University of Southern California Dissertations and Theses

The role of merits briefs in the generation of Supreme Court decisional language

(USC Thesis Other)

The role of merits briefs in the generation of Supreme Court decisional language

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content University of Southern California
Doctoral Dissertation
The Role of Merits Briefs in the
Generation of Supreme Court
Decisional Language
Author: Adam Feldman
Supervisor: Professor
Wayne Sandholtz
A dissertation submitted in fullment of the requirements
for the degree of Doctor of Philosophy (POLITICAL SCIENCE &
INTERNATIONAL RELATIONS)
from the
POLITICAL SCIENCE & INTERNATIONAL RELATIONS
FACULTY OF
THE USC GRADUATE SCHOOL
May 2017
Declaration of Authorship
I, Adam Feldman, declare that this thesis titled, `The Role of Merits Briefs
in the Generation of Supreme Court Decisional Language' and the work
presented in it is my own. I conrm that this work submitted for assessment
is my own and is expressed in my own words. Any uses made within it of
the works of other authors in any form (e.g., ideas, equations, gures, text,
tables, programs) are properly acknowledged at any point of their use. A
list of the references employed is included.
Signed:
Date:
i
ii
\Justice, sir, is the great interest of man on earth. It is the ligament which
holds civilized beings and civilized nations together."
Daniel Webster
iii
Abstract
Merits briefs are litigating parties' main means to convey their construction
of their cases in court. In the Supreme Court, they provide the Justices with
consolidated arguments and assessments of each case. In doing so they are
considered one of the most important inputs for the Justices' opinions and
specically for opinion language. While anecdotal evidence supports this
point, minimal empirical scholarship is devoted to verifying this premise.
This dissertation is designed to look expansively at the relationship between
Supreme Court merits briefs and opinions and in so doing it is designed ll
empirical gaps in our understanding of the inputs into Supreme Court opin-
ions. It also looks specically at the instances where merits briefs played
the largest role in dening the Court's opinion language. It is equally de-
signed to broaden the understanding of the relationship between briefs and
opinions with the addition of new indicators, variables, and data, and by
clarifying the mechanisms involved in the process linking briefs and opin-
ions. This dissertation is broken into three parts based interrelated analytic
goals: examining the relative impact of merits briefs on opinion language,
looking at the factors and mechanisms that are involved in this relationship,
and examining cases where merits briefs have the largest impact on Supreme
Court opinion language. This dissertation uses an original, large-N dataset
of over 9,000 briefs and their respective opinions to test hypotheses regard-
ing factors that make merits briefs more or less impactful on the Court's
opinions.
Keywords: Supreme Court, merits briefs, litigation, Justices, opinions,
language
Acknowledgements
I would like to express my sincere appreciation to my thesis chair, Profes-
sor Wayne Sandholtz, without whose guidance, patience, and expertise this
dissertation would not have been made possible. I would also like to ac-
knowledge my gratitude to the other members of my committee, Professors
Jeb Barnes, Daniel Klerman, and Dan Simon, for the assistance they pro-
vided at all levels of this research. Finally, I would like to thank my dear
friends and family for all of their support.
iv
Contents
Declaration of Authorship i
Abstract iii
Acknowledgements iv
Contents v
List of Figures viii
List of Tables ix
1 Introduction 1
1.1 Historical Background . . . . . . . . . . . . . . . . . . . . . . 6
1.2 The Modern Brief . . . . . . . . . . . . . . . . . . . . . . . . 8
1.3 The Empirical Study of Briefs . . . . . . . . . . . . . . . . . . 10
1.4 Relating Briefs to Opinions . . . . . . . . . . . . . . . . . . . 12
2 Merits Briefs, Amicus Briefs, and Lower Court Opinions 17
2.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.2 Written Opinions . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.3 The Justices' Toolkits . . . . . . . . . . . . . . . . . . . . . . 22
2.4 Data and Methods . . . . . . . . . . . . . . . . . . . . . . . . 27
2.4.1 Dependent Variables . . . . . . . . . . . . . . . . . . . 31
2.4.2 Independent Variables . . . . . . . . . . . . . . . . . . 32
v
Contents vi
2.5 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
2.5.1 Macro-Level . . . . . . . . . . . . . . . . . . . . . . . . 35
2.5.2 Justice-Specic Findings . . . . . . . . . . . . . . . . . 36
2.6 The Role of Merits Briefs . . . . . . . . . . . . . . . . . . . . 42
3 How Briefs and Opinions Relate Across Time 44
3.1 What's Inside: Focusing on the Briefs . . . . . . . . . . . . . 45
3.1.1 The Importance of Brief Quality . . . . . . . . . . . . 45
3.1.2 Understanding Quality . . . . . . . . . . . . . . . . . . 47
3.1.3 Empirical Understandings of Quality . . . . . . . . . . 55
3.1.4 The Role of Brief Quality . . . . . . . . . . . . . . . . 57
3.2 Additional Factors . . . . . . . . . . . . . . . . . . . . . . . . 58
3.2.1 Justices . . . . . . . . . . . . . . . . . . . . . . . . . . 59
3.2.2 Attorneys . . . . . . . . . . . . . . . . . . . . . . . . . 59
3.2.3 Cases . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
3.3 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
3.4 Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
3.4.1 Findings for the Justices . . . . . . . . . . . . . . . . . 74
3.5 Findings by Attorneys and Party Type . . . . . . . . . . . . . 79
3.6 Case Context . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
3.7 What Matters? . . . . . . . . . . . . . . . . . . . . . . . . . . 85
4 The Greatest Impact of Briefs on Opinion Language 88
4.1 Why Focus On Language . . . . . . . . . . . . . . . . . . . . 91
4.2 Measuring Language Overlap . . . . . . . . . . . . . . . . . . 94
4.2.1 Types of Language Shared Generally . . . . . . . . . . 94
4.3 The Cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
4.3.1 Common Source Relationships (High Percentage Over-
lap, Low Words Overlap) . . . . . . . . . . . . . . . . 100
4.3.2 Shared Understanding (Low Percentage Overlap, High
Words Overlap) and Typical Impact (Low Percentage
Overlap, Low Words Overlap) Relationships . . . . . . 103
4.3.3 Lifted Relationships (High Percentage Overlap and
High Word Overlap) . . . . . . . . . . . . . . . . . . . 111
4.3.3.1 Blackmun Cases . . . . . . . . . . . . . . . . 113
4.3.4 Non-Blackmun Cases . . . . . . . . . . . . . . . . . . . 163
4.4 Making Sense of This Relationship . . . . . . . . . . . . . . . 180
Contents vii
5 Conclusion 186
5.1 Main Findings . . . . . . . . . . . . . . . . . . . . . . . . . . 187
5.2 Practical Application . . . . . . . . . . . . . . . . . . . . . . . 188
5.3 Future Research . . . . . . . . . . . . . . . . . . . . . . . . . 188
Bibliography 190
List of Figures
2.1 Total Documents Filed Per Term . . . . . . . . . . . . . . . . 28
2.2 Documents Filed Per Case by Issue Area . . . . . . . . . . . . 29
2.3 Distribution of Overlap Value of All Filings Per Case . . . . . 29
2.4 Mean Overlap Value Per Term for All Filings . . . . . . . . . 30
2.5 Box Plots of Overlap for Parties' Briefs, Amicus Briefs, and
Lower Court Opinions . . . . . . . . . . . . . . . . . . . . . . 31
2.6 Mean Opinion Length by Justice . . . . . . . . . . . . . . . . 37
2.7 Mean Words Overlap by Justice . . . . . . . . . . . . . . . . . 38
2.8 Mean Party Brief Words Overlap by Justice . . . . . . . . . . 39
2.9 Mean Amicus Brief Words Overlap by Justice . . . . . . . . . 39
2.10 Mean Lower Court Words Overlap by Justice . . . . . . . . . 40
3.1 Histogram of % Language Overlap of All Briefs and Opinions:
1946-2014 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
3.2 Marginal Eects of Brief Quality on Overlap Values . . . . . 71
3.3 Box Plots of Language Overlap for All Justices: 1946-2014 . . 75
3.4 Fit Plots of Language Overlap Based on Logged Attorney
Experience . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
3.5 Box Plots of Language Overlap Between Petitioner's Briefs
by Party Type . . . . . . . . . . . . . . . . . . . . . . . . . . 81
3.6 Percent Dierence in Overlap Value Between US Government
and Other Party Types (Petitioners' Briefs) . . . . . . . . . . 82
3.7 Comparison of Language Overlap Between Briefs of Petitioner
and Respondents and Supreme Court Opinions (1946-2013) . 83
3.8 Comparison of Language Overlap by Issue Area . . . . . . . . 84
3.9 Mean Brief/Opinion Overlap Values (1946-2013) . . . . . . . 85
4.1 Typology of Impact of Brief on Opinion Language . . . . . . 98
viii
List of Tables
2.1 Multivariate Analysis . . . . . . . . . . . . . . . . . . . . . . . 35
3.1 Multilevel Model of the Eect of Brief Quality on the Amount
of Opinion Language Shared with Merits Briefs, 1946-2013 . . 69
3.2 Overlap Score Statistics by Top and Bottom 100 Brief Quality
Scores . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
3.3 Multilevel Probit Estimates of Likelihood of Winning . . . . . 72
3.4 Overlap Values and Quality Index by Attorney Experience
Level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
3.5 Mean Overlap and Standard Deviation by Justice . . . . . . . 76
3.6 Justice Ideological Sensitivity (of 2% or Greater) . . . . . . . 78
4.1 Typology of Impact of Brief on Opinion Language . . . . . . 112
ix
I dedicate this work to my mother, Sonia; wife, Gillian;
and children, Leo and Lily, for supporting me throughout
this journey; to the professors that have mentored me
along the way: Wayne Sandholtz, Jeb Barnes, Dan
Klerman, and Dan Simon; and in memory of my father
Lorne David Feldman.
x
Chapter 1
Introduction
[W]hat I am looking for in briefs is a roadmap for me to follow in
deciding the case. I am thinking about how I'm going to decide
the case how I am going to vote. I am thinking about what I
would say if I am assigned the opinion-how I am going to deal
with all of the issues. I am looking for as much help as I can get
from the advocates in the form of the brief, and then to clean up
anything else that is not taken care of in the briefs in the oral
argument.
| Justice Samuel A. Alito
Supreme Court merits briefs are a party's only uninterrupted opportunity
to convey their construction of a case to the Justices on the Supreme Court.
By the time the Court reaches oral argument, the Justices often have made
signicant inferences in cases based on the briefs and use the arguments to
test these inferences (Epstein, Landes and Posner 2010; Johnson, Black and
Wedeking 2009). Given parties' unique opportunities with their briefs, they
are often considered the most important tools in Supreme Court litigation
and one of the most important inputs for the Justices' opinions (Corley
1
2
2008). This input helps dene the Justices' central output: the language
in Supreme Court opinions. This is exemplied in Chief Justice Roberts's
statement (in Garner (2010a, 5)), \Language is the central tool of our trade.
When we're construing the Constitution, we're looking at words. Those are
the building blocks of the law. And so if we're not fastidious, as you put
it, with language, it dilutes the eectiveness and clarity of the law." Given
the signicance of Supreme Court opinion language, a prime component of
the Justices' and clerks' work is getting this language right. The dierence
between choices in language can be the dierence in whether entirely new
avenues of litigation are spawned or not.
And yet our understanding of merits briefs' impact on Supreme Court opin-
ions is minimal. Scholars only recently began utilizing empirical techniques
to study the relationship between briefs and opinions. Such studies show
that the Court's opinions often share more language with experienced liti-
gators and winning parties briefs than with those from novice litigators and
less successful parties (Corley 2008; Corley, Collins and Calvin 2011; Black
and Owens 2012).
These works suggest a causal path between briefs and opinions. While in
some instances causation is easy to infer, the relationship between briefs
and opinions is, in many cases, not linear or causal. In some cases language
from merits briefs comprises the basis for opinion language. In others it
bears only a passing resemblance to the language in the opinion. Similarly
in some instances the language shared between brief and opinion is mere
boilerplate, while in others it is at the heart of the clerks' and Justices'
analyses. The factors in
uencing the variation in this relationship have
remained predominately unexplored.
1
This leads to a lack of generalizability
1
The only quantitative selections aside from my own to look systematically at the
linguistic relationship between the Court's output and merits briefs did so for three terms
during the Rehnquist years (Corley 2008). Based on the sample size from this previous
section it is dicult to make broad generalizations from its ndings or to pinpoint specic
qualities of the Justices or attorneys.
3
and problems validating hypotheses. Beyond this existing empirical work
suggesting that language from merits briefs is a powerful input that impacts
language in Supreme Court, we currently must rely on anecdotal evidence.
Such anecdotes often take the form of judges' and lawyers' statements that
the merits brief is one of the most essential components of the modern
lawyer's arsenal. Examples of statements regarding briefs distinct impor-
tance, like the one from Justice Alito above, are plentiful. On this note,
Carter Philips (1998) who has argued more than seventy-ve cases before
the Supreme Court writes, \The decision-making process is 99.9% based on
briefs." Judge Baldock of the 10th Circuit (2001) acknowledges, \[y]our brief
is the controlling thing. Your oral argument is your chance to be persuasive
and convince us that your position is right." U.S. District Court Judge Mark
Kravitz (2009) declares that briefs must carry the \lion's share of the persua-
sion load" only supplemented by oral argument. Finally in a recent article
Long and Christensen (2013) similarly found that \almost every appellate
judge recognizes that appeals are usually decided on the briefs."
By contrast, there is a large body of literature on judicial behavior that sug-
gests briefs might not matter at all. A central nding of this literature is that
a justice's ideology is the primary determinant of judicial decision-making
(Segal and Spaeth 2002). If Justices decide cases based on their ideologies,
then briefs and oral argument amount only to support for the Justices' pre-
determinations. In addition, Segal and Spaeth argue that lawyers and judges
have a rational, utilitarian interest in maintaining the primacy of briefs in
litigation, even if this primacy is a fa cade.
Supporting this point, there are famous examples typied by the case of
Bush v. Gore,
2
where the Court predictably voted along ideological and
partisan lines, and where one can argue that briefs played little if any role
in determining the Justices' votes. In another example, in the case of Gideon
2
531 U.S. 98 (2000)
4
v. Wainwright,
3
pundits describe how the Court sought \a good vehicle"
to overturn outdated precedent implying that the Court sought a specic
outcome and only needed the proper set of facts in a case to bring this
outcome to fruition (with Gideon the Court overturned the holding of the
case Betts v. Brady
4
(Israel 1963)). In such situations, if the Justices make
up their minds on issues before reviewing the facts of specic cases, briefs
play an insubstantial role in in
uencing the Justices' votes.
The Justices' job of voting for decision outcomes, however, is only a small
piece of their work. The bulk of their time is spent deriving reasons to
support their decisions and this reasoning plays important roles in dening
legal boundaries and setting the standards that guide lower courts in their
interpretation of law.
The importance of opinion language is also clear as is evident from cases
themselves. In Planned Parenthood v. Casey,
5
for example, not only are the
words \armed" and \reversed" meaningful, but perhaps more importantly
the discussion of why the statute was held unconstitutional. To this end the
opinion states,
\This provision in eect requires women, as a condition of ob-
taining an abortion, to provide the Commonwealth with the pre-
cise information we have already recognized that many women
have pressing reasons not to reveal. Like the spousal notice
requirement itself, this provision places an undue burden on a
woman's choice, and must be invalidated for that reason."
6
3
372 U.S. 335 (1963)
4
316 U.S. 455 (1942)
5
505 U.S. 833 (1992)
6
505 U.S. 833, 901 (1992)
5
The specic wording of this opinion sets the standards for future lower
court decisions and for physicians' interpretations of proper abortion rights
(Songer, Segal and Cameron 1994; Hansford and Spriggs 2006).
One of the most direct ways that a brief can contribute to the Court's
decision is to provide language, or as Chief Justice Roberts puts it, \the
central tool of our trade." This Dissertation uses an original dataset of all
orally argued Supreme Court cases between 1946 and 2013 with available
briefs, a signed written opinion, and where exactly one petitioner/appellant
and one respondent/appellee submitted a brief on the merits.
7
From these
cases this Dissertation examines a total of 9,440 briefs and encompasses
the decisions of 36 dierent Supreme Court Justices. The dataset contains
information about each individual brief including the main attorney on the
brief, the attorney's law rm, and the state in which the attorney practices.
The goal of this Dissertation is to generate an understanding of when and
how briefs impact the Court's opinions. Along with explaining how merits
briefs are a main input for constructing opinions, this Dissertation shows
that the varying amount of brief language found in opinions can be at-
tributed to dierences specic to the Justices, attorneys, and to the type of
case.
The Dissertation proceeds as follow: the remainder of the introductory sec-
tion provides information about the historic and modern importance of mer-
its brief in the Supreme Court opinion construction process. Chapter Two
analyzes the importance of merits briefs relative to the two other main writ-
ten tools available to the Justices: amicus briefs and lower court opinions.
It does so using a set of all of the Roberts Court cases decided between
2005 and 2014. Chapter Three focuses specically on merits briefs to show
the means by which they impact Supreme Court opinion language. That
