. South
Melbourne: Oxford University Press.

Davis, John F., and William L. Reynolds. 1974. "Juridical Cripples: Plurality Opinions in
the Supreme Court." Duke Law Journal 1974 (l):59.

Davison, Mark L. 1983. Multidimensional Scaling. New York: Wiley.

Dawson, Daryl, and Grant Anderson. 1996. "Dixon, Sir Owen (1886-1972)." In
Australian Dictionary of Biography. Melbourne: Melbourne University Press.

Dawson, Robert Macgregor, and Norman Ward. 1987. Dawson 's the Government of
Canada. 6th ed. Toronto: University of Toronto Press.

de Vaus, David. 2002. Analyzing Social Science Data: 50 Key Problems in Data
Analysis. Thousand Oaks, CA: Sage.

Deen, Rebecca E., Joseph Ignagni, and James Meernik. 2001. "Executive influence on
the US. Supreme Court: Solicitor General Amicus Cases, 1953-1997." American
Review of Politics 22 (1)23-26.

Del Villar, Gim, and Troy Simpson. 2002. "Circuit System." In The Oxford Companion
to the High Court of Australia, ed. A. Blackshield, M. Coper and G. Williams.
South Melbourne: Oxford University Press.

Dickey, David A., and Wayne A. Fuller. 1979. "Distribution of the Estimators for
Autoregressive Time Series With a Unit Root." Journal of the American
Statistical Association 74 (3 66):427-3 l .

Dillon, Michelle, and John Doyle. 2002. "The Mason Court." In The Oxford Companion
to the High Court of Australia, ed. T. Blackshield, M. Coper and G. Williams.
South Melbourne: Oxford University Press.

Dominello, Francesca. 2001. "Intergovernmental Immunities and Judicial Reasoning." In
The Oxford Companion to the High Court of A ustralia, ed. A. Blackshield, M.
Coper and G. Williams. South Melbourne: Oxford University Press.

274

Dominello, Francesca, and Eddy Neumann. 2002. "Background of Justices." In The
Oxford Companion to the High Court of Australia, ed. T. Blackshield, M. Coper
and G. Williams. South Melbourne: Oxford University Press.

Donald, Peta. 2005. "Susan Crennan Sworn in as New High Court Judge." The World
Today, November 8.

Doogan, Christopher M. 2004. "High Court of Australia Annual Report: 2003-2004."
Canberra: High Court of Australia.

Dorff, Robert H., and Saul Brenner. 1992. "Conformity Voting on the United States
Supreme Court. " Journal of Politics 54 (3):762-75.

Douglas, Roger. 1969. "Judges and Policy on the Latham Court." Politics 4:20.

. 2002. "The Latham Court." In The Oxford Companion to the High Court of
Australia, ed. T. Blackshield, M. C0per and G. Williams. South Melbourne:
Oxford University Press.

 

Douglas, William O. 1948. "The Dissent: A Safeguard of Democracy." Journal of the
American Judicature Society 32:104.

Doyle, John. 1996. "Implications of Judicial Law-Making." In Courts of Final
Jurisdiction: The Mason Court in Australia, ed. C. Saunders. Annandale, NSW:
Federation Press.

Duxbury, Neil. 2001. Jurists and Judges: An Essay on Influence. Oxford: Hart
Publishing.

Edelman, James, and Natalie Gray. 2002. "J. L. Toohey." In The Oxford Companion to
the High Court of Australia, ed. T. Blackshield, M. Coper and G. Williams. South
Melbourne: Oxford University Press.

Elder, John, and Peter E. Kennedy. 2001. "Testing for Unit Roots: What Should Students
be Taught?" Journal of Economic Education Spring: 137-46.

Elliot, Graham, Thomas J. Rothenberg, and James H. Stock. 1996. "Efficient Tests for an
Autoregressive Unit Root." Econometrica 64 (4):813-36.

Engle, Robert F ., and Briton Clive W.J. Granger. 1987. "Cointegration and Error
Correction: Representation, Estimation and Testing." Econometrica 55:251-76.

Epp, Charles R. 1998. The Rights Revolution: Lawyers, Activists, and Supreme Courts in
Comparative Perspective. Ann Arbor: University of Michigan Press.

Epstein, Lee, and Jack Knight. 1998. The Choices Justices Make. Washington DC:
Congressional Quarterly Press.

275

 

. 2000. "Toward a Strategic Revolution in Judicial Politics: A Look Back, A Look
Ahead." Political Research Quarterly 53 (3):625-61.

Epstein, Lee, Jack Knight, and Andrew D. Martin. 2003. "The Norm of Prior Judicial
Experience and Its Consequences for Career Diversity on the US. Supreme
Court." California Law Review 91 (4):903 -65.

Epstein, Lee, and Jeffrey A. Segal. 2000. "Measuring Issue Salience." American Journal
of Political Science 44 (1)266-83.

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 (2)2362.

Epstein, Lee, and Thomas G. Walker. 2001. Constitutional Law for a Changing America:
Rights, Liberties and Justice. 4th ed. 2 vols. Vol. 2. Washington, DC: CQ Press.

Epstein, Lee, Jeffrey A. Segal, Harold. J. Spaeth, and T. G. Walker. 1996. "The Supreme
Court Compendium: Data, Decisions, and Developments, Washington:
Congressional Quarterly." Washington, DC: Congressional Quarterly.

Etherington, David. 2005. "Life at the Bar. " Legal Week Student, March 11.

Evans, Simon. 2002. "Appointment of Justices." In The Oxford Companion to the High
Court of Australia, ed. T. Blackshield, M. Coper and G. Williams. South
Melbourne: Oxford University Press.

Ferejohn, John. 1995. "Law, Legislation and Positive Political Theory." In Modern
Political Economy, ed. J. Banks and E. Hanushek: Cambridge University Press.

F erej ohn, John, and Charies Shipan. 1990. "Congressional Influence on Bureaucracy."
Journal of Law, Economics and Organization 6:1.

F erej ohn, John, and Barry Weingast. 1992. "A Positive Theory of Statutory
Interpretation." International Review of Law and Economics 122263-79.

Festa, Linda, and Leo D. Vichules. 1968. "Cliques on the Supreme Court: Myth or
Reality." Sociological Quarterly 9 (4):540-54.

Franklin, Charles H., and Liane C. Kosaki. 1989. "The Republican Schoolmaster: US.
Supreme Court, Public Opinion, and Abortion." American Political Science
Review 83 (3):751-71.

Fricke, Graham. 1999. "The Knox Court: Exposition Unnecessary." Federal Law Review
27 (1).

 

. 2002a. "Frank Gavan Duffy." In The Oxford Companion to the High Court of
Australia, ed. T. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

276

 

. 2002b. "The Gavan Duffy Court." In The Oxford Companion to the High Court
of Australia, ed. T. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

 

. 2002c. "The Isaacs Court." In The Oxford Companion to the High Court of
Australia, ed. T. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

 

. 2002d. "The Knox Court." In The Oxford Companion to the High Court of
Australia, ed. T. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

Fricke, Graham L. 1986. Judges of the High Court. Melbourne, Vic.: Hutchinson of
Australia.

Gageler, Stephen. 2002a. "Jurisdiction." In The Oxford Companion to the High Court of
Australia, ed. T. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

 

. 2002b. "Legalism." In The Oxford Companion to the High Court of Australia,
ed. T. Blackshield, M. Coper and G. Williams. South Melbourne: Oxford
University Press.

 

. 2002c. "Role of Counsel." In The Oxford Companion to the High Court of
Australia, ed. A. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

Galligan, Brian. 1987. Politics of the High Court : A Study of the Judicial Branch of
Government in Australia. St. Lucia ; New York: University of Queensland Press.

 

. 1991. "Judicial Activism in Australia." In Judicial Activism in Comparative
Perspective, ed. K. M. Holland. New York: St. Martin's Press.

 

. 1995. A Federal Republic .' Australia's Constitutional System of Government.
Cambridge ; New York: Cambridge University Press.

 

. 2002. "World War II." In The Oxford Companion to the High Court of A ustralia,
ed. T. Blackshield, M. Coper and G. Williams. South Melbourne: Oxford
University Press.

Gardiner, David, and Frances McGlone. 1998. Outline of Torts. 2nd ed: Butterworths.

Gaudron, Mary. 2002. "Speech for Women Lawyers Association of New South Wales."
Speech delivered to Women Lawyers Association of New South Wales
Conference on June 13, at Sydney.

George, Tracey, and Lee Epstein. 1992. "On the Nature of Supreme Court Decision
Making." American Political Science Review 86:323-37.