Chapter uses the comprehensive dataset of cases between the 1946 and 2013
7
The dataset itself is designed to be an original contribution to the scholarship.
6
Supreme Court terms. Chapter Four creates a typology based on the cases
in the large-N dataset to show the dierent types of impact briefs have on
Supreme Court opinions. It then uses this typology to select specic cases
to show the large and sometimes dominant role merits briefs play in dening
Supreme Court opinion language.
1.1 Historical Background
Current Supreme Court rules ensure that briefs are direct and succinct.
These rules limit parties' briefs on the merits to 15,000 words.
8
This gives
parties approximately fty pages to persuade the Court with their assess-
ments of facts and law, and the application of law to the facts. The written
brief and a thirty-minute oral argument,
9
which is oftentimes dominated by
the Justices' questions, encompass the party's full presentation of their case
to the Court (Epstein, Landes and Posner 2010).
This is a vastly dierent protocol than existed for much of the history of
the Supreme Court. At the time of McCullough v. Maryland,
10
for instance,
there were no page limits on briefs. Oral arguments also had few restrictions
which is apparent from McCullough's ten days of oral arguments (Martineau
1986). The shift to merits briefs as the centerpiece of parties' persuasive
eorts occurred alongside increasingly stringent time limitations placed on
oral arguments through the 19th and early 20th centuries (Rehnquist 1999).
The contents, form, and role of merits briefs in the Supreme Court also
evolved over the last several hundred years. In the early years of the United
States Supreme Court, appellate practice was based on practices from the
King's Bench in England which were predominately oral in nature (Unah
8
Sup. Ct. R. 33(1)(g) (2013)
9
Sup. Ct. R. 28(3) (2013)
10
17 U.S. 316 (1819)
7
2010). This changed slightly with the rst written requirement of \a state-
ment of the material points of the case" in 1795 (Cozine 1994). The new
rule, however, created only a limited, requisite written component. Briefs
were rst required in 1821 but the modern limits on the amount of material
that could be included were not put into place for decades (Hall, Ely and
Grossman 2005). With few initial requirements for briefs, their lengths var-
ied drastically. Some were so long that they probably did little to economize
the Justices' time. Over time, however, rules for briefs including require-
ments for specic components and page limits became more pronounced.
Briefs may not have initially function as the Justices had hoped when
they instated rules limiting oral arguments and heightening the importance
of briefs (Unah 2010; Ehrenberg 2004). Through slow shifts in the rules
though, by 1884 briefs began to resemble their modern counterparts. At
that time briefs were required to contain arguments, authorities, points of
law, statutes, and citations to pages in the record (Cozine 1994). Combined
with a quickly expanding docket, the shift to curtailed oral arguments with a
consequent greater emphasis on briefs became a necessity (Rehnquist 1999).
Other reasons aside from a desire for time-oriented eciency also initially led
to the prominent role of the written brief in the Supreme Court. There was
a limited number of attorneys trained in oral argument in the early United
States compared to the vast availability of barristers in England (Ehrenberg
2004; Martineau 1986). Justices were prone to illness and requiring them to
sit before extensive oral arguments meant nding time when a quorum was
suciently healthy for this endeavor (Frederick 2005). Culture and heritage
also likely played roles in the shift to the written brief. Americans had a
written Constitution while the British lacked the same. This distinction and
the prominence of the written word in early American culture played major
roles into the movement towards written briefs (Frederick 2005; Kravitz
2009).
8
In contrast to the British institutionalization of oral argument which cre-
ated path dependence making any major change in practice unlikely, the
United States began its advocacy practices with a blank slate (Ehrenberg
2004). With the advent of the printing press and an ethos bent on cultural
dierentiation from its British roots, the United States became a perfect
laboratory for experimenting with litigation based around the written brief
(Martineau 1986). These factors militating towards the centrality of the
written brief converged with the reality that as the Justices' mandatory
docket grew, they lacked sucient time to hear lengthy oral arguments in
all cases (Frederick 2005).
11
The Justices' interest in a more time ecient manner of case presentation
did not mean that briefs necesarily supplied the information the Justices'
sought. Instances like Justice Blackmun's dissent in New York Times v.
United States,
12
make this all the more evident as he wrote, \I therefore
would remand these cases to be developed expeditiously, of course, but on
a schedule permitting the orderly presentation of evidence from both sides,
with the use of discovery, if necessary, as authorized by the rules, and with
the preparation of briefs, oral argument, and court opinions of a quality
better than has been seen to this point."
1.2 The Modern Brief
An undercurrent in Supreme Court discourse questions what the Justices do
with the time they save by hearing over a hundred fewer cases per term than
they did during the 1980's (Starr 2005). Current Justices indicate that they
spend much of their time outside of the courtroom reviewing briefs. Justice
11
Supporting this point, Chief Justice Rehnquist (1983, 1019) stated, \as judicial time
steadily dwindled, the oral side of appeal, and the art of oral advocacy, came to play a
smaller and smaller part in the presentation of cases to appellate courts."
12
403 U.S. 713, 761-62 (1971)
9
Breyer (in Swain (2013)), for instance, said, \[w]e're reading continuously.
If I'm
ying somewhere, I'll have briefs in my bag. I read them on the
airplane. I read them at home. Briefs follow me around during the year."
What are Justice Breyer and the other Justices looking for when they spend
this time reading briefs? There are multiple denitions of the modern brief
that relate to the purposes it serves. With such variation in form over time
the fact that even the purpose of the merits brief is contested should come
as no surprise. The template for briefs is standard (with a few possible
variations) including a table of authorities, table of contents, introduction
with a question framing the case, statement of facts, summary of argument,
argument, conclusion and optional appendix. A current point of contestation
has to do with the expected tone of the brief - that is the extent the brief
should present an objective, neutral perspective compared to the extent it
should present a biased framework in favor of the litigating party (Cozine
1994). One conception is that the brief is designed to \inform" the court
about the facets of the case as well as the party's position (Llewellyn 1962).
This way of looking at the brief denes it as an object that gives the Court
an accurate assessment of the facts and issues involved in the case while
clearly conveying the reasons for the outcome sought by the drafting party.
A second position looks at the brief as more of a pointed, argumentative
piece. Under this characterization the brief is a \written argument" that
either supports or opposes the lower court's judgment (Wiener 2004).
The denition of the brief relates to the purposes it serves. As a docu-
ment drafted by advocates, the brief is primarily designed to persuade. The
litigant and party scripting the brief attempt to persuade the Justices that
they should rely on the brief's analysis and vote in the party's favor (Wiener
2004). When this purpose is taken into account, there is a clearer distinc-
tion between when a brief should be neutral and when it should exert bias
in favor of the writing party.
10
The subtlety involved in drafting briefs is that the judge should not feel
as though the brief neglects relevant aspects of the case, such as facts that
may hurt the drafting party's position, in an eort to win. Such overt
attempts at manipulating the case narrative may very well injure the party's
credibility and in turn dissuade the Justices from relying on that party's
characterization of the case (Marvell 1978). Briefs are unlikely to persuade
if the Justices do not see the presentation as credible and thus do not read
the brief closely or follow the logic of the party's argument (Aldisert 1996). A
main goal of the brief is to consolidate the party's case in a single instrument
and this instrument is not likely to convince the Justices of the merits of a
party's argument if it lacks credibility (Peck 1983).
1.3 The Empirical Study of Briefs
In a vacuum, briefs are only one of the many inputs upon which Justices
base their decisions. With many variables including briefs aecting decision
outcomes, analyzing the unique eect of briefs on judicial decisions poses
major challenges. Based on this, a common counterargument to the intuition
that the language from briefs should play a role in Supreme Court opinion
language is that the Justices have an abundance of other reliable material at
their disposal such as lower court opinions, amicus briefs, and the transcripts
from oral arguments, and so there is no reason for briefs to play a unique
role in the Justices' decisions (Johnson 2001, 333-34).
Statements from judges and the Justices help to separate out the primary
role of briefs. Justice Ginsburg (1998) said, \[a]s between brieng and ar-
gument, there is near-universal agreement among federal appellate judges
that the brief is more important-certainly it is more enduring." Even judges'
statements based on their personal understandings about whether briefs af-
fect decision outcomes may not be wholly accurate. We know this from an
11
area of work within law and psychology that looks at the process of moti-
vated cognition and presents that judges preferences for certain results may
\trigger the operation of cognitive processes that lead to the desired conclu-
sions" (Kunda 1990, 493). The empirical study of briefs is designed to aid
in clarifying the relationship between briefs and opinions beyond that which
is already conveyed by relevant individuals' statements.
As Black and Owens (2012, 97) suggest, merits briefs are uniquely important
in providing information to the Justices, dening issues, framing the Jus-
tices' views of cases, and in giving attorneys the uninterrupted chance to try
to persuade the Justices to decide cases in their favor. At the same time,
the persuasive power of briefs may be minimized by the Justices already
constructed preferences.
In addition, there is a rich legacy of work that suggests not only the out-
come of court cases, but also the decision rationale is consequential when
trying to discern the normative and societal implications of the Court's work
(Shapiro 1962; Monaghan 1979; Carter and Burke 1998; Silverstein 2009).
The empirical study of opinion language and the inputs to courts' linguistic
choices, however, is a relatively new eld (e.g. Wedeking (2010); Pang et al.
(2012)).
Much of the scholarship in political science looking for the inputs to Supreme
Court decision-making focuses on the in
uence and informational role of am-
icus curiae briefs (Segal 1988; Collins 2007) and the role of oral arguments
(Johnson et al. 2009; Johnson 2004; Johnson, Wahlbeck and Spriggs 2006).
Smaller N studies of the role of litigant's briefs in the Supreme Court are
recent additions. Pamela Corley's (2008) work on whether briefs \in
uence"
the content of Supreme Court opinions is one example. Corley uses the same
plagiarism software, WCopynd (Bloomeld 2014), that I use in this Disser-
tation to study the Justices' use of the same or similar wording in opinions
as in the parties' briefs for the 2002 through 2004 terms. Other studies of
12
the relationship between briefs and opinions use dierent means to measure
linguistic relationships such as cosine similarity (Oldfather, Bockhorst and
Dimmer 2012), Wordscore's method of locating texts in policy space, and
Na ve Bayes supervised text classication (Evans et al. 2007). The nding
common to all of these studies is that the language choices in briefs play a
role in the Justices' choice of language in opinions.
1.4 Relating Briefs to Opinions
To understand how briefs directly impact opinions take the example of
Steiner v. Mitchell.
13
In this case, the Court adopted many facts that were
set forth by the respondent but were not in the petitioner's brief. While
the majority of these facts were also mentioned in the lower court opinion,
the Court adopted the wording as it existed in the respondent's brief and
did not focus on material from the lower court opinion that supported the
petitioner's position.
For example, the opinion discusses the risk of lead poisoning that can result
from factory work in a battery plant. It states,
\[t]he risk is `very great' and even exists outside the plant be-
cause the lead dust and lead fumes which are prevalent in the
plant attach themselves to the skin, clothing and hair of the em-
ployees. Even the families of battery workers may be placed in
some danger if lead particles are brought home in the workers'
clothing or shoes."
14
By accepting this portrayal of the facts, the Court not only highlights the
necessity of the workers' safety procedures, but perhaps more importantly
13
350 U.S. 247 (1956)
14
350 U.S. at 250 (1956) and Brief for The Secretary of Labor, 1955 WL 72536 at 3.
13
extends the class of potentially harmed future plaintis to victims' families.
The only two sources of this language are the opinion and the respondent's
brief (the respondent's brief in-part reorganizes language from the lower
court's opinion). Because the respondent's brief predates the opinion, there
is little reason to doubt that the opinion's author Chief Justice Warren in-
corporated the language from the brief in the opinion. By the mere inclusion
of such language found in the respondent's brief, the Court provided a basis
for a potentially large class of compensation claims.
Choices of whether or not to include language like in the above examples
are typical decisions the Justices make, all of which may aect an array of
rights in previously unforeseen ways. Individuals with greater abilities to
mold Supreme Court opinion language thus have power to shape the law
and those aected by it.
The existence of signicant ties that bind merits briefs and courts' opinions
are known within the legal community. In fact, there is a general expectation
that briefs will play a signicant role in dening opinion language. Professor
Llewellyn (1962, 638) explained that a brief should provide, \[the] court with
something that it can lift verbatim, into the opinion taking care of all prior
authority, phrasing the whole satisfactorily, and applying it to the case in
hand."
Judges further inform us of the various links between briefs and opinions
and specically on the expectation that brief language will help to shape
opinion language. Judge Prettyman (1953) asserts that a well-designed brief
can be the main reference point for the judicial opinion. It may well provide
part or all of the rationale for a court's decision (Tate 1978). In his work
on appellate advocacy, Judge Aldisert (1996) provides First Circuit Judge
Michael Boudin's description that judges often value the litigants' extensive
involvement at the case level and the expertise they can share with the court.
In this way the merits brief provides judges with the materials on which
14
they can formulate their decisions. Since briefs are organized to contain the
entirety of a party's argument and to explain the party's position in the
case, they are holistic compared to the piecemeal accounts provided by oral
argument that may veer o from the main substance of the case. On this
point Judge Kravitz (2009, 261) notes, \No matter how well we judges may
take notes, it is dicult to absorb, let alone retain, numerous factual details
or other case-specic information..." Even with oral argument transcripts,
merits briefs provide the Justices with much more complete pictures of cases.
The role of briefs in shaping opinion language is also documented at the
qualitative level. Epstein and Kobylka (1992, 114) look at the death penalty
case of Gregg v. Georgia.
15
By comparing the documents in the case, the
authors contend that, \the Justices borrowed heavily from Powell's dissent
in Furman and from Bork's [the Solicitor General] brief, `[u]sing the solicitor
general's arguments they also indicated that evolving standards of decency
did not support an abolitionist outcome, nor did judicial restraint.'" In a
similar vein, Judge Posner (1993) highlights several instances when Judge
Cardozo (prior to his tenure on the Supreme Court), reworded statements
from merits briefs into tactful opinion language. Posner presents examples of
this from Cardozo's decisions in both the Ultramares,
16
and Macpherson,
17
cases.
The brief is the attorney's main opportunity to convince the judge of the
correctness of his or her client's position (Martineau 1986). This is not
to say that briefs necessarily win cases. Judge Sloviter of the 3rd Circuit
describes that even a persuasive and eloquent brief will not sway her decision
if the facts or law are not in a party's favor (in Aldisert (1996)). Eective
briefs may, however, convince the Justices and clerks to share more opinion
language with those briefs.
15
428 U.S. 153 (1976)
16
174 N.E. 441 (1932)
17
111 N.E. 1050 (1916)
15
The Justices obviously do not decide cases solely based on the merits briefs,
but these briefs can be and often are the main element in the decision mak-
ing process. Alongside the fact that they are often a reference point for
clerks and Justices in a case, they are present for the Justices' use from the
beginning until the end of the case. In the Supreme Court, the Justices and
their clerks often make their rst impressions of cases based on the merits
briefs (Rehnquist 1999; Hughey 2010). The briefs are used in preparation
for oral argument and help the Justices reach tentative conclusions before
the opinion is assigned to a justice (Tate 1978; Marvell 1978). While oral ar-
guments may paint an incomplete picture of a case, the Justices can always
review the briefs for a full assessment of the parties' positions. The Justices
also hear several oral arguments in sequence, which can lead them to forget
particular details of a case (Ginsburg 1998). Briefs, however, preserve the
arguments in their entirety. In this way, briefs are often reviewed by Justices
and clerks multiple times after oral argument; especially during the process
of drafting opinions (Dwyer, Feldman and McBridge 2007; Hughey 2010).
The Justices are overt in their statements that they tend to form under-
standings of cases and of their likely positions on the issues prior to oral
arguments and based on the briefs (Swain 2013). If the Justices make pre-
liminary decisions based on the briefs, then they may use oral arguments as
a means to test the validity of their intuitions (Epstein, Landes and Posner
2013). Since the Justices and clerks may initially learn about their cases
through the briefs, briefs can provide a wealth of information about the case
and create priors for these Justices that are dicult to undo (Cohen 2009).
Briefs also impact opinion language due to their proximity to the opinion
drafters. Clerks' preliminary bench memos are often drafted alongside the
merits briefs (Posner 2013a; Dwyer, Feldman and McBridge 2007). Briefs
are ever present after oral arguments. Some judges and clerks report they
review the briefs many times after arguments when coming to their decisions
(Dwyer, Feldman and McBridge 2007). Justice Thomas describes how oral
16
arguments might lead him to make a few new assessments of a case but,
\nothing that totally changes what [the justice and his clerks have] done,
based on the briefs [emphasis added]" (in Swain (2013, 90)). Justices also
often base their decision of whether to join the majority opinion or not
based on the briefs (Tate 1978). As dierent Justices' chambers may have
diering uses for merits briefs, in certain circumstances merits briefs may
lead to initial impressions in cases, while in others they may aid Justices and
clerks in forming concrete viewpoints and can in such cases play formative
roles in shaping opinion language.
Chapter 2
Merits Briefs, Amicus Briefs,
and Lower Court Opinions
2.1 Overview
During his 2005 Senate conrmation hearings, then court of appeals judge
John Roberts declared, \Judges and Justices are servants of the law, not
the other way around. Judges are like umpires. Umpires don't make the
rules; they apply them."
1
This legalistic formulation of judging which takes
a mechanistic approach implying little agency on the part of the judge is
met with heavy skepticism today. In one instance, Judge Posner (Posner
2013a, 6) from the Seventh Circuit Court of Appeals critiqued Chief Justice
Roberts' statement saying, \The umpireal conception of the judge breeds
passivity. The judge does not pick the players, the plays, etc. He just
watches...Of course no informed person, least of all Chief Justice Roberts, be-
lieves that an adequate description of a Justice of the U.S. Supreme Court."
1
The entire transcript of the hearing is available from http://www.senate.gov/
reference/Supreme_Court_Nomination_Hearings.htm, accessed 12/4/2015.
17
18
An honest assessment of Supreme Court decision-making is likely in between
these two poles of this debate. The Justices choose the cases they hear from
a litany of petitions for certiorari (Cameron, Segal and Songer 2000). This
provides them the immediate ability to take cases with particular mixes of
fact and law; cases which are more or less constrained by existing prece-
dent. They, like other political actors, have policy preferences which play
roles in their voting behavior (Segal and Spaeth 2002; Epstein and Knight
1998). Under either conception of judging, once the Justices decide to hear
a case they are loosely constrained by the documents led in the case and
by the lower court record (e.g. Dworkin (1978, 31)). In this vein, Justice
Alito (2009, 35) helps to clarify Justice Roberts' above-quoted statement:
\I think what the Chief Justice was getting at was simply an analogy and
not anything that he intended to be exact. Obviously, there are important
dierences between umpires and judges, but I think what he was getting at
is a very valid point, which is that umpires have rules to apply, and judges
have rules to apply."
Supplementing case-specic constraints, the Justices have additional incen-
tives to rely on the case material presented to them. If we assume they
are rational actors, then we can also presume they try to maximize their
utility functions (Epstein, Landes and Posner 2013, 48). This leads to an
expectation that they will act eciently and thus not create additional and
unnecessary work for themselves.
2
Of course this utility in eciency is coun-
terbalanced by an interest in the quality of the judicial output and concern
about an opinion's reception (Lax and Cameron 2007; Baum 2009a). Part
of the eciency of Supreme Court judging is achieved by the Justices lim-
iting themselves to the aggregate of the accumulated case material. With
this material, the Justices have a circumscribed universe of tools at their
2
Part of this eciency is created through the Justices varying uses of their clerks
(Peppers and Zorn 2008). When I refer to Justices I include the work of their clerks in
this unit as without case specic information it is impossible to separate out the case level
work performed by the Justices and their clerks.
19
disposal. These tools include the factual record and lower court opinions,
briefs, and other documents led in the case. While the Justices are not
bound by these resources, especially if the they regard them as insucient,
alternatives involve added and potentially super
uous work. In the majority
of cases, the Court has what Justice Holmes referred to as \the implements
of decision" at hand and the Justices become, in a sense, passive actors to
the extent that they do not have to or choose to seek additional material
(Davis 1940).
While sitting on a panel for the American Philosophical Society, Justice
Souter (2010, 34) aptly described the quintessential situation the Justices
face: \We have language in front of us and we don't know the limits of that
language. It is not clear to us. We take whatever resources we may have,
and we try to come up with a reasonable construction of that language-
a reasonable meaning of what may not be clear." This Chapter focuses
on the core resources that the Roberts Court Justices and their clerks use
when drafting Supreme Court opinions. It does so by measuring the Court's
reliance on wording in briefs and lower court opinions between the 2005 and
2014 Court Terms. To accomplish this goal, the Chapter compares language
in over 13,000 documents in the Court's docket during this period with
their respective opinions. The Chapter then looks at the relative impact of
parties' briefs and lings, amicus curiae briefs, and lower court opinions on
the Court's opinion language.
The analyses in this Chapter occur at both the macro and micro levels. They
look at the relative eects of these lings on the Court's overall opinion
language and then break this down by individual Justice. The Chapter
initially proceeds by looking at the Justices' goals for their opinions. Then
it examines the process of opinion construction and the tools available to
the Court. Next it goes through the methodology and the results for the
entire Court and by individual justice. The concluding section discusses
how an increased understanding of the Justices' toolkits provides us with an
20
enhanced understanding of Supreme Court decision-making, and by looking
at where additional data could further improve our understanding of opinion
construction.
2.2 Written Opinions
The written opinion is a hallmark of the appellate legal process in the United
States. In the Supreme Court it provides reasons for judgements and it
publicly transmits the letter of the law. In this sense it works as a constraint
on both the Court as the purveyor of the law and on lower courts as well as
the the public as consumers of the law.
Opinions are not arbitrary exercises in power as they are predicated on a
history of prior case law. Such precedent varies in its constraining abil-
ity. With each case, Supreme Court rulings have the potential to build
on the Court's accumulated knowledge over time (White 1995). On top of
this framework, there is an expectation of error correction and of learning
from past experience (Hellman 1982). Landmark decisions overruling prior
judgements provide evidence of such corrective ability. One example of this
is the Court's decision in Lawrence v. Texas,
3
where the Court overruled
Bowers v. Hardwick.
4
When referring to the Bowers decision, language in
Lawrence includes, \That statement [regarding a fundamental right to en-
gage in homosexual sodomy], we now conclude, discloses the Court's own
failure to appreciate the extent of the liberty at stake."
5
The Court in Brown
v. Board of Education,
6
describes assertions adduced in prior cases regarding
the ability to implement separate but equal schooling facilities in a similar
manner:
3
539 U.S. 558 (2003)
4
478 U.S. 186 (1986)
5
539 U.S. at 562
6
347 U.S. 483 (1954)
21
\[w]hatever may have been the extent of psychological knowledge
at the time of Plessy v. Ferguson, this nding is amply supported
by modern authority. Any language in Plessy v. Ferguson con-
trary to this nding is rejected. We conclude that in the eld
of public education the doctrine of `separate but equal' has no
place. Separate educational facilities are inherently unequal."
While the Court does not regularly overrule its own precedent, it has the
unique ability to do so.
Reasoning from cases in opinions also provides legitimacy to rulings (Shapiro
1981; Carter and Burke 2007). It provides litigants with a seemingly objec-
tive explanation for the Court's results as opinions are guided by existing
law. To do so opinions often appear as engaged in \dialogues" with past
decisions (Lebovits, Curtin and Solomon 2008). Consequently opinions also
provide guidance for future litigants looking to raise similar issues (White
1995, 1367). With this purpose in mind, the Court may draft opinions with
more or less certainty depending on the degree of
exibility that the Justices
wish to leave in a particular area of law (Corley and Wedeking 2014). In
any case, the the Justices have the capacity to dene the contours of their
opinions for future application.
Even with the established practice of engaging with past case law, the inher-
ent ambiguity in the law's language aords the Justices signicant discretion
in their judgements (e.g. Hart (1961); Posner (2013a)). When the Justices
have strong views on certain issues, they often vote according to those views
rather than with existing precedent on an issue (Segal and Spaeth 1996).
Within these permeable strictures of precedent, the Court has wide latitude
to redene the direction of the the law. This freedom is consequential to
forming opinions that accord with the views of a majority of the Justices.
Justice Ginsburg (1992, 1194) highlights the importance of interjustice con-
sensus in her statement, \In writing for the court, one must be sensitive to
22
the sensibilities and mindsets of one's colleagues, which may mean avoiding
certain arguments and authorities, even certain words." Here the importance
of linguistic
exibility is apparent, as it can be used to satisfy the views of
other Justices.
2.3 The Justices' Toolkits
This section examines how the Court achieves this balance between tting
opinions within the bounds of existing law and achieving the policy output
that a majority of the Justices seek. How do they do this? While judging
was once viewed as an art of discovery, it is now perceived as the forming
of inferences and the creation of internal coherence between such inferences
(Simon 1998, 20). Supreme Court cases seldom have clear answers and the
indeterminacy in the law creates situations ripe for the Justices' processes
of decision-making. The Justices begin cases with many resources at their
disposal, the most important of which are generally the parties' merits briefs
(Kravitz 2009).
The Justices, however, generally look at the briefs only after their clerks do
so. Judge Posner (2013b, 3) describes the amount of workload delegation
to Supreme Court clerks from his experience clerking for Justice Brennan,
\I was stunned to discover that Supreme Court Justices didn't write their
own opinions. Most judges don't much like to write and therefore to review
and edit law clerks' opinion drafts." Justice Ginsburg (2012, 396) further
notes the importance of clerks reading briefs prior to the Justices, \[Clerks'
bench memos] job is to give me a road map through the case, and then I
can read the briefs. They also tell me which of the green briefs (Amicus)
I can skip." In the Justice-clerk relationship the Justice maintains ultimate
responsibility for the written opinion even if the outcome is in
uenced by
the clerks. The work of clerks, however, is not severable from that of the
23
Justices without greater individual level details of the relative duties of each
Justice and clerk and an understanding of the weight each Justice gives to
clerks' work products.
The cert pool, where clerks from one Justice's chamber prepare case memos
for most or all of the Justices relating to which cases the Court should exam-
ine during a forthcoming term further confounds this relationship (Palmer
2001). This close-knit relationship between the Justices and clerks makes it
evident that at the merits level, clerks are essential in the process of deciding
which individual case lings ultimately are insinuated into the prose of the
nal written opinion.
Within individual cases many documents are submitted for the Justices'
review. Most important are often the parties' merits briefs. On this note
Justice Thomas (2011, 70) said, \In my view, most of the heavy lifting,
most of the work, is done in the written briefs, written arguments by the
parties." Due to the amount of time they spend on individual cases, the
lawyers generally have time and resources to gather what Justice Ginsburg
(1985, 208) referred to as, \...the universe of information we might consider."
Several studies provide empirical evidence of the close connection between
language in briefs and in Supreme Court opinions (Black and Owens 2012;
Corley 2008).
Oral arguments can sway the Justices' decisions, especially in close cases
(Johnson, Wahlbeck and Spriggs 2006). Although there is no clear link be-
tween oral arguments and decision language, Supreme Court opinions often
reference statements in oral arguments and the lawyers' prepared state-
ments often mirror sections from their briefs.
7
Oral arguments are also an
important opportunity for parties to clarify arguments made in their briefs
(Aldisert 1992; Ehrenberg 2003).
7
See e.g. Bush v. Gore, 531 U.S. 98, 108 (2000) (\At oral argument, respondents
estimated there are as many as 110,000 overvotes statewide").
24
Alongside the parties' arguments, non-parties amicus curiae briefs also play
roles in Supreme Court decision-making. These briefs are traditionally
thought to have the greatest impact on policy questions (Collins 2004). Am-
icus briefs often articulate positions shared by segments of the population
and are directed towards specic policy concerns that may be aected by
case outcomes. Large-N studies show that they may aect the outcome of
some Supreme Court cases (Songer and Sheehan 1993; Kearney and Merrill
2000). They also have the potential to sway the Court's choice of opinion
language (Collins, Corley and Hamner 2015).
The Supreme Court is typically the second appellate court to hear a case
after either a federal or state court of appeal. Many of the parties' arguments
and the entire factual record are generally developed by this point. The
language already developed by the lower courts helps to shape the language
in Supreme Court opinions (Corley, Collins and Calvin 2011). Based on
the Court's ability to select specic cases to review, the disposition in the
lower court may play a signicant role in the selection of cases as well. The
Court is more likely to grant certiorari and overturn a lower court's ruling,
for instance, if a majority of Justices disagree with a lower court decision
(Epstein and Knight 1998, 80). Thus, the lower courts that ruled on a case
and the dispositions therefrom often provide the Court with information
regarding which cases to select on certiorari and provide the Justices with
preconceptions about the direction they may vote if the Court grants cert
(Songer, Segal and Cameron 1994). When a case was previously heard by
a federal court of appeal, the Justices' views of the lower court judges that
authored the majority and/or dissenting opinions may further aect their
choice of cases to take on cert and create predispositions for their ultimate
rulings and for the language in opinions (Haire, Lindquist and Songer 2003).
Focusing specically on the language of opinions, the Court utilizes two types
of language from briefs and lower court opinions: (1) citation to authorities
and the record and (2) original language. Cited authority ranges in its degree
25
of in
uence and constraining ability. Some sources such as international
law tend to have little impact on opinion content (O'Brien 2006). Similarly,
references to non-legal material such as social science statistics may be relied
upon in opinions but generally only play a role in decisions with broad policy
implications (Margolis 1999; Hasko 2002).
Connections to existing law and to stare decisis tend to have the most preva-
lent eect on Supreme Court opinions (Knight and Epstein 1996). While
the Justices may have strategic uses for the insertion of specic precedent
into their opinions, case law is fundamental in framing analyses in opinions
(Baum 2009b, 74). The Court's choice of precedent is also in
uenced by the
group preferences of the majority and is often related to the views expressed
in separate opinions in the same cases (Lupu and Fowler 2013). Indetermi-
nacy in the law of the cases the Supreme Court hears allows for this element
of choice. Precedent plays a major role in framing and shaping opinions,
even when the Justices' strategic goals are stripped out of the equation.
Along with shared citations, opinions also often borrow original language
from briefs. This creates what Judge Cardozo referred to as a \style ag-
glutative" where sections of opinions strongly resemble briefs for winning
parties in cases (Lasky 1961). In several instances, the Court's language is
primarily an assemblage of language taken directly from a merits brief led
in the case (infra Chapter 4).
Such uses of language raise questions regarding the expected sources of
Supreme Court opinion language. Some nd this agglutative style norma-
tively troubling (Lebovits, Curtin and Solomon 2008). It creates situations,
for example, where the writing styles of particular Justices become unnotice-