277

Gerber, Scott D., and Keeock Park. 1997. "The Quixotic Search for Consensus on the
US. Supreme Court: A Cross-Judicial Empirical Analysis of the Rehnquist Court
Justices." American Political Science Review 91 (2)390.

Gibbs, Harry. 1986. "Appellate Procedures in the High Court." Australian Bar Review
2: 1. '

. 2002. "Samuel Walker Griffith." In The Oxford Companion to the High Court of
Australia, ed. T. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

 

Giles, Michael W. 1977. "Equivalent Versus Minimum Winning Opinion Coalition Size:
A Test of Two Hypotheses." American Journal of Political Science 21 (2):405.

Gillman, Howard, and Cornell W. Clayton. 1999. The Supreme Court in American
Politics: New Institutionalist Interpretations. Lawrence, KS: University Press of
Kansas.

Ginsburg, Ruth Bader. 1994. "Gifts of Speech." Speech delivered to Annual Dinner of
the American Law Institute on May 19, at Sweet Briar, VA.

Gleeson, Anthony Murray. 1998. "The Role of the Judge and Becoming a Judge." Speech
delivered to National Judicial Orientation Programme on August 16, at Sydney.

 

. 2000a. "Judicial Legitimacy." Speech delivered to Australian Bar Association
Conference on July 2, at New York.

 

. 2000b. "The Keystone of the Federal Arch." In Radio National 's Boyer Lectures.
Australia: Australian Broadcasting Corporation.

 

. 2001a. "A Changing Judiciary." Speech delivered to Judicial Conference of
Australia on April 7, at Uluru.

. 2001b. "Courts and the Rule of Law on November 7, at Melbourne.

 

 

. 2002a. "The Birth, Life and Death of Section 74." Speech delivered to The
Samuel Griffith Society on June 14, at Sydney.

 

. 2002b. "Global Influences on the Australian Judiciary." Speech delivered to
Australian Bar Association Conference on July 8, at Paris.

. 2002c. "The Oxford Companion to the High Court of Australia." Speech
delivered to Launch on February 13, at Canberra.

 

 

. 2003a. "Judicial Selection and Training: Two Sides of the Same Coin." Speech
delivered to Judicial Conference of Australia on May 31, at Darwin.

278

 

. 2003b. "Legality - Spirit and Principle." Speech delivered to Second Magna
Carta Lecture on November 20, at Sydney.

 

. 2004a. "The High Court of Australia: Challenges for its New Century." Paper
read at Constitutional Law Conference, February 20, at Sydney, NSW.

 

. 2004b. "The High Court: Challenges for a New Century." Sydney Morning
Herald, February 21.

 

. 2005. "The State of the Judicature." Speech delivered to 19th Biennial
Conference of Lawasia on March 24, at Gold Coast.

Gleeson, Murray. 1997. "The Role of the Judiciary in a Modern Democracy." " Speech
delivered to Judicial Conference of Australia Annual Symposium on, at Sydney,
NSW.

Glendon, Mary Ann, Michael W. Gordon, and Paolo Wright-Carozza. 1999. Comparative
Legal Traditions in a Nutshell. 2nd ed. St. Paul, Minn: West Group.

Glick, Henry R. 1983. Courts, Politics, and Justice. New York: McGraw Hill.

Goldberger, A. S. 1991. A Course in Econometrics. Cambridge, MA: Harvard University
Press.

Goldman, Sheldon, and Austin Sarat, eds. 1978. American Court Systems: Readings in
Judicial Process and Behavior. San Francisco: W.H. Freeman.

Goldring, John. 1997. "The Path to Engineers." In How Many cheers for Engineers?, ed.
M. Coper and G. Williams. Sydney: Federation Press.

Goldsmith, Henry. 2006. "No Bar to Employment: Judicial Appointment and the
Employed Bar. " Speech delivered to Employed Barristers' Conference on March
17, at United Kingdom.

Granger, Briton Clive W.J., and Paul Newbold. 1977. Forecasting Economic Time Series.
New York: Academic Press.

Granger, Clive W.J. 1980. "Long Memory Relationships and the Aggregation of
Dynamic Models." Journal of Econometrics 14 (2):227-38.

Green, Donald P., and Ian Shapiro. 1994. Pathologies of rational choice theory: a
critique of applications in political science. New Haven: Yale University Press.

Griffiths, John. 1985. "Australian Administrative LAw: Institutions, Reforms and
Impact." Public Administration 63 (4):445-63.

Grossman, Joel B. 1968. "Dissenting Blocs on the Warren Court: A Study in Judicial
Role Behavior. " Journal of Politics 30 (4): 1068-90.

279

Grossman, Joel B., and Joseph Tanenhaus, eds. 1969. Frontiers of Judicial Research.
New York: John Wiley.

Groves, Matthew, and Russell Smyth. 2004. "A Century of Judicial Style: Changing
Patterns in Judgment Writing on the High Court 1903-2001." Federal Law
Review 11.

Guamieri, Carlo. 2004. "Appointment and Career of Judges in Continental Europe: The
Rise of Judicial Self-Government." Legal Studies 24: 169-87.

Guilfoyle, Kate. 2002. "M. H. McHugh." In The Oxford Companion to the High Court of
Australia, ed. T. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

Hagle, Timothy M. 1993. "'Freshman Effects' for Supreme Court Justices." American
Journal of Political Science 37 (4)21142-57.

Halpem, Stephen C., and Kenneth N. Vines. 1977. "Institutional Disunity, the Judges'
Bill and the Role of the US. Supreme Court." Western Political Quarterly 30
(4):471-83.

Hammond, Thomas H., Chris W. Bonneau, and Reginald S. Sheehan. 2005. Strategic
Behavior and Policy Choice on the US. Supreme Court. Stanford,CA: Stanford
University Press.

Hand, Learned. 1958. The Bill of Rights. Cambridge: Harvard University Press.

Handberg, Roger. 1978. "Opinion Behavior on the United States Supreme Court 1965-
1977. " Jurimetrics Journal 18:369.

Harrell, Frank E. 2001. Regression Modeling Strategies: With Applications to Linear
Models, Logistic Regression, and Survival Analysis.

Hasluck, Micholas. 1998. "Deconstructing the High Court." Quadrant 42: 12.

Haultain, Lynne. 1997. "The Changing of the Guard on the High Court." In The Law
Report: Radio National Transcripts.

Hayne, Kenneth. 2002. "Owen Dixon." In The Oxford Companion to the High Court of
Australia, ed. T. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

 

. 2004. "Advocacy in the High Court of Australia." Speech delivered to Western
Bar Association on.

Hayne, Kenneth M. 2003. "The High Court of Australia." Speech delivered to Round
Table of the International Association of Constitutional Law on Oct. 4, at
Melbourne.

280

 

Haynie, Stacia L, and C. Neal Tate. 1998. "Studying Judicial Behavior Comparatively:
Theory and Methods, Sermons and Lessons."

Haynie, Stacia L. 1992. "Leadership and Consensus on the US. Supreme Court." Journal
of Politics 54:1158.

Haynie, Stacia L., Reginald S. Sheehan, Donald R. Songer, and C. Neal Tate. 2008. High
Courts Judicial Database 2007 [cited March 1 2008]. Available fi'om
http://sitemason.vanderbilt.edu/site/dSYnTZ/data_sets.

Haynie, Stacia L., C. Neal Tate, Reginald S. Sheehan, and Donald R. Songer. 2001.
"Winners and Losers: A Comparative Analysis of Appellate Courts and Litigation
Outcomes." Speech delivered to Annual Meeting of the American Political
Science Association on September 2, at San Francisco, CA.

Heerey, Peter. 2004. "Aesthetics, Culture, and the Whole Damn Thing." Law and
Literature 15 (3):295-312.

Heise, Michael. 2002. "The Past, Present, and Future of Empitical Legal Scholarship:
Judicial Decision Making and the New Empiricism." University of Illinois Law
Review 20022819.

Hellevik, Ottar. 2007. "Linear Versus Logistic Regression When the Dependent Variable
is a Di chotomy." Quality and Quantity Online First.

Hettinger, Virginia A., Stefanie A. Lindquist, and Wendy L. Martinek. 2003a.
"Acclimation Effects and Separate Opinion Writing on the United States Courts
of Appeals." Social Science Quarterly 84 (4):792.

 

. 2003b. "Separate Opinion Writing on the United States Courts of Appeals."
American Politics Research 31 (3):215-50.

 

. 2004. "Comparing Attitudinal and Strategic Accounts of Dissenting Behavior on
the US. Courts of Appeals." American Journal of Political Science 48 (1):123-37.

Heydon, Dyson. 2003. "Judicial Activism and the Death of the Rule of Law." Quadrant
47 (2)210.

Hirst, John. 1995. " Stuart Macintyre and Samuel Griffith: The Report of the Civics
Expert Group." Paper read at Conference of the Samuel Griffith Society, March
31, at Sydney, NSW.

 

. 1998. Discovering Democracy: A Guide to Government and Law in Australia.
Carlton, VIC: Curriculum Corporation.

Hochschild, Adam S. 2000. "The Modern Problem of Supreme Court Plurality Decision:
Interpretation in Historical Perspective." Journal of Law and Policy 4:261.

281

Hocking, Barbara Ann, and Alison Smith. 1996. "The Potential of the Law of Torts to
Assist in the Protection of Children." Murdoch University Electronic Journal of
Law 3 (1):7.

Hocking, Jenny. 1997. Lionel Murphy: A Political Biography. Melbourne: Cambridge
University Press.

Hodgins, Bruce W., Don 1. Wright, and W. H. Heick. 1978. Federalism in Canada and
Australia: The Early Years. Waterloo, ONT: Wilfiid Laurier University Press.

Hoekstra, Valerie, and Timothy R. Johnson. 2003. "Delaying Justice: The Supreme
Court's Decision to Hear Rearguments." Political Research Quarterly 56 (3)2351-
60.

Holloway, Ian. 2001. "Keith Arthur Aickin." In The Oxford Companion to the High
Court of Australia, ed. T. Blackshield, M. Coper and G. Williams. South
Melbourne: Oxford University Press.

Horton, Jonathan. 2002. "Justices' Associates: Some Observations." University of
Queensland Law Journal 22 (1):114.

Howard, J. Woodford. 1968. "On the Fluidity of Judicial Choice." American Political
Science Review 62 (1):43-56.

Hudson, Phillip, and Annabel Crabb. 2002. "NSW Judge to Sit on High Court." The Age,
Dec. 17.

Hughes, Charles Evans. 1928. The Supreme Court of the United States: Its Foundation,
Methods and Achievements, An Interpretation. New York: Columbia University
Press.

Hulme, S.E.K. 1994. "Hit and Myth in the Law Courts." Speech delivered to Upholding
the Australian Constitution on, at Brisbane, QLD.

Hunt, Erling Messer. 1930. American Precedents in Australian Federation. New York,

Hurwitz, Mark S., and Drew Noble Lanier. 2004. "I Respectfully Dissent: Consensus,
Agendas, and Policymaking on the US. Supreme Court, 1888-1999." Review of
Policy Research 21 (3):429-45.

Jackson, DE 2001. "The Australian Judicial System: Judicial Power of the
Commonwealth." University of New South Wales Law Journal 24 (3)259.

Jackson, David. 1996. "The ROle of the Chief Justice: A View from the Bar. " In Courts of
Final Jurisdiction: The Mason Court in Australia, ed. C. Saunders. Annandal e,
NSW: Federation Press.

282

 

. 2002a. "The Brennan Court." In The Oxford Companion to the High Court of
Australia, ed. T. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

 

. 2002b. "Leave to Appeal." In The Oxford Companion to the High Court of
Australia, ed. T. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

Jackson, David, and Joan Priest. 2002. "Harry Talbot Gibbs." In The Oxford Companion
to the High Court of Australia, ed. T. Blackshield, M. Coper and G. Williams.
South Melbourne: Oxford University Press.

Jackson, Elizabeth. 2003. High Court Discussion Panel. ABC Online, October 6, 2003
2003 [cited October 7 2003]. Available from
http://www.abc.net.au/worldtoday/content/2003/s960836.htrn.

Jefferson, Thomas. 1822. Seriatim Opinions and the History of Parties, October 27.

Johansen, Saren, and Katarina Juselius. 1990. "Maximum Likelihood Estimation and

Inferences in Cointegration - with Applications to the Demand for Money."
Oxford Bulletin of Economics and Statistics 52:169-210.

Johnson, Charles A. 1979. "Lower Court Reactions to Supreme Court Decisions: A
Quantitative Examination." American Journal of Political Science 23 (4):792.

 

. 1987. "Law, Politics, and Judicial Decision MAking: Lower Federal Court Uses
of Supreme Court Decisions." Law & Society Review 212325—40.

Johnson, Rebecca, and Marie-Claire Belleau. 2006. "Judicial Dissent at the Supreme
Court of Canada: Integrating Qualitative and Quantitative Empirical Approaches."
Paper read at First Annual Conference on Empirical Legal Studies, June 30.

Johnson, Timothy R. 1997. "Oral Arguments, Information, and Supreme Court Decision
Making." Paper read at Annual Conference on the Scientific Study of Judicial
Politics, at Atlanta, GA. '

 

. 2003. "The Supreme Court, the Solicitor General, and the Separation of
Powers." American Politics Research 31 (4):426-51.

 

. 2004. Oral Arguments and Decision Making on the United States Supreme
Court. Albany, NY: SUNY Press.

Jones, Frank. 2002. "Unreported Judgments." In The Oxford Companion to the High
Court of Australia, ed. A. Blackshield, M. Coper and G. Williams. South
Melbourne: Oxford University Press.

Joseph, S., and M Castan. 2001. Federal Constitutional Law: A Contemporary View.
Pyrmont, NSW: Lawbook.

283

Jucewicz, Joseph, and Lawrence Baum. 1990. "Workload Influences on Supreme Court
Case Acceptance Rates, 1975-1984." Western Political Quarterly 43 (1): 123-35.

Kalowski, Henrik. 2002. "Mary Genevieve Gaudron." In The Oxford Companion to the
High Court of Australia, ed. T. Blackshield, M. Coper and G. Williams. South
Melbourne: Oxford University Press.

Kelman, Maurice. 1985. "The Forked Path of Dissent." The Supreme Court Review
19851227.

Kenny, Susan. 2004. "Interveners and Amici Curiae in the High Court." Speech delivered
to Public Law Weekend on, at Canberra, ACT.

Kildea, Paul. 2003. "Federal Civil Justice System Strategy Paper. " Canberra: Attomey-
General's Department.

Kirby, Michael. 1983. Reform the Law: Essays on the renewal of the Australian Legal
System. Melbourne: Oxford University Press.

 

. 1992. "Ex Tempore Reasons." Australian Bar Review 9:93.

 

. 1996. "AP. Mason - From Trigwell to Teoh." Melbourne University Law
Review 20:1087.

 

. 1998a. "Kitto and the High Court of Australia - Change and Continuity." Speech
delivered to Sir Frank Kitto Lecture on, at Armidale.

 

. 1998b. "Murray Gleeson - Law Student." Speech delivered to Funtion to Honour
the Appintrnent to the Office of Chief Justice of Australia the Hon. Murray
Gleeson AC on July 16, at Sydney.

 

. 1998c. "Times of Transition." Speech delivered to Kitto and the High Court of
Australia - Change and Continuity on May 22, at Armidale.

 

. 1999. "Constitutional Interpretation and Original Intent - A Form of Ancestor
Worship? " Speech delivered to University of Melbourne Law Students' Society
on September 9, at Melbourne.

 

. 2001. "Good News, Bad News, No News." Speech delivered to Australian
Institute of Judicial Administration on, at Sydney.

. 2003. Interview. Canberra, ACT, ?

 

 

. 2004. "A Blaze in the Sky - the Centenary Conference of the High Court of
Australia." Speech delivered to University of Notre Dame Law Review on
February, at South Bend, IN.

284

. 2005a. "Dissent and the Irnporance of Judicial Diversity." Speech delivered to
Institute of Judicial Studies of New Zealand Conference: Equality and Diversity
in Our Community on Aug. 18-19, at Rotorua, NZ.

 

 

. 2005b. "Isaac Isaacs: A Sesquicentenary Reflection." Speech delivered to
Samuel Alexander Lecture on Aug. 4, at Melbourne.

 

. 2005c. "Judicial Dissent." Speech delivered to Law Students' Society of James
Cook University on February 26, at Cairns.

 

. 2006. "Dame Ann Ebsworth Remembered." Speech delivered to Dame Ann
Ebsworth Memorial Lecture on Feb. 32, at London.

 

. 2007a. "Maximising Special Leave Performance in the High Court of Australia."
University of New South Wales Law Journal 30 (3 ).

 

. 2007b. "Twelve Years in the High Court - Continuity and Change." Paper read
at Michael Kirby Lecture Series, March 30, at Lismore, AU.