able. Opinions' uses of original brief language may be a function of clerks'
work as well. In delegating portions of opinion construction to clerks, the
Justices are not necessarily entirely cognizant of the source of the language
the clerks use. Generally when Justices' excessive delegation of tasks to
26
clerks is made public or where Justices' work products are overtly criticized,
the Justices feel compelled to put more eort into and take greater ownership
over their opinion construction (Choi and Gulati 2004, 1086).
This Chapter has several hypotheses informed by the existing ndings and
theories described above. First, the intensity of the relationship between the
author of a document led in a case and the case itself should in
uence the
amount of impact the document has on an opinion's language. Based on this,
the Chapter's rst hypothesis is that across Justices, the Court will share
the greatest amount of language with parties' merits briefs, the next most
language with lower court opinions, and the least amount with amicus briefs.
There should be a relationship between the amount of information conveyed
to the Court in a case and the absolute amount of overlapping words that
are inserted into an opinion. Based on this the second hypothesis is: There
should be an inverse relationship between the amount of language the Court
shares with parties briefs and amicus briefs depending on the number of these
briefs led in a case. As more parties' briefs are led the Court should share
more language with parties briefs and less with amicus briefs and vice-versa
for amicus briefs. The third hypothesis is similar but related to Solicitor
General lings: When the Solicitor General les a merits brief in a case
the Court should share more wording with the parties' lings and less with
amicus lings and vice-versa for amicus lings when the Solicitor General
les an amicus brief.
The next hypothesis has to do with complex cases: Since the Justices likely
have the less dened preferences and preexisting views as case complexity
increases, the Court should rely more heavily on parties' explanations and
on lower court ndings in these cases. The nal hypothesis deals with case
salience: Amicus briefs should play a larger role in dening opinion lan-
guage as the salience of cases increase and consequently so do their policy
implications. The converse should be true for parties' briefs in salient cases.
27
2.4 Data and Methods
This Chapter examines the similarities in language between, on the one hand
- parties' merits briefs, amicus briefs, and lower court opinions - and on the
other hand, Supreme Court written opinions in nearly 700 briefed and orally
argued cases between the 2005 and 2014 Supreme Court Terms. The anal-
ysis involves the comparison of over 13,000 documents with the Supreme
Court opinions. As with several other papers examining shared language in
political/legal documents, this Chapter uses WCopynd (Bloomeld 2014)
to perform comparisons (Grimmer 2010; Collins, Corley and Hamner 2015).
WCopynd was created as a tool to locate pairwise plagiarism and has a
variety of local settings. Like previous studies, I set the minimum number
of words in an overlapping phrase to six and the minimum percentage of
overlapping language to 80% (infra Chapter 3). The intention behind these
settings is for the program to only locate phrases with suciently mean-
ingful overlapping language. This Chapter looks at the number of words in
Supreme Court opinions that overlap with language from briefs and lower
court opinions.
Before delving into the disaggregated amount of language opinions share
with these types of Court documents, I look at the amount of language that
the Court shares with all lings in each case across the ten terms I examine
in this Chapter. To get a sense for the number of total briefs led each term,
Figure 2.1 shows the annual number of combined party and amici lings.
Figure 2.1 shows that the absolute number of lings has been relatively
stable since the 2005 Term although there is some
uctuation with over
1,500 lings in 2012 and under 1,000 lings in 2011.
The stability in the overall number of lings leads to the presumption that
the amount of overall information conveyed to the Justices in the cases they
heard during these terms was relatively consistent. Another way of looking
28
Figure 2.1: Total Documents Filed Per Term
0 500 1,000 1,500 2,000
Number of Total Briefs Filed
2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
Term
at the number of lings, however, presents the possibility of greater variation
between cases. Figure 2.2 shows the average number of lings in each case
by issue area.
8
Here we see large variation between the values in some of the issue areas.
Two areas where there are clearly the most lings per case are for those
dealing with Due Process and First Amendment issues while cases dealing
with issues related to Attorneys and Criminal Procedure average close to
twenty fewer lings per case.
How consistent is the Court in its uses of these lings? Figure 2.3 is a his-
togram that shows the per-case distribution of the percentage each opinion
overlaps with all of the party and amici lings in the case.
The distribution in Figure 2.3 is similar to a bell curve. While the bulk of
the distribution is around 20% per opinion, the overlap values range from
8
I use the separation into issue areas as coded in the Supreme Court Database (Spaeth
2015).
29
Figure 2.2: Documents Filed Per Case by Issue Area
0 10 20 30
Mean Total Filings Per Case
Due Process
First Amendment
Privacy
Economic Activity
Federalism
Civil Rights
Unions
Judicial Power
Federal Taxation
Criminal Procedure
Attorneys
Issue Area
Figure 2.3: Distribution of Overlap Value of All Filings Per Case
0 5 10 15
Percent of Briefs (Across All Terms)
0 10 20 30 40
Overlap Value
around 5% to over 40%. Evidence of large variation between the amount
of overlapping language in opinions alone, however, does not point to the
source of this variation. There are clear dierences in the average amount
of opinion language that overlaps with language from the briefs in a case on
30
a yearly basis. Figure 2.4 graphs the percentage overlap value per term.
Figure 2.4: Mean Overlap Value Per Term for All Filings
16 18 20 22 24
Mean Per Case Overlap Percentage
2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
Term
This mean value
uctuates quite a bit as the it dips below 18% during
several terms and exceeds 22% at its high point in 2012. Although the
trend is jagged over time, it clearly increases from 2005 to 2014.
Figure 2.5 shows the relative impact of parties' briefs, amicus briefs, and
lower court opinions on the language in the Roberts' Courts opinions.
As predicted, the Court shares the largest portion of its opinions with lan-
guage from parties' briefs, then lower court opinions, and the smallest por-
tion with amicus briefs. Not only are parties and the lower courts generally
more directly involved with the cases, but they also provide descriptions of
the facts along with legal points and arguments that the Court may incor-
porate into its opinions.
While this provides an initial sense of the importance of the lings for the
Court's composition of opinion language it leaves important questions unan-
swered. The subsequent analyses in this Chapter look beyond the fraction
31
Figure 2.5: Box Plots of Overlap for Parties' Briefs, Amicus Briefs, and
Lower Court Opinions
of each opinion that shares language with the aggregate of lings and in-
stead look primarily at the absolute number of overlapping words (based on
the same WCopynd settings) between opinions and the various documents
submitted to Court in a case.
Along with the number of lings in each case, the models in the Chapter
also control for the presence of the Solicitor General, the complexity and
salience of a case and the ideological preferences of each justice.
2.4.1 Dependent Variables
To understand the Justices' use of opinion construction resources, the multi-
variate analysis is organized into separate, related models examining the use
of language from each source - merits briefs, amicus briefs, and lower court
decision - in the Court's majority opinions. As the focus of this Chapter
is opinion construction, it marks a departure from previous works looking
at language overlap. In other studies using similar methods, the dependent
32
variables were constructed from the overlapping language between individ-
ual briefs and opinions leading sometimes to dozens of observations for each
case where many briefs were led (Corley, Collins and Hamner 2013). This
Chapter is concerned with the cumulative impact of the aggregate of - the
parties' merits briefs, amicus briefs, or lower court opinions in a case- on
the Supreme Court's majority opinion language. To measure this, there are
three comparisons per case - one with all language in all of the parties' mer-
its briefs against the Court's opinion language, one with all the language in
the combined amicus briefs against the Court's opinion language, and lastly
one with any lower court decisions in the case against the Court's opinion
language.
These three comparisons on a case-by-case basis provide the number of over-
lapping words between the resources of interest and the opinion. The main
measurement is the number of overlapping words rather than the percentage
of each opinion that overlaps with language from a given resource as other
studies have used (Corley 2008). The percentage of the opinion that overlaps
with a resource is not as benecial for the purposes of this Chapter because
this measure does not account for longer or shorter opinions that rely on
more or less of a resource. With longer opinions, even when there are many
instances of overlapping language, the percentage of overlap still tends to
remain low because this measure is relative to the length of the opinion. The
overlapping words metric (with the addition of a control for opinion length)
provides a means to account for the raw amount of overlapping language
the opinion shares with a given resource of opinion construction.
2.4.2 Independent Variables
The Justices' choices in the quantity and type of language they use from
each resource is mediated by additional factors in each case. To control
for the impact of these factors several additional variables were integrated
33
into each model. The rst as mentioned above is the length of the opinion.
Opinion Length is a count variable based on the total number of words in a
case's majority opinion. The inclusion of this variable prevents opinions with
more words from skewing the results based on the larger absolute number
of words shared with those opinions.
As more lings in a case give the Court additional raw material to work with
in fashioning an opinion, the next set of variables, Party Filings and Amicus
Filings, control for the number of lings in each case. The expectation
behind these variables is that as party lings increase the Court should
share more language with wording from the parties' briefs. This should lead
to a positive relationship in the party brief model. The same should be true
for amicus lings in the amicus overlap model.
Along with the number of lings in each case, there are several controls
for factors that vary at the case-level. The rst, Complexity, measures the
number of issues in each case. This variable is derived from the Supreme
Court Database's counts of legal provisions and additional issues in each
case (Benesh and Reddick 2002).
The next case-level variable is Political Salience. The intuition behind this
variable is that the Justices will have more pronounced preferences, will
share less language with parties' briefs, and will share more language with
amicus briefs in politically salient cases (Bailey, Kamoie and Maltzman
2005). To measure the political salience of cases, this Chapter uses a dummy
variable for each case that is mentioned on the front page of the New York
Times the day after the opinion was released (Epstein and Segal 2000a).
The third case level variable is Legal Salience. To measure the legal salience
of the case, this Chapter uses a dummy variable that is coded 1 in every case
where the Court rules a law unconstitutional or overrules its own precedent
(Maltzman, Spriggs and Wahlbeck 2000). This Chapter uses the coding
34
in the Supreme Court Database (2016) to identify when cases have either
characteristic of a legally salient case.
The nal case-level variable, Civil Liberties, identies cases that t into this
issue area (those coded as criminal procedure, civil rights, rst amendment,
due process, privacy and attorney in the Supreme Court Database). The
Justices tend to have more clearly dened preferences in civil liberties cases
than they do in cases dealing with other issues (Epstein, Landes and Posner
2010). Recent Supreme Court rulings in marriage equality cases, with the
Court often fracturing along ideological lines, present prime examples of
this relationship. These include Obergefell v. Hodges,
9
, United States v.
Windsor,
10
and Hollingsworth v. Perry.
11
The Court tends to share more language in its majority opinions with briefs
led by the Oce of the Solicitor General than with briefs from similarly
situated litigators (Black and Owens 2012). To control for this inherent
advantage this Chapter creates variables that identify when the Solicitor
General is involved at the merits level of a case or supports a party with an
amicus brief (Solicitor General Merits and Solicitor General Amicus).
To control for whether the Justices' ideological preferences aect whether
they share more or less language with resources of opinion construction in
a given case, this Chapter uses an Ideological Compatibility variable. This
variable was generated using the Justices' Martin and Quinn Scores (2002),
which place the Justices on a conservative-liberal continuum. It then uses
the case level coding from the Supreme Court Database for whether the
ideological direction of the Court's ruling was liberal or conservative. This
dummy variable is coded 1 in instances where the majority authoring Jus-
tice's ideology is consistent with the ideological direction of the Court's
decision.
9
135 S. Ct. 2584 (2015)
10
133 S.Ct. 2675 (2013)
11
133 S.Ct. 2652 (2013)
35
2.5 Results
2.5.1 Macro-Level
Table 2.1 presents three OLS regression models for overlapping words in an
opinion.
12
Table 2.1: Multivariate Analysis
(1) (2) (3)
Lower Court Words Party Briefs' Words Amicus Briefs' Words
Total Opinion Words 0.0481*** (0.00723) 0.0999*** (0.00796) 0.0424*** (0.00549)
Party Filings -4.929 (5.104) 1.007 (4.831) -10.36** (4.023)
Amicus Filings 3.513** (0.850) 0.138 (1.159) 6.805*** (0.872)
Complexity 69.43*** (13.20) 22.49 (28.24) 11.72 (17.79)
Political Salience -28.63 (32.60) -56.66 (53.09) 3.305 (29.05)
Legal Salience 10.80 (49.70) 31.80 (41.47) 90.71* (49.91)
Civil Liberties -14.93 (17.36) -61.75** (18.52) -0.124 (16.05)
Solicitor General Merits -22.66 (33.76) 79.21** (25.93) -42.71** (16.08)
Solicitor General Amicus -40.86 (25.10) -35.82 (24.01) 52.24** (15.84)
Ideological Compatibility 13.83 (37.37) 16.06 (22.09) 13.66 (11.16)
Constant 82.58 (58.28) 237.4** (68.63) -48.51 (29.75)
N 654 668 668
Robust standard errors in parentheses clustered on majority opinion author
* p<.1, ** p<.05, *** p<.001
As the number of cases with written and accessible lower court opinions was
slightly smaller than the total number of written opinions, there are fourteen
fewer observations for the model based on lower court opinion words.
The rst control is for the overall number of words in each opinion. The use
of this control prevents opinions with more words from skewing the results
based on the larger absolute number of words shared with these opinions.
Focusing on the hypotheses, the number of amicus and party lings are
highly signicant (and move in the predicted directions) in the amicus model.
Here an increase in party lings decreases the number of words from amicus
12
For robustness purposes I also ran Seemingly-Unrelated Regressions which correct
for correlated error terms. The results were substantially similar to those in Table 2.1.
36
briefs that overlap with the opinions and an increase in the number of amicus
lings leads to more words shared between opinions and amicus briefs.
Consistent with the complexity hypothesis, the number of words that over-
lap between Supreme and lower court opinions increases with increased case
complexity. The assumption continued that the number of words overlap-
ping with party lings would increase as well with increasing case complex-
ity. Although the coecient for the complexity in the party lings model is
positive, it is not signicant.
Looking at case salience, opinions share more language with amicus briefs
in legally salient cases where the Court either overrules its own precedent
or declares a federal statute unconstitutional. The coecient for parties'
briefs is negative in these instances but not signicant. Overlapping words
between opinions and parties briefs' is negative and signicant in civil lib-
erties cases. This is consistent with the case salience hypothesis as these
cases are oftentimes categorized as the most salient type as often dealing
with individual rights (Epstein and Segal 2000a).
Finally, the Solicitor General variables have the predicted eects. When
the Solicitor General is an attorney on the merits, the amount of words
the Court's opinions share with the parties' briefs increases signicantly.
When the Solicitor General les an amicus brief there is a bidirectional
eect where the Court not only shares more language with amicus briefs,
but it also shares less with parties' briefs.
2.5.2 Justice-Specic Findings
The Justices and their clerks each have distinct preferences and practices
regarding when and how to use the language provided to them in each case.
This leads to diering amounts of shared language in each category (parties'
briefs, amicus briefs, and lower court opinions) by justice.
37
Before the opinion construction process can begin, however, an initial deci-
sion is made that aects the Court's end-product or the majority's written
decision: the assignment of the opinion's authorship. The opinion is either
assigned by the Chief Justice or the Senior Associate Justice in the ma-
jority if the Chief Justice is in the minority. The relationship between the
assigning and authoring Justices as well as the Justices topical specializa-
tions leads to patterns of authorship among the Justices (Lax and Cameron
2007). This process along with the Justices and clerks' unique writing styles
lead to categorical dierences in the styles of opinion writing. One way to
observe the dierences in opinions between the Justices is by their average