Kirrnan, Igor. 1995. " Standing Apart to Be a Part: The Precedential Value of Supreme
Court Concurring Opinions. " Columbia Law Review 95 (8)2083.

Kirwin, Lucy, and Jeremy Masters, eds. 2002. Australian Guide to Legal Citation. 2nd
ed. Melbourne: Melbourne University Law Review.

Kluger, Richard. 1976. Simple Justice. New York: Random House.

Kneebone, Susan. 1998. Tort Liability for Public Authorities. Sydney: LBC Information
Services.

Knoke, David, and Peter J. Burke. 1980. Log-Linear Models. Vol. 20. Beverly Hills, CA:
Sage Publications.

Kobetsky, Michael, and Rick Krever. 2002. "Taxation Law." In The Oxford Companion
to the High Court of Australia, ed. T. Blackshield, M. Coper and G. Williams.
South Melbourne: Oxford University Press.

Kort, Fred. 195 7. "Predicting Supreme Court Decisions Mathematically." American
Political Science Review 51 :1.

 

. 1963. " Simultaneous Equations and Boolean Algebra in the Analysis of Judicial
Decisions." Law and Contemporary Problems 28:143.

Krishnakumar, Anita S. 1999. "On the Evolution of the Canonical Dissent." Rutgers Law
Review 52:781.

Kwiatkowski, Denis, Peter C. B. Phillips, Peter Schmidt, and Yongcheol Shin. 1992.
"Testing the Null Hypothesis of Stationarity Against the Alternative of a Unit

285

Root: How sure are we that economic time series have a unit root? " Journal of
Econometrics 54:159-78.

Kyri akakis, Joanna. 2005. "Freeport in West Papua: Bringing Corporations to Account
for International Human Rights Abuses Under Australian Criminal and Tort
Law." Melbourne UniversityLaw Review 31 (l):95.

La Nauze, John Andrew. 1972. The Making of the Australian Constitution. Carlton:
Melbourne University Press.

Lani er, Drew Noble, and Sandra L. Wood. 2001. "Moving On Up: Institutional Position,
Politics, and the Chief Justice." American Review of Politics 22293-127.

Lavarch, Michael. 1996. "The Court, the Parliament and the Executive." In Courts of
Final Jurisdiction: The Mason Court in Australia, ed. C. Saunders. Annandale,
NSW: Federation Press.

Lax, Jeffrey R., and Charles M. Cameron. 2007. "Bargaining and Opinion Assignment on
the US. Supreme Court." Journal of Law, Economics, & Organization 23
(2):276-302.

LCA. 1997. "Formal Consultation Process for High Court Appointments Unnecessary."
Canberra: Law Council of Australia.

 

. 2001. "Challenges for the Legal Profession. " In Law Council of Australia
Discussion Paper. Canberra.

 

. 2003. "Snapshot of the Legal Profession." Canberra: Law Council of Australia.

Lebo, Matthew J ., Robert W. Walker, and Harold D. Clarke. 2000. "You Must
Remember This: Dealing With Long Memory in Political Analyses." Electoral
Studies 19 (1)231-48.

Lehane, John. 2002. "W. M. C. Gummow." In The Oxford Companion to the High Court
of Australia, ed. T. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

Leigh, Andrew. 2000. "Behind the Bench: Associates in the High Court of Australia."
Alternative Law Journal 25 (6):295-99.

Lijphart, Arend. 1999. "Australian Democracy: Modifying Majoritarianism?" Australian
Journal of Political Science 34 (3):313-26.

Lindgren, Kevin E. 2007. "Is the Australian Law Journal an Australian Law Journal?"
Paper read at The Judges' Review Conference, Mar. 16, at Sydney.

286

Lindsay, David. 2002. "Protection of Privacy Under the General Law Following AB. C. v.
Lenah Game Meats Pty. Ltd : Where to Now?" Privacy Law and Policy Reporter
2002245.

Luntz, Harold. 2002. "Tort Law." In The Oxford Companion to the High Court of
Australia, ed. A. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

Luntz, Harold, and D. Hambly. 2006. T orts: Cases and Commentary. 5th rev. ed.
Sydney: LexisNexis Butterworths.

Lynch, Andrew. 2002a. "Dissent: Towards a Methodology for Measuring Judicial
Disagreement in The High Court of Australia." Sydney Law Review 24:470.

 

. 2002b. "Dissenting Judgments." In The Oxford Companion to the High Court of
Australia, ed. T. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

 

. 2003a. "Dissent: The Rewards and Risks of Judicial Disagreement in the High
Court of Australia." Melbourne University Law Review 29.

 

. 2003b. "The Gleeson Court on Constitutional Law - An Empirical Analysis of its
First Five Years." University of New South Wales Law Journal 26:32.

Lynch, Andrew, and George Williams. 2005. "The High Court on Constitutional Law:
The 2004 Statistics." Paper read at Constitutional Law Conference, February 18,
at Sydney.

MacKinnon, James G. . 1991. "Critical Values for Cointegration Tests." In Long-Run
Economic Relationships: Readings in C ointegration, ed. R. F. Engle and C. W. J.
Granger. Oxford: Oxford University Press.

Maddala, G. S. 1983. Limited-Dependent and Qualitative Variables in Econometrics.
New York, NY: Cambridge University Press.

Maddala, G. S., and In-Moo Kim. 1998. Unit Roots, Cointegration, and Structural
Change. Cambridge: Cambridge University Press.

Magalhaes, Pedro C. 1998. "Judicial Behavior in Constitutional Courts: The Case of
Portugal." Speech delivered to Scientific Study of Judicial Politics on, at East
Lansing.

Maitra, Pushkar, and Russell Smyth. 2005. "Determinants of Retirement on the High
Court of Australia." Economic Record 81 (254):193.

Maltese, John Anthony. 1998. The Selling of Supreme Court Nominees. Edited by M.
Nelson. Paperback ed. Baltimore, MD: Johns Hopkins University Press. .

287

Maltzman, Forrest, James F. Spriggs, II, and Paul J. Wahlbeck. 2000. CraflingLaw on
the Supreme Court: The Collegial Game. Cambridge: Cambridge University
Press.

Maltzman, Forrest, and Paul J. Wahlbeck. 1996a. "May It Please the Chief: Opinion
Assignments in the Rehnquist Court." American Journal of Political Science
40:421-43.

 

. 1996b. "Strategic Policy Considerations and Voting Fluidity on the Burger
Court." American Political Science Review 90 (3):581-92.

 

. 2004. "A Conditional Model of Opinion Assignment on the Supreme Court."
Political Research Quarterly 57 (4):551-63.

 

. 2005. "Opinion Assignment on the Rehnquist Court." Judicature 89 (3): 121-28.

Marks, Brian. 1989. A Model of Judicial Influence on Congressional Policymaking:
Grove City College v. Bell. Ph.D. Dissertation, Washington University.

Marr, David. 1980. Barwick. Sydney: Allen & Unwin.

Marriner, Cosima. 2007. "Straight-Shooter Jumps the Bar." Sydney Morning Herald,
Aug. 18.

Martin, Paul, and Stuart Soroka. 2004. "Dissent and the Decline of Unanimity on the US.
Supreme Court: Methodological Issues and Alternative Explanations." Paper read
at Annual Meeting of the American Political Science Association, September, at
Chicago, IL.

Mason, Anthony. 2000. "The Evolving Role and Function of the High Court of
Australia." In The Australian Federal Judicial System, ed. B. Opeskin and F.
Wheeler. Melbourne: Melbourne University Press.

 

. 2002a. "The Barwick Court." In The Oxford Companion to the High Court of
Australia, ed. T. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

 

. 2002b. "The Griffith Court." In The Oxford Companion to the High Court of
Australia, ed. T. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

 

. 2002c. "Judiciary Act." In The Oxford Companion to the High Court of
Australia, ed. T. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

 

. 2002d. "Role of the Chief Justice." In The Oxford Companion to the High Court
of Australia, ed. A. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

288

 

. 2003a. "Celebrating the High Court's Centenary." Speech delivered to Centenary
of the High Court Conference on 21 February, at Sydney, NSW.

 

. 2003b. "The High Court of Australia: A Personal Impression of its First 100
Years." Melbourne University Law Review 2003 :33.

 

. 2003c. Interview. Sydney, March 10.

Matson. 1993. "The Common Law Abroad: English and Indigenous Laws in the British
Commonwealth." International and Comparative Law Quarterly 42 (4):753-79.