opinion lengths as presented in Figure 2.6.
Figure 2.6: Mean Opinion Length by Justice
0 2,000 4,000 6,000 8,000
Mean Words in an Opinion
Kennedy
Roberts
Alito
Souter
Kagan
Breyer
Sotomayor
Stevens
Ginsburg
Scalia
Thomas
The variables most strongly associated with an opinions' length are political
and legal salience.
13
Not surprisingly then, Justices Kennedy and Roberts,
the Justices that authored the most politically and legally salient cases dur-
ing this time period, also wrote the longest opinions. There is variation
13
When the independent variables in the model are regressed on total opinion length
the variables with the greatest magnitudes are political and then legal salience. They are
both also highly signicant.
38
from this norm also evident as Justices Souter and Kagan average in the
top ve of the Justices for opinion length even though both only authored
one legally salient opinion (Justices Souter and Kagan also authored two
and one politically salient opinions respectively during this period).
With a baseline of the Justices' relative opinion lengths we can examine the
Justices uses for linguistic resources in the form of briefs and lower court
opinions.
Figure 2.7: Mean Words Overlap by Justice
0 500 1,000 1,500
Mean Overlap Words of All Filings Per Case
Kennedy
Roberts
Sotomayor
Alito
Ginsburg
Stevens
Thomas
Souter
Breyer
Scalia
Kagan
This breakdown by Justice presents information about their tendencies to
directly utilize material from all lings in a case.
The number of words on average per opinion that the Justices share with
these linguistic resources is strongly associated with the Justices' average
opinion lengths although there are also visible stylistic dierences between
the Justices. Like the order for opinion length, the Justices that share the
most language with these resources are Justices Kennedy and Roberts with
Justice Alito sharing the fourth most words on average per opinion. Justice
39
Figure 2.8: Mean Party Brief Words Overlap by Justice
0 200 400 600 800 1,000
Mean Overlap Words of Parties’ Briefs
Sotomayor
Kennedy
Roberts
Alito
Ginsburg
Thomas
Stevens
Souter
Breyer
Kagan
Scalia
Figure 2.9: Mean Amicus Brief Words Overlap by Justice
0 100 200 300
Mean Overlap Words of Amicus Briefs
Kennedy
Roberts
Alito
Stevens
Ginsburg
Sotomayor
Thomas
Scalia
Souter
Breyer
Kagan
Sotomayor, however, shares the third greatest number of words per opin-
ion with these resources and Justice Ginsburg the fth. Justice Sotomayor
shares the most words on average per opinion of all the Justices with parties'
briefs. Justice Ginsburg is in the middle for the average amount of language
40
Figure 2.10: Mean Lower Court Words Overlap by Justice
0 200 400 600
Mean Overlap Words of Lower Court Opinions
Kennedy
Roberts
Sotomayor
Alito
Stevens
Thomas
Ginsburg
Scalia
Souter
Breyer
Kagan
she shares with parties' briefs, amicus briefs, and with lower court opinions
combined.
The reasons for the consistencies and dierences across Justices are also at-
tributable to the Justices and clerks' individual choices in drafting opinions,
which are not easily measurable and can ultimately lead to much of the
dierential language sharing.
Case examples help to contextualize the type of utility these resources pro-
vide for the Justices. The opinion in the trademark case of Hana Financial
v. Hana Bank,
14
was authored by Justice Sotomayor and is her opinion
which shares the largest percentage of language with the combination of
briefs in the case of the cases she authored. The opinion is 2,051 words
and of those, 753 or 36% overlap with the briefs. Upon examination of the
overlapping language in the case, it is apparent that this language primarily
relates to the lower court's conclusions and to the relevant Supreme Court
14
135 S.Ct. 907 (2015)
41
precedent.
15
Much of the rest of the overlap is based on references to rel-
evant cases such as the statement, \two marks may be tacked when the
original and revised marks are `legal equivalents.'"
16
Consequently there is
little evidence of salient overlapping original language in this case. Original
language that overlaps with the briefs in this case primarily relates to the
Petitioner's arguments, including that, \tacking is a question of law that
should be resolved by a judge,"
17
which the Court ultimately dismisses.
By contrast the overlapping language is much more meaningful in Justice
Kennedy's opinion in Boumediene v. Bush,
18
which, of his opinions, shares
the greatest number of overall words with the briefs. This opinion is 22,231
words in length and while 17% of the opinion overlaps with the briefs, this
equates to 3,845 total words. The case concerned the rights of detainees
at Guantanamo Bay to seek habeas-corpus relief. Although the shared lan-
guage predominately comes from a source cited to by briefs, the sources
are suciently unique to infer that the likely link between the opinion and
the original source was a brief in the case. These include Alexander Hamil-
ton's recitation in Federalist No. 84 taken from Blackstone's Commen-
taries: \connement of the person, by secretly hurrying him to jail, where
his suerings are unknown or forgotten, is a less public, a less striking, and
therefore a more dangerous engine of arbitrary government." The language
here clearly supports the conclusion ultimately reached by the Court that
the due-process protections aorded to many Guantanamo detainees were
insucient.
15
For instance the opinion shares language with the briefs repeating jury instruction, \A
party may claim priority in a mark based on the rst use date of a similar but technically
distinct mark where the previously used mark is the legal equivalent of the mark in question
or indistinguishable therefrom such that consumers consider both as the same mark..."
(135 S. Ct. at 910).
16
Id.
17
135 S. Ct. at 911
18
128 S. Ct. 2229 (2008)
42
In another instance in Boumediene,
19
the Court examines the extent that
these due-process protections are provided to non-American detainees. To
support the proposition that these provisions should extend to all detainees
held in Guantanamo, the Court (in language overlapping with the briefs)
cites to a 1984 British case, Khera v. Secretary of State which derives its
language from a principle described in Somersett's Case decided in 1772:
\Habeas corpus protection is often expressed as limited to `British subjects.'
Is it really limited to British nationals? Suce it to say that the case law has
given an emphatic `no' to the question." Like the quote before it, this passage
was used to strengthen the claim that additional protections were necessary
for all Guantanamo detainees. Such instances of shared language show how
briefs may provide the Court with the tools that ultimately support and
dene the conclusions the Justices reach as a matter of law.
2.6 The Role of Merits Briefs
This Chapter provides one of the rst empirical comparisons of the Court's
relative language sharing habits with its three main linguistic resources -
parties' briefs, amicus briefs, and lower court opinions - as well as with the
combination of all three. It also is the rst to break these sharing practices
down by individual justice and to look at how the Justices dierentially
utilize the aggregate of these resources.
The roles that already constructed linguistic resources play in dening the
Roberts Court's opinion language is more easily explained on the macro
level than on the individual Justice level. The results from the macro level
models conrm the Chapter's hypotheses about the factors that aect the
number of words the Court shares with the linguistic resources.
19
128 S. Ct. at 2248
43
The ndings from the macro level models may also help structure our un-
derstanding of individual justice behavior. The role of case salience, for
instance, in the average number of overlapping words by justice per opinion
assists in constructing a hierarchy of the Justices in terms of overlapping
words per opinion. The factors from the macro level model, however, are
insucient to fully explain the Justices' behavior.
There are two factors that if empirically documented could provide addi-
tional insight into the individual Justices' practices. One is a greater under-
standing of individual clerks' writing practices. Inevitably clerks make many
of the initial decisions regarding which linguistic resources to use in opin-
ions. A clearer understanding of the clerks' individual choices and behavior
could more precisely explain the types of information within the Justices'
purviews when they engage in the construction of opinions.
A second related factor is the type and extent of oversight provided by
each Justice. There are already qualitative examinations of the Justices'
oversight practices (Weiden and Ward 2006; Peppers and Ward 2012). If
combined with an understanding of clerks' behavior and choices, this in-
formation could lead to a more precise understanding of the paths that the
linguistic resources take on their way to their eventual placement in Supreme
Court opinions. While the Justice-level ndings in this Chapter could be
enhanced with this knowledge, the ndings in this Chapter could also help
validate the accuracy of our understanding of the role of clerks in opinion
construction. This is the next layer that when uncovered will provide a more
expansive and detailed understanding of opinion construction that not only
accounts for the principals in the form of the Justices but also their agents
in the form of their clerks.
Chapter 3
How Briefs and Opinions
Relate Across Time
Not all briefs play an equal role as aids in opinion construction. What are
the reasons for the disparities? One exercise espoused by Judge Posner
(1998) among others is for the litigant to put him or herself in the judge's
shoes (see also Stern (1981)). For instance, if the advocate were the judge,
what information would be most compelling and how should it be presented?
There are several features of a brief that provide the Justices with strong
contextual cues that a brief is likely to be strongly informative and helpful
in drafting the opinion.
This Chapter focuses on four sets of factors in the brief-opinion relation-
ship that may lead to more and less overlapping language between merits
briefs and the Court's opinions (language overlap). These are brief-related,
Justice-related, attorney-related, and case-related. The existence of signi-
cant variation in the amount of language that is shared between briefs and
the opinions across cases shows that there are instances where merits brief
44
45
are more and less of a utility for Justices and clerks in the opinion drafting
process.
3.1 What's Inside: Focusing on the Briefs
3.1.1 The Importance of Brief Quality
\I'll read good briefs, and I'll understand, that's a good brief.
And you'll try to think back at exactly what it was that made
it a good brief, and the sentence structure, and how it
owed
together. And to a certain extent, your mind internalizes that."
| Chief Justice John G. Roberts
Beginning day-one in law school, aspiring lawyers are taught that success
in their future careers is dependent upon becoming good legal writers. This
maxim often becomes a reality as attorneys, especially at the appellate level,
spend the bulk of their time preparing written documents such as briefs and
motions. As mastery of the written word is an essential attribute of a
talented attorney, we might expect little variation in the quality of written
briefs at the Supreme Court level - the venue of practice for the most talented
and experienced legal practitioners. The question of whether the quality of
legal briefs aects the Justices' decisions, however, is unexplored scholarly
terrain. A higher quality brief may, for instance, persuade the Justices
to adopt a greater amount of language in the opinion that accords with
language from the brief. In this Chapter I examine nearly 9,500 merits briefs
from 1946 through 2013 to help answer the question of whether the quality
of legal writing in merits briefs aects Supreme Court opinion content.
One reason why the quality of writing may aect Supreme Court opinion
content is that it can aect the Justices' and clerks' perceptions of the
46
strength of a lawyer's case. Justice Roberts acknowledges perceptible gra-
dations in the quality of Supreme Court briefs: \The quality of briefs varies
greatly. We get some excellent briefs; we get a lot of very, very good briefs.
And there are some where the rst thing you can tell in many of them is
that the lawyer really hasn't spent a lot of time on it...[emphasis added]"
(Roberts in Garner (2010b, 6)). Attorneys that wish to give the parties they
represent the best chances of favorable outcomes are behooved to take the
quality of the brief into account.
Justice Alito echoes Justice Robert's exhortations about the consequential
nature of briefs' quality, \Certainly, I appreciate good writing. It makes my
job so much easier. I've seen briefs that are extremely well written and some
that are abysmally written" (Alito in Garner (2010b, 170)). Justice Scalia
makes this point as well, \When you write well, you capture the attention of
your audience much better than when you write poorly" (Scalia in Garner
(2010b, 53)). Since the Justices take note of the dierences in brief quality,
lawyers are not performing their best as advocates without paying proper
attention to the quality of their written work.
A main reason that the question of whether brief quality matters is unan-
swered may have to do with the ambiguity surrounding how we conceive of
the concept of \quality." Across disciplines, writing quality is understood as
ranging from spelling and word choice to punctuation and ease of readability.
A second reason may have to do with diering views within the legal commu-
nity about what constitutes a high quality brief. While judges continuously
emphasize the importance of specic aspects of briefs such as clarity and
organization, surveys of lawyers show that they do not accord the same sig-
nicance to the written argument (Posner and Yoon 2010, 340). This may
explain some of the disjuncture about the importance of brief quality within
the legal community. Notwithstanding the lack of a clear conceptualization
47
of \quality," judges do not dispute that the quality of briefs aects their
perceptions of attorneys' cases.
How do we measure a concept that has so many potential dening features;
a concept that is dicult to measure with any objectivity due to often
subjectively dened features such as style (Moss 2013, 59)? This Chapter
tackles this question by examining multiple dimensions of briefs to assess a
spectrum of features related to the concept of brief quality. Some of these
aspects are particular to legal writing while others are central to writing
quality generally.
3.1.2 Understanding Quality
How do we know that the quality of merits briefs makes a dierence? For one
thing Supreme Court opinions tell us so. For example, in the case Zablocki v.
Redhail,
1
Justice Thurgood Marshall writing for the majority states, \With
regard to safeguarding the welfare of the out-of-custody children, appellant's
brief does not make clear the connection between the State's interest and the
statute's requirements." This case concerns a Wisconsin statute that places
specic requirements on parents with previous child support obligations that
wish to marry. Here, the Court does not give credence to the state's position
because the state does not present a sucient nexus between points in its
brief.
Lack of clarity in a brief can also confuse the Justices regarding a party's ar-
gument. In Atwater v. Lago Vista,
2
Justice Souter writing for the majority
states, \...it is unclear from Atwater's briefs whether the rule she proposes
would bar custodial arrests for ne-only oenses even when made pursuant
to a warrant..." In Atwater, the Court did not fully comprehend the peti-
tioner's argument from the brief, which may have ultimately aected the
1
434 U.S. 374, 389 (1978)
2
532 U.S. 318, fn. 15 (2001)
48
Court's response to the petitioner's pleas and led the Court to question the
strength of the petitioner's case.
Finally both merits briefs in a case may lack the quality expected by the
Justices. Without briefs that properly inform the Justices of the case merits,
the Justices may not feel properly able to adjudicate claims, especially with
far reaching implications. Justice Blackmun's dissent in New York Times v.
United States,
3
makes such a point. In his dissent Justice Blackmun writes,
\I therefore would remand these cases to be developed expedi-
tiously, of course, but on a schedule permitting the orderly pre-
sentation of evidence from both sides, with the use of discovery,
if necessary, as authorized by the rules, and with the preparation
of briefs, oral argument, and court opinions of a quality better
than has been seen to this point."
4
Justice Blackmun sought higher quality brieng in this instance to make a
more informed decision in a case with vast First Amendment implications.
In these examples it is apparent that the Justices are concerned with the
quality of the merits briefs. The Justices and their clerks also seek merits
briefs that clearly lay out all of the party's points and arguments and these
examples present evidence that the Justices may not rule favorably on a
party's contentions without such clarity.
5
But what exactly does quality
mean in such cases? To gain leverage on the concept of brief quality, this
Chapter examines statements from legal scholars and from judges. Based
on these statements, it develops testable hypotheses that examine whether
3
403 U.S. 713 (1971)
4
Id. at 761-62
5
Although infrequent, there are occasionally opportunities for the parties to clarify
statements from their briefs in oral argument. In Illinois v. Caballes, 543 US 405 (2004)
for example, Justice Ginsburg gives respondent's attorney a chance to clarify a statement
from the brief when begins a question, \There was something you said in your brief that
I thought was unclear. So may I ask you..." 2004 WL 2663949 at *37.
49
the features of brief quality that judges discuss are actually associated with
more successful briefs.
Attorney experience is often anecdotally associated with brief quality. It is
possible for instance that attorneys make marginal improvements in their
brief writing with each successive brief they write for the Supreme Court.
Justice Thomas relayed this notion in his statement, \I think you learn
over time. You gain kind of a comfort with it. It's like a jazz musician or
something. You get a feel for it. You don't just know the law, but you have
a feel for it. You have a feel for what the judges are trying to do. And