Maveety, Nancy. 1996. Justice Sandra Day O'Connor: Strategist on the Supreme Court.
Lanham, MD: Rowman & Littlefield.

 

. 2003. "Concurrence and the Study of Judicial Behavior in American Political
Science. " Juridica International 12173-85.

Mazzocchi, Jo. 2003. Australia's High Court All Male Again. Canberra, February 11.

Mc Iver, John P. 1976. "Scaling Judicial Decision: The Panel Decisionmaking Process of
the US. Courts of Appeals." American Journal of Political Science 20 (4):749.

McCarthy, Alana 2004. "The Evolution of the Role of the Attorney-General." Murdoch
University Electronic Journal of Law 1 l (4):30.

McCormick, Peter James. 1994. "Career Patterns and the Delivery of Reasons for
Judgment in the Supreme Court of Canada (1949-1993)." Supreme Court Law
Review 5 (2)2499-521.

McDonald, Leighton. 2002. "The Myth of Legal Merit and Ability." Canberra Times,
December 23.

McGarvie, Richard D. 1999. Democracy: Choosing Australia's Republic. Melbourne:
University of Melbourne Press.

McGinley, GP]. 1987. "The Search for Unity: The Impact of Consensus Seeking
Procedures in Appellate Courts." Adelaide Law Review 11:203.

McGuire, Kevin T. 1994. The Supreme Court Bar: Legal Elites in the Washington
Community. 1 ed. Charlottsville: University of Virginia Press.

McGuire, Kevin T., Charles E. Smith, Jr., and Gregory A. Caldeira. 1999. "Ambiguities
in Measuring and Modeling the US. Supreme Court." Speech delivered to
American Political Science Association on, at Atlanta.

McHugh, Michael. 2002. "The High court and the Oxford Companion to the High
Court." Speech delivered to Constitutional Law Conference on February 15, at
Sydney.

289

. 2004. "The Constitutional Jurisprudence of the High Court: 1989-2004." Speech
delivered to Sir Anthony Mason Lecture in Constitutional Law on Nov. 26, at
Sydney, NSW.

 

. 2005. "Working as a High Court Justice." Speech delivered to Women Lawyers
Association of New South Wales on 17 August, at Newcastle.

 

McIntyre, Greg, David Ritter, and Paul Sheiner. 1999. "Administrative Avalanch: The
Application of the Registration Test under the Native Title Act 1993 (Cth)."
Indiginous Law Bulletin 4 (20):8-12.

McLaughlin, John Kennedy. 2002. "Cyril Ambrose Walsh." In The Oxford Companion to
the High Court of Australia, ed. T. Blackshield, M. Coper and G. Williams. South
Melbourne: Oxford University Press.

McMillan, John. 2002. "Judicial Restraint and Activism in Administrative Law." Federal
Law Review 30 (2):335.

 

. 2003. Interview. Canberra, ACT, 24 February.

Meagher, Dan. 2002. "Guided By Voices? Constitutional Interpretation on the Gleeson
Court." Deakin Law Review 2002214.

Meek, Michael. 1999. The Australian Legal System. 3 ed. Pyrmont, NSW: LBC
Information Services.

Miles, Jeremy, and Mark Shelven. 2001. Applying Regression and Correlation: A Guide
for Student and Researchers. Thousand Oaks, CA: Sage.

Milionis, Alexandros E. 2004. "The Importance of Variance Stationarity in Economic
Time Series Modelling: A Practical Approach." Applied Financial Economics 14
(4):265-78.

Mishler, William, and Reginald S. Sheehan. 1993. "The Supreme Court as a
Counterrnaj oritarian Institution? The Impact of Public Opinion on Supreme Court
Decisions." American Political Science Review 87287-101.

 

. 1996. "Public Opinion, the Attitudinal Model, and Supreme Court Decision
Making: A Micro-Analytic Perspective." Journal of Politics 58 (1)2169-200.

Moens, Gabriel A. 1997. "The Role of the States in High Court Appointments." Speech
delivered to Upholding the Australian Constitution on March 7-9, at Canberra.

Moisidis, Cosmas. 2002. "Achieving World's Best Practice in the Writing of Appellate
Judgments." Law Institute Journal 76 (10):30.

290

Murphy, Lionel. 1980. "The Responsibility of Judges." In Law, Politics and the Labor
Movement, ed. G. Evans. Melboum: Legal Service Bulletin in Association with
the Australian Society of Labor Lawyers.

 

. 1983. "National Press Club Speech on August 17, at Canberra.

Murphy, Walter F. 1964. Elements of Judicial Strategy. Chicago: University of Chicago
Press.

Myers, R. F. 1990. Classical and Modern Regression with Applications. Boston: PW S-
KENT.

Nagel, Steward S. 1961. "Political Party Affiliations and Judges' Decisions." American
Political Science Review 55:843-50.

Narayan, Paresh Kumar, and Russell Smyth. 2005. "The Consensual Norm on the High
Court of Australia: 1904-2001 ." International Political Science Review 26
(2)2147-68.

 

. 2007. "What Explains Dissent on the High Court of Australia? An Empirical
Assessment Using a Cointegration and Error Correction Approach." Journal of
Empirical Legal Studies 4 (2)2401-25.

Ng, Serena, and Pierre Perron. 1995. "Unit Root Tests in ARMA Models with Data-
Dependent Methods for the Selection of the Truncation Lag." Journal of the
American Statistical Association 90:268-81.

Novak, Linda. 1980. "The Precedential Value of Supreme Court Plurality Decisions."
Columbia Law Review 80 (4)2756.

O'Brien, David. 1985. "Managing the Business of the Supreme Court." Public
Administration Review 45 2667-78.

O'Brien, David M. 1999. "Institutional Norms and Supreme Court Opinions: On
Reconsidering the Rise of Individual Opinions. " In Supreme Court Decision-
Making: New Institutionalist Approaches, ed. C. W. Clayton and H. Gillman.
Chicago: University of Chicago Press.

O'Brien, Robert M. 2007. "A Caution Regarding Rules of Thumb for Variance Inflation
Factors." Quality and Quantity 41 (5)2673 -90.

Ostberg, C. L., and Matthew E. Wetstein. 2004. "Equality Cases and the Attitudinal
Model in the Supreme Court of Canada." Paper read at Canadian Political Science
Association Meeting,Jun. 5, at Winnipeg.

Ostberg, C. L., Matthew E. Wetstein, and Craig R. Ducat. 2002. "Attitudinal Dimensions
of Supreme Court Decision Making in Canada: The Larner Court 1991-1995."
Political Research Quarterly 55 (1):235-56.

291

 

. 2003. "Acclimation Effects on the Supreme Court of Canada: A Cross-Cultural
Examination of Judicial Folklore." Social Science Quarterly 84 (3):704—22.

Pacelle, Richard L., and Bryan W. Marshall. 2001. "Strategic or Sincere? The Supreme
Court in Constitutional Decisions." Paper read at American Political Science
Association, Aug. 30 - Sep. 3, at San Francisco, CA.

Pampel, Fred C. 2000. Logistic Regression: A Primer. Edited by M. S. Lewis-Beck.
Thousand Oaks: Sage Publications.

Parish, Ken. 2000. "The High Court and the Nature of Judicial Power." In Australian
Public Law. Darwin.

Patapan, Hai g. 2000a. "Governance and the High Court." In Institutions on the Edge?
Capacity for Governance, ed. M. Keating, J. Wanna and P. Weller. St. Leonards:
Allen & Unwin.

 

. 2000b. Judging Democracy : The New Politics of the High Court of A ustralia.
Cambridge [England] ; New York: Cambridge University Press.

 

. 2000c. "The Politics of Interpretation." Sydney Law Review 22:247.

 

. 2002. "High Court Review 2001: Politics, Legalism and the Gleeson Court."
Australian Journal of Political Science 37 (2):241-53.

Pearson, Linda. 2007. "The Impact of External Administrative Law Review: Tribunals."
In University of New South Wales Faculty of Law Research Series. Sydney.

Pelly, Michael. 2005. "McHugh's Angels: 10 Women Fit for High Court, Says Top
Judge." Sydney Morning Herald, Aug. 19.

 

. 2008. "Kirby the Outsiders Lonely Last Lap on a Bench where Legalism Rules."
The Australian, Feb. 8.

Perry, H. W., Jr. 1994. Deciding to Decide: Agenda Setting in the United States Supreme
Court: Harvard University Press.

Perry, H.W. 1991. Deciding to Decide: Agenda Setting in the United States Supreme
Court. Cambridge: Harvard University Press.

Peterson, Steven A. 1981. "Dissent in American Courts." Journal of Politics 43 (2)1412.

Phillips, Peter C. B., and Pierre Perron. 1988. "Testing for a Unit Root in Time Series
Regression." Biometrika 75 (2):335-46.

Pierce, Jason L. 2006. Inside the Mason Court Revolution. Durham, NC: Carolina
Academic Press.

292

Pinello, Daniel R. 1999. "Linking Party to Judicial Ideology in American Courts: A
Meta-analysis. " Justice Systems Journal 20 (3)221 9-5 7.

Popple, James. 1999. "Case Management in the High Court of Australia: The First Year
of a New System." Paper read at Sixth National Court Technology Conference,
14-16 September, at Los Angeles.

. 2003. "Number of Justices." In The Oxford Companion to the High Court of
Australia, ed. A. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

 

Posner, Richard A. 1995. Overcoming Law. Cambridge, MA: Harvard University Press.

Posner, Richard A., and Eric Bennett Rasmusen. 1999. "Creating and Enforcing Norms,
with Special Reference to Sanctions." International Review of Law and
Economics 19 (3):369-82.

Post, Robert. 2001. "The Supreme Court Opinion as Institutional Practice: Dissent, Legal
Scholarship, and Decisionmaking in the Taft Court." Minnesota Law Review
85: 1267.

Powell, Lewis F., Jr. 2004. "What Really Goes on at the Supreme Court." In Judges on
Judging: Views fiom the Bench, ed. D. O'Brien. Washington, DC: CQ Press.

Priest, Susan. 2002. "Strike of 1905." In The Oxford Companion to the High Court of
Australia, ed. T. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

Pritchett, C. Herman. 1948. The Roosevelt Court: A Study in Judicial Politics and
Values, 193 7-194 7. New York: Macmillian.

 

. 1954. Civil Liberties and the Vinson Court. Chicago: University of Chicago
Press.

 

. 1961. Congerss versus the Supreme Court: University of Minnesota Press.

Przeworski, Adam. 1995. "The Role of Theory in Comparative Politics: A Symposium."
World Politics 48: 1-49.

Przeworski, Adam , and Henry Teune. 1970. The Logic of Comparative Social Inquiry.
Malabar, Fl.: Krieger Publishing Company.

Quick, John, and Robert Randolph Garran. 1901. The Annotated Constitution of the
Australian Commonwealth. Sydney: Angus & Robertson.

Ramseyer, J. Mark. 1994. "The Puzzling (In) Dependence of Courts: A Comparative
Approach." Journal of Legal Studies 23 2721-47.

293

Ramseyer, J. Mark, and Eric B. Rasmusen. 1997. "Judicial Independence in a Civil Law
Regime: The Evidence from Japan." Journal ofLaw, Economics, & Organization
13 (2):256-86.

Rehnquist, William H. 1983. "Oral Advocacy: A Disappearing Art." Mercer Law Review
35: 1015-28.

 

. 2004. "The Supreme Court's Conference." In Judges on Judging: Views from the
Bench, ed. D. O'Brien. Washington, DC: CQ Press.

Richards, Alexandra. 2000. "Sir Harry Gibbs Wrong on Merit in Judicial Appointments."
Canberra: Australian Women Lawyers.

Riker, William H. 1962. The Theory of Political Coalitions. New Haven: Yale University
Press.

Robinson, Peter M. 1995. "Gaussian Semiparametric Estimation of Long Range
Dependence." The Annals of Statistics 23 (5): 1630-61.

Rohde, David. 1972a. "Policy Goals and Opinion Coalitions in the Supreme Court."
Midwest Journal of Political Science 16 (2)2208.

 

. 1972b. "Policy Goals, Strategic Choice and Majority Opinion Assignments in
the US. Supreme Court." Midwest Journal of Political Science 16 (4):652.

 

. 1977. "Some Clarifications Regarding a Theory of Suprrne Court Coalition
Formation. " American Journal of Political Science 21 (2):405.

Rose, Richard, and Doh Chull Shin. 2001. "Democratization Backwards: The Problem of
Third-Wave Democracies." British Journal of Political Science 30 (2):299-3 22.

Rosenberg, Gerald N., and John M. Williams. 1997. "Do Not Go Gently into That Good
Right: The First Amendment int he High Court of Australia." The Supreme Court
Review 1997:439-195.

Rubenstein, Kim. 2002. "In High Court Selection, Like Promotes Like." The Age,
December 20.

Said, S. E., and David A. Dickey. 1984. "Testing for Unit Roots in Autoregressive
Moving-Average Models with Unknown Order." Biometrika 71:599-607.

Sartori, Giovanni. 1970. "Concept Mi sinformation in Comparative Politics." American
Political Science Review 64 (4): 1033-53.

Sawer, Geoffrey. 1957. Cases on the Constitution of the Commonwealth of A ustralia. 2nd
ed. Sydney: The Law Book Co. of Australasia.

. 1975. The Australian Constitution. Canberra: Australian Govt. Pub. Service.

 

294

Sawer, Marian. 1999. "Overview: Institutional Design and the Role of the Senate." In
Representation and Institutional Change: 50 Years of Proportional
Representation in the Senate, ed. M. Sawer and S. Miskin. Canberra, ACT:
Department of the Senate.

Sawyer, Geoffrey. 1967. Australian Federalism in the Courts. Melbourne: Melbourne
University Press.

Schaefer, Walter V. 1966. "Precedent and Policy." University of Chicago Law Review 34
(1):3-25.

Schmidhauser, John R. 1962. "Stare Decisis, Dissent, and the Background of the Justices
of the Supreme Court of the United States." University of Toronto Law Journal
14:194.

Schubert, Glendon. 1959. Quantitative Analysis of Judicial Behavior. Glencoe, Ill: The
Free Press.

 

. 1964. "The Power of Organized Minorities in a Small Group." Administrative
Science Quarterly 9 (2)2133-53.

 

. 1968. "Political Ideology on the High Court." Politics 3:21.

 

. 1969a. "Judicial Attitudes and Policy-Making in the Dixon Court." Osgood Hall
Law Journal 7:1.

 

. 1969b. "Two Causal Models of Decision-Making by the High Court of
Australia." In Comparative Judicial Behavior: C ross-C ultural Studies of Political
Decision-Making in the East and West, ed. G. Schubert and D. J. Danelski. New
York: Oxford University Press.

Schubert, Glendon A. 1965. The Judicial Mind : The Attitudes and Ideologies of Supreme
Court Justices, 1946-1963. Evanston, 111.: Northwestern University Press.

Schubert, Glendon, and David J. Danelski. 1969. Comparative Judicial Behavior: Cross-
Cultural Studies of Political Decision-Making in the East and West. New York:
Oxford University Press.

Schuck, Peter H., and E. Donald Elliot. 1990. "To the Chevron Station: An Empirical
Study of Federal Administrative Law." Duke Law Journal 1990:984.

Schwartz, Edward P. 1992. "Policy, Precedent, and Power: A Positive Theory of

Supreme Court Decision-Making. " Journal of Law, Economics and Organization
8 (2) :2 1 9.

Scott, John T., James F. Spriggs, and James R. Zink. 2003. "Courting the Public: Judicial

Behavior and Public Views of Court Decisions." Speech delivered to Annual
Meeting of the Midwest Political Science Association on, at Chicago.

295

Seddig, Robert G. 1991. "John Marshall and the Origins of Supreme Court Leadership."
Journal of Supreme Court History 1991 (1).

Segal, Jeffrey A. 1984. "Predicting Supreme Court Cases Probabilistically: The Search
and Seizure Cases, 1962-1981." American Political Science Review 78:891-900.

Segal, Jeffrey A., and Albert D. Cover. 1989. "Ideological Values and the Votes of US.
Supreme Court Justices." American Political Science Review 83 (2):557-65.

Segal, Jeffrey A., Harold. J. Spaeth, Lee Epstein, and Thomas G. Walker. 2007. The
Supreme Court Compendium: Data, Decisions & Developments. 4th ed: CQ
Press.

Segal, Jeffrey Allan, and Harold J. Spaeth. 1993. The Supreme Court and the Attitudinal
Model. Cambridge ; New York: Cambridge University Press.

 

. 2002. The Supreme Court and the Attitudinal Model Revisited. Cambridge ; New
York: Cambridge University Press.

Selway, Bradley. 2002. "The Use of History and Other Facts in the Reasoning of the
High Court of Australia." University of Tasmania Law Review 20 (2):129-58.

Selway, Bradley, and John M. Williams. 2005. "The High Court and Australian
Federalism." Publius 35 (3)2467.