then you know where you can give a little ground without giving up your
case" (Thomas in Garner (2010b, 109). Similarly Judge Robert Baldock
from the Tenth Circuit Court of Appeals underscored the importance of
working through multiple drafts of each brief stating, \Now those who have
an appellate practice, who appear before us a lot, their briefs are usually
very well done. But with the lawyer who only comes once in a great while,
often the product is not good and it's no help. That really irritates me when
I haven't been helped at all by a brief and it has wasted my time" (Baldock,
Lucero and Mandell-King 2001, 274).
Another feature that experience may help develop is an attorney's credibil-
ity. Experienced Supreme Court litigators tend to be highly successful on
the merits (Lazarus 2008). The credibility of OSG (Oce of the Solicitor
General) attorneys, for example, is often cited as a reason for the Justices'
trust in them (Wiener 2004). This credibility may generate a presumption
of high quality briefs that in turn benets the OSG's likelihood of success.
As this is a presumption of credibility often based on an attorney's experi-
ence, it may be rebutted by evidence that trust is unwarranted. Attorneys
that present inaccurate information in their briefs or that attempt to de-
ceive judges with skewed portrayals of the facts may lose this credibility
and develop notoriety for their lack of candor (Stark 1983; Marvell 1978).
50
Accordingly, for Justice Stevens, honesty is the most important quality of
a brief (Stevens in (Garner 2010b)). Judges on federal benches at all levels
consistently highlight the necessity of honesty in briefs and how deception
can lead to the loss of the judge's trust, which may consequently injure the
attorney's reputation (Reinhart in Aldisert (1996, 86)).
Attorneys may also lose credibility through attacks on other parties or en-
tities involved in the case (Edwards 1992, 63). Since the attorney's goal is
to persuade, even if the facts are not in the party's favor, attorneys need
to tread lightly when trying to portray facts in the most favorable light to
their clients while providing accurate statements that do not belittle other
actors involved in the litigation (Phillips 1998).
One feature of a brief that judges repeatedly say can win or lose a case is
its clarity. When lawyers do not lay out all of their points clearly, judges
may miss important aspects of the party's position. On this note, in an
instance when Chief Justice Roberts was asked if a bad brief can lose a case,
he replied, \It sure can - because [the Justices] may not see your strong case.
It's not like judges know what the answer is. I mean, we've got to nd it out.
And so when you say can bad writing lose a strong case, if it's bad writing,
we may not see that you've got a strong case" (Roberts in Garner (2010b,
23)). Answering the same question Justice Alito conveyed, \It can because
you may totally fail to convey the point that you want to make to the court.
The court just might miss your point. There have been times when I've read
a brief, and reread a brief, and I just didn't see what it was saying" (Alito in
Garner (2010b, 177)). The relative clarity of the party's brief may impact
the judge's view of the case. Judge Diane Wood from the Seventh Circuit of
Appeals explained, \...if one side has presented a beautifully organized and
written brief, and the other leaves me trying to decide if they've hit the side
of the barn or not...there's an inherent advantage to the side that's done it
well" (Wood in Garner (2013, 103)).
51
The necessity of brief clarity is also one of few factors that has remained
of central importance in judges' analyses of advocacy over the last hun-
dred years (Anderson 2010, 4) and judges describe that the best briefs they
read are often the clearest (Robbins 2002). Along these lines, when writ-
ing regarding judges' expectations for briefs, Judge Roger Miner (1992, 23)
unsurprisingly described, \We expect clarity, well-organized arguments, and
understandable sentence structure. All too often, we nd rambling narra-
tives, repetitive discussions, non-sequiturs, and conclusions unsupported by
law or logic."
Veteran attorneys tend to be keenly aware of the need to set their points
out clearly on the page. Supreme Court advocate Robert Stern (1981) de-
scribes that one of the most common faults from inexperienced brief writers
is they do not make their arguments suciently coherent for judges to un-
derstand. According to this logic, when lawyers' writings are muddled in
lengthy, inconsequential prose, judges may lose track of the main point of
the argument(s).
While briefs are argumentative in nature, the tone of the brief is still an
element that may aect the chances of the brief's success. Accordingly,
surveyed judges requested \appropriate adversarial tone" from brief writers
(Robbins 2002, 264). This can be a ne line for attorneys to follow, especially
when confronting contentious issues. There are several factors that judges
point to, however, that may contribute to an overly negative tone.
One clear admonishment concerns written attacks directed towards other
attorneys or ocers of the court. Judge Harry Edwards (1992) and Judge
Robert Martineau (1985), for instance, both note that such attacks may
immediately detract from a judge's focus on the main argument. These are
not the only judges to acknowledge the toll a negative tone can have on a
brief. Judge Harry Pregerson (2008, 436) refers to a \shrill tone in a brief"
52
as \ineective and counterproductive" and Judge Miner (1992, 24) explains
that personal attacks tend only to \weaken the brief."
Taken together, the judges' remarks convey that briefs including attacks and
a negative tone are detrimental to their persuasive powers. Regular use of
intensiers may also be viewed as \loser language" that can attach to the
judge's view of the attorney's position (Long and Christensen 2011). While
such language may diminish the eectiveness of the brief, an adversarial tone
is still expected. Based on this assessment, judges are looking for a balance
between argumentation and overly negative and con
ictual statements.
The last two points of brief quality relate to the judge's focus. The rst
of these factors is brevity. While briefs that lack complete arguments and
treatments of the facts will not serve attorneys' purposes, judges do not
expect or wish attorneys to expound lengthy prose to make their points.
Justice Ginsburg (Garner 2010b, 137) pointed this out as a
aw in many
attorneys' cases, \Lawyers somehow can't give up the extra space, so they ll
the brief unnecessarily, not realizing that eye-fatigue and even annoyance will
be the response they get for writing an overlong brief." In response to what
he nds to be the biggest shortcoming in briefs, Justice Scalia responded,
\Prolixity, probably...You don't have to use the 40 pages if that's what you're
allotted. Use as much as is necessary to make your point" (Scalia in Garner
(2010b, 53).
The Justices clearly demand succinct briefs that only deal with aspects of
the case relevant to the Court's inquiry. While the Supreme Court docket
has substantially shrunk over the last several decades (Owens and Simon
2011), the Justices still want to focus on the salient features of each case.
The question on certiorari should be the focus of the brief and straying from
this might show the Justices that the attorney is not concerned with issues
relevant to the case. Justice Scalia is clear on this point, \The framing of
the question is crucial...I have seen that happen: not included within the
53
question presented. So you make that argument and, you know, too bad"
(Scalia in Garner (2010b, 73). Justice Breyer underscores this point stating,
\...we've taken the case to decide an issue - one issue usually, maybe two -
and we are not looking so much at the whole case" (Breyer in Garner (2010b,
161)). Along with a shrinking docket, there has been a substantial growth in
the number of cert petitions and amicus briefs led over recent years. The
Justices and their clerks need to process this additional information, which
counterbalances their lighter caseload.
The Justices and their clerks are not only concerned with the length of briefs.
They also prefer interesting over dull writing (Miner 1992, 20). Engaging
writing may gain the Justices' attention and focus Justices on the statements
made in the brief. This further highlights the importance of a coherent
rather than verbose brief (Peck 1983). Scholars point to the paramount
signicance of good general writing skills that make the points in the brief
easy for judges to understand (Stern 1981).
The Justices generally agree that well-written briefs gain their attention.
For Justice Scalia, this ability to focus the Justice's attention is the main
advantage to good writing as he explains with the example, \My attention
was xed on that brief. I'd been reading a lot of other briefs, and they did
not grab me the way this one did. That's the payo. That's the payo. It is
clear" (Scalia in Garner (2010b, 73). Justice Alito connects good writing to
the lawyer's persuasive ability, \I think there is a clear relationship between
good, clear writing and good, clear thinking. And if you don't have one, it's
very hard to have the other" (Alito in Garner (2010b, 170)).
There are a slew of examples of attorneys using long-winded and hard to
follow sentences. These may confuse the Justices about the attorneys' ob-
jectives and cause the Justices and clerks to lose focus on such briefs. How
54
do such phrasings look? In the school desegregation case of Bradley v. Rich-
mond School Board,
6
the School Board's attorneys include a sentence in the
brief,
\In this context, any limitations which failed to extend the scope
of the award back to the time of this Court's decision in Brown
v. Board of Education, 347 U.S. 483 (1954) would be arbi-
trary and productive of the incongruous result that many of the
school authorities who, with the massive resources of state trea-
suries at their disposal, openly deed this Court's earlier man-
dates against segregated education would escape the reach of any
charge for the payment of fees incurred in the torturous litiga-
tion which those seeking admission to schools on an equal basis
were forced to undergo."
7
This one hundred word sentence makes several contentions, which when
combined become quite dicult to follow. Justices and clerks have to parse
such convoluted sentences to make sense of the details of the argument and
such complexity may lead to the Court's diminished focus on the brief and
potentially to less consideration of the points therein.
As scholars and judges allude, there is an expectation that many of the
features of high-quality briefs are connected. For instance, well-written briefs
should be succinct and clear. In this Chapter's analysis, I expect many of
these factors to be connected if these assessments are correct. Based on the
expectations set forth by the Justices themselves, the primary brief quality
hypothesis is: The Court will share more of its opinion language with higher
quality briefs.
6
416 U.S. 696 (1974)
7
1973 WL 172306 at *12
55
Attorney credibility should play a large independent role in the Justices'
adoption of language from briefs. While attorneys learn more about the Jus-
tices' preferences from increased experience in the Court (McGuire, Vanberg
and Yanus 2007), by the time attorneys begin their practice, in the Supreme
Court the quality of their legal writing should be fairly solidied.
8
Even if
the Justices share more language on the margins with higher quality briefs,
there is evidence that they share more language with briefs from attorneys
with greater credibility as well (Black and Owens 2012). The second hypoth-
esis is: Controlling for dierences in brief quality, Supreme Court opinions
will share more language with more experienced attorneys' briefs.
3.1.3 Empirical Understandings of Quality
There is little empirical work in the area of merits brief quality, especially
with large-N samples. Much of the work that purports to analyze brief
quality either does so with unclear conceptual denitions or with proxy
measures for quality. One example of this is in Kearney and Merrill's (2000,
801) paper, \The In
uence of Amicus Curiae Briefs on the Supreme Court."
In that paper, the section on brief quality begins, \Because reading and
assessing the quality of more than 12,000 individual amicus briefs was a
task far beyond our endurance, we had to come up with a proxy for briefs
that contain information valued highly by the Court." The proxy measure,
also adopted in other scholarly works, is based on attorney experience (Box-
Steensmeier, Christenson and Hitt 2013; Corley, Collins and Calvin 2011;
Corley 2008). What these studies of briefs lack are measurements based on
the actual words of the brief.
Another proxy measure that is designed to account for brief quality is
whether the brief is submitted by the OSG. OSG briefs are often touted
8
Based on available scholarship I would expect greater dierences in brief quality based
on attorney experience in lower courts.
56
as the highest quality (Pacelle Jr. 2003, 45). The high quality of briefs
from the OSG is often suggested without a denition of the concept of brief
quality, and so the measure of quality may be wrapped up with the high re-
gard the Justices hold for the SG's credibility (Black and Owens 2012; Segal
1988). To this point, Justice Ginsburg said, \It's never a problem with the
SG. Even if I disagree with the argument, I know that the brief will give an
honest account of the authorities. That's very important; I know I can trust
the SG's brief" (Ginsburg in Garner (2010b, 137).
One area of scholarship that tackles the question of the relationship between
brief quality and attorney success is experimental in nature. These stud-
ies compare judges' responses to dierent linguistic framings. Several such
studies, for instance, determined that judges found similar arguments in
plain English more persuasive than in legal jargon (Flammer 2010; Benson
and Kessler 1986). These studies focus more precisely on the relationship
between words in attorneys' briefs and judges' decisions, yet they are purely
hypothetical in nature. This paper moves beyond the hypothetical by inves-
tigating the relationship between the quality of existing briefs and success
before the Court.
Recent forays into the relationship between brief quality and attorney suc-
cess focus on the readability of briefs. These studies utilize readability algo-
rithms such as Flesch Reading Ease scale to determine the ease of reading of
existing briefs (Coleman and Phung 2010). The algorithms generate mea-
sures based on the relative numbers of words, sentences, and syllables in a
text. The relationship between readability and the chances of a brief's suc-
cess is still unclear. In one study utilizing readability measures, for instance,
the authors found no correlation between the readability of briefs based on
such algorithms and case outcome (Long and Christensen 2012).
Finally several papers employ automated content analysis to deduce cer-
tain aspects of legal texts. These articles use Linguistic Inquiry and Word
57
Count (LIWC) to gauge the linguistic complexity of legal opinions and briefs
(Owens and Wedeking 2012, 2011; Corley, Collins and Hamner 2013). LIWC
uses dictionaries of words that measure a sample text along multiple cate-
gories or dimensions. The program then provides an output that includes
the percentage of words in a text that belong to each category. The above-
mentioned studies, for instance, cluster several word categories that relate
to cognitive complexity to generate their metrics.
3.1.4 The Role of Brief Quality
Statements from judges make clear that they place focus and emphasis on
high quality briefs. High-quality briefs do not necessarily win cases however.
Justice Breyer is keyed into this point, \...if you don't have a sound view
as to how these cases should come out, how the law should t together, I
doubt that you could make up for it by good writing. If you're very clear,
you might just be very clearly wrong" (Breyer in Garner (2010b, 159)).
In close cases, a high quality brief may persuade the Justices and clerks
to focus their attention on the arguments from that particular brief. On
the other hand, when the law is clear, the brief is limited in its ability to
shift the decision outcome. Even if the law is clear, well-written briefs can
in
uence the Court to share more language with a particular brief regardless
of whether the brief is written for the winning or losing party (Corley 2008;
Black and Owens 2012).
Based on this notion, this Chapter relies on an outcome variable derived
from the percentage of an opinion's language that is shared with a given
merits brief. Although winning briefs typically share more language with
the Court's opinions, this is not always the case (Corley 2008). A high-
quality brief should persuade the Justices and clerks' to place greater focus
on it during their deliberations. In eect, a lawyer may be able to compen-
sate for a position that is not likely to win with a well-written brief that
58
persuades the Court to share a maximum amount of opinion language with
the brief. For attorneys and parties concerned with the Court's shifts in the
law over time, such incremental benets can have large downstream payos
(Hathaway 2001). Persuasion from losing briefs may also involve limiting
the magnitude of a negative outcome.
Here, quality may play a similar role to attorney credibility. While credible
attorneys with high levels of experience have the ability to discriminate
between cases they handle that are more or less likely to win, inevitably they
will represent clients with losing cases. While their briefs cannot change
the facts or law relevant to the case, they can persuade the Court that
their argument is sound. As clerks and Justices are apt to read briefs from
experienced counsel more closely, especially those from the OSG, these briefs
can persuade the Court to rely on them in resolving issues extraneous to the
main outcome, or in limiting the eect of the main outcome, if the outcome
is so clear to the Court based on factors extraneous to the briefs.
3.2 Additional Factors
The limited existing anecdotal and empirical evidence documenting the re-
lationship between briefs and opinions presents an impression that Justices
and clerks use language from merits briefs in varying amounts in their opin-
ions (Corley 2008; Black and Owens 2012; Epstein and Kobylka 1992). In
this Chapter I also test this linguistic relationship on a much vaster scale
than in any prior work. I break this relationship down into several addi-
tional, testable hypotheses below.
59
3.2.1 Justices
The Justices are political actors (Epstein and Knight 1998) and have prefer-
ences on issues, which are re
ected in the consistency of their voting habits
(Segal and Spaeth 1996). It is unsurprising then that ideology is a strong
predictor of the Justices' votes on the merits (Segal and Spaeth 2002). The
extent of the role of ideology in determining whether an opinion shares more
or less language with a brief is less clear. Previous work on the subject sug-
gests that ideology should play a large role in the Justices choice of which
briefs they share more and less language with in their opinions leading to the
prediction that: Justices with strong ideological preferences will draft opin-
ions that share more language with briefs submitted by parties with whom
they are ideologically aligned.
3.2.2 Attorneys
Repeat players have an advantage in conveying their cases to the Justices
and clerks (Galanter 1974; McGuire 1993). They have already established
themselves before the Justices and have a preconception of what the Justices
are looking for in a brief. The reputation of the lawyer and law rm and their
history before the Court may provide the Justices and clerks cues as to the
credibility of their writings (Aldisert 1996). The development of a \Supreme
Court Bar" of repeat Supreme Court litigators in recent years exemplies
this specialized Supreme Court expertise (Lazarus 2008; Roberts 2005).
If repeat opportunities before the Court matter, I expect attorneys and law
rms with more experience in the Court to prepare briefs with which the
Court shares more of its opinion language. This leads to the expectation
that: the Justices will adopt a greater amount of language from briefs sub-
mitted by more experienced attorneys and law rms.
60
Beyond the likelihood that experience in the Court matters, there is evidence
that suggests attorneys from the OSG (Oce of the Solicitor General) fare
better before the Court than their experience alone would dictate (Black and
Owens 2012). Attorneys from the OSG have several advantages over other
experienced litigators. They have an institutional connection to the Court
through the government. This may also lead to strategic decisions that mil-
itate the Justices' decisions in favor of the OSG (Epstein et al. 2007). The
SG is also the most consistent repeat player in the Court, which establishes
the SG's credibility and rapport with the Justices (McGuire 1995). This
leads to the expectation that: the Court will share a larger amount of lan-
guage on average with briefs submitted by the OSG than with briefs from
other Supreme Court litigators.
Conversely, the Court may systematically share less language with briefs
from parties by virtue of these parties' systematically less persuasive tradi-
tions of advocacy. There is evidence that states tend to fare less successfully
in the Court than other similarly situated parties as they have been crit-
icized, \by scholars and Justices alike as poor litigators in the Supreme
Court" (Morris 1986, 300). This leads to the expectation that: the Court
will tend to share a low amount of language with briefs submitted by states'
attorneys.
3.2.3 Cases
The Justices have more expertise in certain areas of law than in others.
Lawyers often know when the Justices have experience with particular areas
of law and with this knowledge they draft briefs that are more or less tech-
nical in their language (Robbins 2002). The Justices regularly hear cases
with constitutional issues such as civil rights or discrimination claims. The
Court also hears cases in a variety of areas where specic expertise is re-
quired. This leads to the expectation that: opinions will share less language
61
with merits briefs in cases that deal with civil liberties than they do in other
areas of law.
Cases that are viewed as particularly important to the Justices may persuade
the Justices and clerks to utilize more extensive resources in coming to their
decisions. The Justices are incentivized to expend more energy on cases that
are viewed as more consequential (Epstein, Landes and Posner 2013). This
leads to the expectation that: the Court will share less language with merits
briefs in more salient cases.
When there are a greater number of issues in a case, there is reason to
suspect that the Court will look for a larger number of sources than it does
in less complex cases (Maltzman, Spriggs and Wahlbeck 2000) leading to
the expectation that: the Justices will share less language with merits briefs
in more complex cases than they do in less complex cases.
3.3 Methods
To test hypotheses regarding the amount of opinion language shared with
the brief, I use two-level hierarchical models on a newly developed dataset
composed of all orally argued cases with exactly two merits briefs for the
1946 through 2013 Supreme Court Terms. The main dependent variable is
the percentage of language in the opinion that is also located in a merits
brief.
After collecting briefs and opinions for these cases I examined the language
overlap between each individual brief and the corresponding opinion. This
measure of language overlap suggests that the Court either relied on the lan-
guage in the brief or when reviewing the brief found the language suciently
relevant to be included in the opinion. I created separate text les for each
brief and opinion in every case. I then ran the briefs and opinions through
62
the software WCopynd 4.1.1 (Bloomeld 2014). This program allows for
pairwise comparison of documents to analyze instances of shared language.
The user inputs a base document (the opinion) with secondary documents
(the case briefs) to locate similarities in the language used. I maintained the
program's default settings in a similar manner to Corley (2008) and Owens
and Black (2012) so that the program would highlight exact or extremely
similar language. Accordingly, the program was set to pick up phrases of at
least 80% overlapping language between the brief and opinion. The mini-
mum length of each phrase was set to six words. These settings are designed
to ensure the program focuses on common language in phrases of sucient
length to be meaningful.
The multilevel model is appropriate when there are strong correlations be-
tween observations (Kenny 1996). In the case of the amount of language
opinions share with briefs, there tends to be a high level of correlation be-
tween the amount of language opinions share with both briefs in the same
case. To deal with this feature of brief/opinion relationships, I model both
briefs in a case as nested in a dyad. I do this by creating separate obser-
vations for each brief/opinion relationship, but also by creating a common
identication for both briefs in a case.
Since scholars cite multiple aspects of writing as consequential for a brief's
quality, I measure the quality of briefs along multiple dimensions. The
indicators of quality discussed by scholars break down into two categories:
those dealing with word-choices and those dealing with sentence structure.
Judges and Justices seem to seek language and structure that make the
writing easily comprehensible.
To move beyond past measures, I used dictionary-based software. The rst
tool, StyleWriter 4, provides the indicators for the bulk of the factors as-
sociated with brief quality including: wordiness, lively language, passivity,
63
and sentence complexity. StyleWriter is writing editor software with set-
tings that can be modied for specic industries and purposes such as law.
StyleWriter has a built in 200,000 graded word list and 50,000 word and
phrase style and usage checker to analyze the use of plain language.
9
Although StyleWriter measures the quality of writing, it lacks measure-
ment for one very important dimension - sentiment. Current works in many
academic disciplines utilize sentiment analysis to measure the tone of doc-
uments. I use a modied version of SentiWordNet (Baccianella, Esuli and
Sebastiani 2010) to measure the sentiment of the briefs in the dataset.
To focus on the impact of factors relating to a brief's quality that impact the
amount of opinion language that overlaps with each brief, I generated multi-
ple control variables.
10
The rst of these controls is Complexity. Complexity
is a measure of the number of legal provisions relied upon and issues raised
in the case as coded in the Supreme Court Database (Benesh and Reddick
2002). As case complexity rises, the Justices may look to a larger pool of
resources in drafting the opinion. Additional complexity should decrease a
brief's overlap value.
11
Next, Legal Salience is a dummy variable that is coded 1 in cases where the
Court strikes down a law as unconstitutional or overturns its own prece-
dent (as coded in the Supreme Court Database) (Maltzman, Spriggs and
Wahlbeck 2000). Political Salience examines when the case is salient to the
public and to elites. When a case is discussed on the front page of the New
York Times the day after the decision is handed down (Epstein and Segal
2000a), I code Political Salience as 1. Both salience variables should also
9
StyleWriter editing software can be found at
http://www.editorsoftware.com/StyleWriter.html.
10
Some of the variables are based on those used in previous studies looking at similar
relationships (e.g. Corley (2008)).
11
In Chapter 2, I noted that the Justices and clerks should look to merits briefs and
lower court opinions more in complex cases relative to amicus brief use. In this chapter
the focus is isolated to the parties' merits briefs.
64
lead the Justices to focus on a larger pool of resources and lower the briefs'
overlap values.
The next variables relate to party type and issue area. The rst is Solicitor
General. It is coded 1 when the party on the brief is the United States or
an executive branch agency represented by the OSG. It does not account for
individual government employees. The other party variable is State. While
I expect a federal government briefs to carry a strong positive coecient,
the state variable should move in the negative direction due to the docu-
mented, poor-quality of states' briefs and the often overloaded dockets that
states' attorneys face (Morris 1986). To combine constitutional issue areas,
I clustered the Civil Liberties variables together.
12
Based on the assessment
that many civil liberties cases are highly salient for the Justices and that
the Justices' have distinct preferences in such cases, I expect this variable
to have a negative coecient (Martin and Quinn 2002).
I coded a variable for briefs associated with parties that won by a unanimous
decision of the Justices as 1. This is due to the expectation that the role of
ideology is minimized in unanimous cases thus enabling the Justices to reach
consensus on the opinion's language with fewer con
icting voices (Epstein,
Landes and Posner 2012a). Unanimous should carry a positive coecient.
In contrast, I expect ideological friction among the Justices to play a larger
role in contested decisions. I coded a dummy variable for cases where the
split of the Justices' votes is Five to Four Vote and I expect this variable to
carry a negative coecient.
Next, to account for the petitioners' advantage due to the certiorari process
and aggressive grants (Epstein and Knight 1998), I coded a dummy variable,
Petitioner's Brief, as 1 for each brief for the petitioning party and I expect
this variable to carry a positive coecient. Based on the Justices' votes on
12
Supreme Court Database Issue Areas: Criminal Procedure, Civil Rights, First
Amendment, Due Process, and Privacy.
65
the merits, I coded a dummy variable Winning Brief for the winning party
in a case. As the Justices decide the winner of the case in conference prior
to drafting the opinion, I expect the winning brief to generally set the bar
for the amount of language the Court will share with the briefs in the case.
To measure an attorney's experience in the Supreme Court, the Attorney
Experience variable tracks the number of times an attorney is listed as the
attorney of record on Supreme Court briefs. I generated this variable based
on a Westlaw search of briefs for each attorney. Because the distribution of
experience is skewed to the low end with a few signicant outliers, I use the
natural log of this experience variable. The pre-logged number increases by
1 in chronological sequence each time the attorney is listed as attorney of
record on a merits brief.
I next include a control variable for the Justices' Ideological Compatibility
with the briefs. This variable accounts for the ideological compatibility
between the Justice and brief and controls for the ideological direction of
the brief. To code this variable I use Martin-Quinn (M.Q.) Scores that
measure the Justices' ideologies based on their prior votes.
13
I coded the
dummy variable 1 when the majority opinion writer's ideological direction
accorded with the ideological direction of the brief in the observation and 0
otherwise.
14
To control for a Justice's relative workload and for the possibility that the
number of clerks in a Justice's chamber aects the amount of overlap in
a Justice's opinions, I added the variable Clerks-Per-Chamber. I expect
that more clerks function as a resource to gather greater amounts of in-
formation and so the overlap value should decrease as the number of clerks
13
These scores vary by term. The scores are negative for liberal and positive for con-
servative and range from approximately -6 on the liberal side to near 6 on the conservative
side. I only coded for ideological compatibility when the Justices' scores were either less
than -1 or more than 1 indicating that the justice is not ideologically neutral.
14
Based on the direction of the lower court decision as coded in the Supreme Court
Database.
66
increase.
15
Although clerks may have subjective views on the utility of briefs
for constructing opinions, multiple clerks in the same chamber should lead
to opinion construction based on a greater diversity of sources.
16
Central to the analyses in this Chapter, I generated several indicators to
measure the quality of briefs. As these are all indicators of the same overall
concept, I used factor analysis to collapse the factors into a latent variable
called Brief Quality.
17
All but one of the quality indicators were derived
using StyleWriter's dictionary-based indices that measure characteristics of
writing quality.
The rst indicator, Passivity measures the number of passive verbs in the
document based on the total number of sentences. Passivity makes writing
less clear and incoherent. Examples of passivity include verbs preceded by
\are" and \be."
The next two measures based on StyleWriter's term dictionaries are Lively
Language and Wordiness.
18
Wordiness measures the inclusion of some of
the most common non-pronoun words that are often used to link parts-
of-speech.
19
Overuse of these words may detract from the writing quality.
Many of these are also known as stopwords that are often removed from
other forms of text analysis. These words include \the", \and", \to", \of",
\is", and \for."
15
There is an underlying issue of a justice's supervision of clerks in this process. Al-
though there is a possibility of \rogue" clerks that do tend to rely more or less on brief
language when assisting in opinion drafting, I expect that as a general matter an increasing
number of clerks will also function as a check on other clerks to ensure they are performing
their duties in the manner expected of them.
16
Notwithstanding the tradition that individual clerks are traditionally assigned to
focus on particular cases.
17
Factor analysis was also used to validate this approach showing that all the brief
quality indicators break down to a single factor.
18
StyleWriter refers to these as \Pep" words.
19
StyleWriter refers to these as \Glue" words.
67
Because StyleWriter's indices are term-based, I have a count variable that
measures Sentence Complexity. This variable is a simple average of the num-
ber of words per sentence across a document. Increased sentence complexity
tends to create sentences that are dry and harder to follow.
Finally, with SentiWordNet I am able to measure the overall sentiment or
tone of each brief. SentiWordNet measures whether a document has a posi-
tive or negative polarity based on the WordNet database of synsets. Synsets
are synonymous terms grouped together in the database. SentiWordNet de-
termines the polarity of a document by the proportion of classiers assigned
a label within it (Baccianella, Esuli and Sebastiani 2010).
Although SentiWordNet has been validated in a variety of studies as a sen-
timent classier (e.g. Kundi et al. (2014)), it lacks one essential tool for
sentiment classication: negation identication. Negation identication is
essential to sentiment classication due to a negation's ability to change the
meaning of terms immediately following it. A simple example is the com-
parison of the phrases \the verdict was accurate" and \the verdict was not
very accurate." Although \accurate" can convey a positive sentiment, the
negation changes the meaning of the term.
To deal with negations, I created a regular expression that eliminated negated
terms from sentiment classication to prevent these negations from con-
founding the results. The sentiment scores are, therefore, based solely on
non-negated terms, both positive and negative in polarity.
Figure 3.1 presents a histogram all of the briefs' overlap values from the
dataset.
While most of the opinions share around 10% of their language with briefs,
the tail of the curve extends to the right until almost 40% language overlap.
At the end of the tail there are eight instances were an opinion shares 50%
68
Figure 3.1: Histogram of % Language Overlap of All Briefs and Opin-
ions: 1946-2014
or more of its language with a brief (the top instance of overlap is 59%).
20
We might infer from this striking similarity in language between a portion
of briefs and opinions that there is something inherent in specic briefs or
cases that motivates this high level of brief language used in such opinions.
The mean overlap value across the dataset is 9.69% while the median falls
at 8%.
3.4 Analysis
Table 3.1 presents the multilevel model results of the relationship between
brief quality as well as several other variables and the overlapping language
between opinions and briefs.
20
In all of the instances of 50% or greater language overlap there was a substantially
lower percentage of language overlap between the opinion and the other party's brief.
69
Table3.1: Multilevel Model of the Eect of Brief Quality on the Amount
of Opinion Language Shared with Merits Briefs, 1946-2013
Variable
Complexity -0.253