Shapiro, Martin M. 1995. "From the Section Chair. " Law and Courts 5:1.

Shapley, L. S., and Martin Shubik. 1954. "A Method for Evaluating the Distribution of
Power in a Committee System." American Political Science Review 48:78 7-92.

Sheehan, Reginald S. 1990. "Administrative Agencies and the Court: A Reexamination
of the Impact of Agency Type on Decisional Outcomes." Western Political
Quarterly 43 :875-85.

 

. 1992. "Federal Agencies and the Supreme Court: An Analysis of Litigation
Outcomes, 1953-1988." American Politics Quarterly 20:478-500.

Sheldon, Charles H. 1967. "Public Opinion and High Courts: Communist Party Cases in
Four Constitutional Systems." Western Political Quarterly 20:341.

Sheller, Simon. 2002. "MD. Kirby." In The Oxford Companion to the High Court of
Australia, ed. T. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

Simpson, Amelia, and Troy Simpson. 2002. "Personal Relations." In The Oxford
Companion to the High Court of Australia, ed. T. Blackshield, M. Coper and G.
Williams. South Melbourne: Oxford University Press.

296

Simpson, Troy. 2002. "Conferences." In The Oxford Companion to the High Court of
Australia, ed. A. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

Smith, R. 2000. "Unit Roots and All That: The Impact of Time-Series Methods on
Macroeconomics." In Macroeconomics and the Real World, ed. R. Backhouse
and A. Salanti. New York: Oxford University Press.

Smithey, Shannon Ishiyama. 2002. "Activism on the South African Constitutional
Court." Speech delivered to Midwest Political Science Association on, at
Chicago.

Smyth, Russell. 2000. "The 'Haves' and the 'Have Nots': An Empirical Study of the
Rational Actor and Party Capability Hypotheses in the High Court 1948-99."
Australian Journal of Political Science 35 (2):255-74.

. 2001. "Judicial Interaction on the Latham Court: A Quantitative Study of Voting
Patterns on the High Court 193 5-1950." Australian Journal of Politics and
History 47 (3)2330-48.

 

. 2002a. "Acclimation Effects for High Court Justices 1903-1975." University of
Western Sydney Law Review.

 

. 2002b. "Explaining Voting Patterns on the Latham High Court 1935-50."
Melbourne University Law Review 200225.

 

 

. 2002c. "Historical Consensual Norms in the High Court." Australian Journal of
Political Science 37 (2):255-66.

. 2003a. "Case Complexity and Citation to Judicial Authority: Some Empirical
Evidence from the New Zealand Court of Appeal." Murdoch University
Electronic Journal of Law 10 (1).

 

 

. 2003b. "Explaining Historical Dissent Rates in the High Court of Australia."
Commonwealth and Comparative Politics 41 (2)283-1 14.

. 2004. "What Explains Variations in Dissent Rates?: Time Series Evidence from
the High Court." Sydney Law Review 262221.

 

 

. 2005a. "Do Judges Behave as Homo Economicus, and If So, Can we Measures
Their Performance? An Antipodean Perspective on a Tournament of Judges."
Florida State University Law Review 3221299-330.

 

. 2005b. "The Role of Attitudinal, Institutional and Environmental Factors in
Explaining Variations in the Dissent Rate on the High Court of Australia."
Australian Journal of Political Science 40 (4): 5 19-40.

297

Smyth, Russell, and J ayaprakash Narayan. 2004. "Hail to the Chiefl Leadership and
Structural Change in the Level of Consensus on the High Court of Australia."
Journal of Empirical Legal Studies 1 (2):399-427.

Snyder, Eloise. 1958. "The Supreme Court as a Small Group." Social Forces 36 (3)2232-
8.

 

. 1959. "Uncertainty and the Supreme Court's Decisions." American Journal of
Sociology 65 (3)2241.

Solomon, David Harris. 1999. The Political High Court : How the High Court Shapes
Politics. St Leonards, NSW, Australia: Allen & Unwin.

Songer, Donald R, and Stefanie A. Lindquist. 1996. "Not the Whole Story: The Impact of
Justices' Values on Supreme Court Decision Making." American Journal of
Political Science 1049-1 063.

Songer, Donald R. 1978. "The Relevance of Policy Values for the Confirmation of
Supreme Court Nominees." Law & Society Review 132927-49.

Songer, Donald R., and Martha Anne Humphries. 1999. "Law and Politics in Judicial
Oversight of Federal Administrative Agencies." Journal of Politics 61 (1):207-20.

Songer, Donald R., and Susan W. Johnson. 2002. "Attitudinal Decision Making in the
Supreme Court of Canada on.

 

. 2007. "Judicial Decision Making in the Supreme Court of Canada: Updating the
Personal Attribute Model." Canadian Journal of Political Science 40:91 1-3 4.

Spaeth, Harold J. 1989. "Consensus in the Unanimous Decisions of the US. Supreme
Court." Judicature 72 (5):274-92.

 

. 1997. The United States Supreme Court Judicial Database, 1953-1995 . Ann
Arbor: Inter-University Consortium for Political and Social Research.

Spaeth, Harold J ., and Douglas R. Parker. 1969. "Effects of Attitude Toward Situatio
upon Attitude toward Obj ect." Journal of Psychology 732173-82.

Spaeth, Harold J., and Jeffrey A. Segal. 2001. Majority Rule or Minority Will: Adherence
to Precedent on the US. Supreme Court. Cambridge: Cambridge University
Press.

Spigelman, James. 2004. "Anniversary Dinner Address." Speech delivered to Federal
Law Review - 40th Anniversary Dinner on Oct. 28, at Canberra, ACT.

Spriggs, James F., 11. 1996. "The Supreme Court and Federal Administrative Agencies: A

Resource-Based Theory and Analysis of Judicial Impact." American Journal of
Political Science 40 (4)21122-51.

298

 

. 2008. "An Empirical Analysis of the Trends, Determinants, and Effects of
Majority Opinions of the US. Supreme Court." In Selected Works of James F.
Spriffs II: ExpressO.

Spriggs, James F., Forrest Maltzman, and Paul J. Wahlbeck. 1999. "Bargaining on the
US. Supreme Court: Justices' Responses to Majority Opinion Drafts." Journal of
Politics 61 (2)2485.

Spriggs, James F., and Paul J. Wahlbeck. 1997. "Amicus Curiae and the Role of
Information at the Supreme Court." Political Research Quarterly 50 (2)2365.

Stager, Walter. 1925. "Dissenting Opinions, Their Purpose and Results." Virginia Law
Register 11 (7)395.

Stephens, Tony. 2002. Sir William Deane: The Things that Matter. Sydney: Hodder.

Stone, Adrienne. 2000. "The Freedom of Political Communication Since Lange." In The
High Court at the Crossroads: Essays in Constitutional Law, ed. A. Stone and G.
Williams. Leichhardt, N.S.W.: Federation Press.

Stone, Alec. 1992. The birth of judicial politics in France .' the Constitutional Council in
comparative perspective. New York: Oxford University Press.

Stras, David R. 2007. "Why Supreme Court Justices Should Ride Circuit Again."
Minnesota Law Review 91 : 1710-51 .

Stretton, Dean. 2005. "The Birth Torts: Damages for Wrongful Birth and Wrongful
Life." Deakin Law Review 16:10.

Tabachnick, Barbara G., and Linda S. Fidell. 2001. Using Multivariate Statistics. 4th ed.
Needham Heights, MA: Allyn & Bacon.

Tanenhaus, Joseph. 1966. "The Cumulative Scaling of Judicial Decisions." Harvard Law
Review 79 (8)21583-94.

Tarr, G. Alan. 2003. Judicial Process and Judicial Policymaking. 3rd ed. Belmont:
Thomson Wadsworth.

Tate, C. Neal. 1981a. "Judicial Institutions in Cross-National Perspective: Toward
Integrating Courts into the Comparative Study of Politics." In Comparative
Judicial Systems, ed. J. R. Schmidhauser: Butterworths.

 

. 1981b. "Personal Attribute Models of the Voting Behavior of US. Supreme
Court Justices: Liberalism in Civil Liberties and Economics Decisions, 1946-
1978." American Political Science Review 75 (2):355-67.

 

. 1983. "The Methodology of Judicial Behavior Research: A Review and
Critique." Political Behavior 5 (1)251 .

299

 

. 1992. "The Development of Comparative Judicial Politics." Perspectives on
Political Science 21 (3): 138-44.

 

. 1996. "Comparative Judicial Politics in the 19903." Speech delivered to Western
Political Science Association on, at San Francisco.

 

. 2002a. "Past, Present, and Future with the "Comparative Advantage": Part II,
1990-2002 Interlogue." Law and Courts Section Newsletter 12 (3)21-13.

 

. 2002b. "Past, Prsent, and Future with the "Comparative Advantage": Part I,
1892-1989." Law & Courts 12 (2):l-10.

Tate, C. Neal, and Panu Sittiwong. 1989. "Decision Making in the Canadian Suprrne
Court: Extending the Personal Attributes Model across Nations." Journal of
Politics 51 (4)2900-16.

Tate, C. Neal, Alec Stone, William Kitchin, Nils Stjernquist, Peter Russell, Mary L
Volcansek, Carl Baar, David J. Danelski, and Donald W. Jackson. 1990.
"Symposium on Judicial Review in Comparative Perspective." Policy Studies
Journal 19 (1):76-171.

Tate, C. Neal, and Torbj 'o’rn Vallinder. 1995. The Global Expansion of Judicial Power.
New York: New York University Press.

Thompson, Elaine. 1980. "The "Washminster" Mutation." Australian Journal of Political
Science 15 (2):32-40.

Thompson, J. 1982. "Principles and Theories of Constitutional Interpretation and
Adjudication: Some Preliminary Notes." Melbourne University Law Review
132603.

Twomey, Anne. 2002. "The Gibbs Court." In The Oxford Companion to the High Court
of Australia, ed. T. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

Tyree, Alan. 1977. "The Geometry of Case Law. " Victoria University of Wellington Law
Review 4:403.

 

. 1981. "Fact Content Analysis of Case Law: Methods and Limitations."
Jurimetrics Journal 22: 1.

Ulmer, S. Sidney. 1970. "Dissent Behavior and the Social Background of Supreme Court
Justices." Journal of Politics 32 (3):580-98.

 

. 1971. "Earl Warren and the Brown Decision." Journal of Politics 33 2689-702.

300

 

. 1973. "Social Background as an Inidcator to the Votes of Supreme Court
Justices in Criminal Cases: 1947-1956 Terms." American Journal of Political
Science 17:622-30.

Ulmer, S. Sidney, and William Nicholls. 1978. "The Integration of Dissent Behavior in
the United States Supreme Court." Jurimetrics Journal 19:170.

Verba, Sidney. 1967. "Some Dilemmas in Comparative Research." World Politics 20
(1):111-27.

Voigt, Stefan, Michael Ebeling, and Lorenz Blume. 2006. "Improving Credibility by
Delegating Judicial Competencewthe Case of the Judicial Committee of the Privy
Council." Journal of Development Economics 82 (2):348-73.

Volcansek, Mary L. 1993. Judicial Impeachment: None Called it Justice. Champaign:
University of Illinois Press.

von Nessen, Paul E. 1992. "The Use of American Precedents by the High Court of
Australia, 1901-1987." Adelaide Iflw Review 14:181.

Wahlbeck, Paul J., James F. Spriggs, II, and Forrest Maltzman. 1998. "Marshalling the
Court: Bargaining and Accommodation on the United States Supreme Court."
American Journal of Political Science 42 (1):294-315.

 

. 1999. "The Politics of Dissents and Concurrences on the US. Supreme Court "
American Politics Research 27 (4):488-517.

Wald, Patricia M. 1999. "A Response to Tiller and Cross." Columbia Law Review
99:23 5-61.

Walker, Bret. 2002a. "A.M. Gleeson." In The Oxford Companion to the High Court of
Australia, ed. T. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

Walker, Geoffrey de Q. 2002b. "The Seven Pillars of Centralism: Federalism and the
Engineers' Case." Speech delivered to Conference of the Samuel Griffith Society
on June 14-16, at Sydney.

Walker, Kristen. 2002c. "A. F. Mason." In The Oxford Companion to the High Court of
Australia, ed. T. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

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 50 (2):361-89.

301

Walker, Thomas G., and W. E. Hulbary. 1978. "Selection of Capable Justices." In The
First One Hundred Justices, ed. A. E. Blaustein and R. M. Mersky. Hampden,
CT: Shoe String Press.

Walsh, Peter. 1997. "Labor and the Constitution: Forty Years On." Paper read at
Conference of the Samuel Griffith Society, Oct. 26, at Perth.

Waltrnan, J erold L., and Kenneth M. Holland. 1987. The Political role of law courts in
modern democracies. New York: St. Martin's Press.

Wasby, Stephen L. 1970. The Impact of the United States Supreme Court: Some
Perspectives. Homewood, IL: The Dorsey Press.

Wells, William A. N. 1991. Law, Judges and Justice. North Ryde, NSW: Butterworths.

Wetstein, Matthew E., and C. L. Ostberg. 2005. "Strategic Leadership and Political
Change on the Canadian Supreme Court: Analyzing the Transition to Chief
Justice." Canadian Journal of Political Science 38 (3):653-73.

Wheeler, Fiona. 2003. Discussion. Canberra, ACT, 18 February 2003.

White, Byron R. 1982. "The Work of the Supreme Court: A Nuts and Bolts Description."
New York State Bar Journal 54:346.

Willheim, Ernst. 2002. "Collective Responsibility." In The Oxford Companion to the
High Court of Australia, ed. A. Blackshield, M. Coper and G. Williams. South
Melbourne: Oxford University Press.

Williams, George. 1999. "The High Court and the Media." University of Technology,
Sydney Law Review 1999 (1): 136-46.

 

. 2000. "The Amicus Curiae and Intervener in the High Court of Australia: A
Comparative Analysis." Federal Law Review 282365-402.

Williams, John M. 2002. "Lionel Keith Murphy." In The Oxford Companion to the High
Court of Australia, ed. A. Blackshield, M. Coper and G. Williams. South
Melbourne: Oxford University Press.

Winterton, George. 1998. "Barwi ck the Judge." University of New South Wales Law
Journal 21:109.

 

. 2002. "Garfield Edward John Barwick." In The Oxford Companion to the High
Court of Australia, ed. T. Blackshield, M. Coper and G. Williams. South
Melbourne: Oxford University Press.

Wood, Rebecca. 2002. "Dimensions of Decision-Making: Determining the Complexity of
Politics on the High Court of Australia." Speech delivered to Southern Political
Science Association on November 7-9, at Savannah, GA.

302

 

. 2008. "Institutional Considerations in Locating Norms of Consensus: A Cross-
National Investigation." Paper read at Annual Meeting of the Midwest Political
Science Association, Apr. 4, at Chicago, IL.

Wood, Sandra L., Linda Camp Keith, Drew Noble Lanier, and Ayo Ogundele. 1998a.
""Acclimation Effects" for Supreme Court Justices: A Cross-Validation, 1888-
1940." American Journal of Political Science 42 (2):690-7.

 

. 1998b. "The Supreme Court, 1888-1940: An Empirical Overview." Social
Science History 22 (2):201-24.

Woodward, Robert, and Scott Armstrong. 1979. The Brethren. New York: Avon Books.

Young, Neil J ., and Stephen McLeish. 2004. "Appellate Advocacy." Speech delivered to
Victorian Bar Compulsory Continuing Legal Education Program on December
14, at Victoria.

Yule, G. Udny. 1926. "Why do we Sometimes get Nonsense Correlations Between Time
Series: A Study in Sampling and the Nature of Time Series." Journal of the Royal
Statistical Society 89: 1-64.

Zines, Leslie. 1997. The High Court and the Constitution. 4th ed. Sydney: Butterworths.

 

. 2000. "The Present State of Constitutional Interpretation." In The High Court at
the Crossroads: Essays in Constitutional Law, ed. A. Stone and G. Williams.
Leichhardt, N.S.W.: Federation Press.

 

- . 2002a. "The Dixon Court." In The Oxford Companion to the High Court of
Australia, ed. T. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

 

. 2002b. "The Gleeson Court." In The Oxford Companion to the High Court of
Australia, ed. T. Blackshield, M. Coper and G. Williams. South Melbourne:
Oxford University Press.

 

. 2002c. "Legalism, Realism and Judicial Rhetoric in Constitutional Law." Speech
delivered to New South Wales Bar Association on October 16, at Sydney.

Zivot, Eric, and Jiahui Wang. 2003. "Modeling Financial Time Series with S-PLUS."
T echnometrics 45 (4):373-4.

 

. 2006. Modeling Financial Time Series with S-PL US: Springer.

Zorn, Christopher J. W. 2005. "A Solution to Separation in Binary Response Models."
Political Analysis 13 (2):]57-70.

303