(0.0722)
Legal Salience -0.724

(0.200)
Political Salience -1.379

(0.188)
Solicitor General 4.006

(0.279)
State -0.518

(0.110)
Civil Liberties -0.398

(0.146)
Petitioner's Brief 1.100

(0.101)
Attorney Experience (Log) 0.315

(0.0889)
Winning Brief 1.880

(0.165)
Ideological Compatibility 0.479

(0.168)
Unanimous 0.654

(0.207)
Five to Four Vote -0.765

(0.179)
Clerks Per Chamber -0.0818 (0.0772)
Brief Quality 0.794

(0.0700)
Constant 8.019

(0.296)
Variance of Constant 1.279

(0.0305)
Variance of Residual 1.467

(0.0285)
N 9498
ICC .408
PRE .2005
Robust standard errors in parentheses clustered on Supreme Court term.

p< 0:05,

p< 0:01,

p< 0:001
Model t using maximum likelihood
Before examining the results, there are several checks I performed to ensure
that the model is correctly specied and that the multilevel model accounts
for the presumed correlation between merits briefs' overlap values in a case.
First, a likelihood ratio test between the multilevel model and a linear re-
gression is signicant at the 0.001 level. The variance of the residuals is
signicant at the 0.001 level, which also helps support the presumption that
the model is accurately specied. The reduction of error (PRE) in the
two-level model over the one-level model is 20.05%. Finally, the intraclass
70
correlation coecient (ICC) is .408.
21
As Table 3.1 shows, brief quality aects the amount of opinion language the
Court shares with merits. The control variables all move in the predicted
directions. The variables with the greatest magnitudes are for winning briefs,
the presence of the Solicitor General, and for petitioner's briefs. These three
variables have positive coecients indicating they lead to a likelihood of
more shared language between merits briefs and opinions.
Both politically and legally salient cases, as well as more complex cases,
lead to decreased shared language between briefs and opinions. This is
not surprising given the expectation that the Justices and clerks look to
additional sources in drafting opinions when these factors are present. Also,
as predicted, the Court shares less language with states' briefs.
Increased attorney experience positively aects the amount of language opin-
ions shared with merits briefs. This nding supports the proposition that
credibility plays a role in the amount of language the Justices and clerks
share in their opinions with the briefs. The Justices tend to share more
language with more experienced attorneys' briefs. Opinions also share more
language with briefs for winning parties in unanimous decisions and tend to
share less language with both briefs in cases decided by ve-to-four votes. Fi-
nally, and as expected, the Justices tend to share more language with briefs
based on their ideological compatibility with the direction of the briefs.
Most importantly, the result for brief quality is signicant and moves in the
predicted direction. To understand the magnitude of the eect of brief qual-
ity, Figure 3.2 below depicts the marginal eects of increased brief quality
on opinions' overlap values.
21
This measure was derived with non-robust standard errors, yet its magnitude suggests
a substantial portion (over forty-percent) of the residual variance is due to the dyadic pairs
conditional on the top-level factors of the two-level model.
71
Figure 3.2: Marginal Eects of Brief Quality on Overlap Values
Note: Dashed outer lines represent 95% condence intervals.
The gure shows that as brief quality increases, so does the amount of
language the Justices and clerks share with briefs. There is more than a
fourfold increase in the amount of language that opinions tend to share
with briefs that meet the upper bounds of the brief quality when compared
with briefs at the lower bounds.
22
This suggests that an increased focus
on writing quality does indeed enhance the Court's reliance on particular
merits briefs.
Table 3.2 underscores the dierence between the amount of language the
Court shares with high and low quality briefs. This table examines statistics
from the top 100 and bottom 100 briefs based on the quality measure.
Table 3.2 shows that the average amount of language that the Court shares
with the top 100 briefs in the dataset based on brief quality is almost double
that for the bottom 100. The overlap values for the top 100 briefs based on
22
The margins command in Stata bases its results on mean values for all other variables.
72
Table3.2: Overlap Score Statistics by Top and Bottom 100 Brief Quality
Scores
Statistic Top 100 Bottom 100
Mean 10.32 5.67
Median 9.00 4.50
Variance 28.60 38.85
Skewness 1.37 2.91
brief quality are also less dispersed as the variance and skewness are both
smaller than those for the bottom 100 briefs. These values indicate that the
top 100 briefs' overlap values are less driven by outliers than the bottom
100.
To help conrm the importance of brief quality Table 3.3 looks at the asso-
ciation between higher brief quality and winning in the Supreme Court.
Table 3.3: Multilevel Probit Estimates of Likelihood of Winning
Variable
Petitioner's Brief 0.551

(0.0325)
Solicitor General 0.402

(0.0631)
State 0.220

(0.0343)
Ideological Compatibility 0.378

(0.0381)
Attorney Experience (Log) 0.0635

(0.0160)
Brief Quality 0.0373

(0.0137)
Constant -0.538

(0.0225)
N 9498
Robust standard errors in parentheses

p< 0:05,

p< 0:01,

p< 0:001
The dependent variable in Table 3.3 is a dichotomous variable coded as 1 if
the brief is for the winning party and 0 if the brief is for the losing party.
The results of this model show that the quality of briefs is in fact signicant
in winning cases as brief quality positively aects the likelihood of winning.
The likelihood of winning a case increases by approximately 20% by moving
73
from the low end of the brief quality spectrum to the high end.
23
As the
quality of briefs enhances both the language the Court shares with the brief
as well as the likelihood of winning the case, briefs' quality play a major role
in both of these Supreme Court case outcomes.
Do more experienced Supreme Court attorneys draft higher quality briefs?
On one hand, one might expect that by the time attorneys le briefs in
Supreme Court cases they have already honed their legal writing skills in
law school and through legal practice. Without additional input and writ-
ing training, an attorney's writing ability should be well dened by the time
practitioners begin writing Supreme Court briefs. On the other hand, ex-
perienced Supreme Court practitioners may have gained a specic skill set
that only those with such experience can acquire. The results in Table 3.4
provide evidence of the relationship between Supreme Court attorney expe-
rience and brief quality and show that to some extent both suppositions are
correct.
Table 3.4: Overlap Values and Quality Index by Attorney Experience
Level
Overlap Value
(mean)
Brief Quality
(mean)
N
Overall 9.55 0.000 9498
Repeat Player 11.543 0.053 3415
Non-Repeat Player 8.439 -0.030 6083
Dierence 3.104 .083
Note: Dierence in means tests for overlap and quality both show means
are statistically dierent from each other at any the .001% level. Repeat
Player refers to attorneys with more than one brief led in the Supreme
Court.
23
This is based on predicted probabilities where all other variables in the models are
set to their mean values.
74
According to Table 3.4, there is a slight dierence in the quality of briefs from
more experienced attorneys where their briefs, on average, are of higher qual-
ity. This may be attributed to their increased Supreme Court brief writing
experience. The dierence in mean overlap scores between these two groups,
however, is quite large at over three-percent per opinion. Dierences in brief
quality alone cannot account for this large dierence in overlap values. This
leads to the conclusion that an experienced attorney's credibility before the
Court enhances the overlap values of their briefs more than quality alone
would indicate.
3.4.1 Findings for the Justices
One way to analyze the dierential utility of merits briefs is by examining
the data broken down by individual Justice. Figure 3.3 presents box plots
for each justice.
Several of the Justices including Douglas, Murphy, Whittaker, Minton, and
Sotomayor have median values clearly over 10%. Indeed each of the Justices'
interquartile ranges exceeds 10% except for Justices Jackson and Kagan.
There is a clear decrease in the maximum values of language overlap per
Justice over time potentially indicating that the Court's shrinking docket
led to less reliance on the merits briefs.
24
The range in median language overlap value across Justices is almost as large
as the value for the justice with the smallest overlap value (Justice Kagan
has a median overlap value of 6.5%). On the other end of the spectrum,
Justice Murphy has the largest median overlap value at 12.5%. Since the
Burger Court, the Justices with median overlap values of 10% or greater are
Justices Thomas, Ginsburg, and Sotomayor.
24
Although in Chapter 2 I note that overall overlap values (combining overlap with
amicus briefs, merits briefs, and lower court opinions) have increased in the short term,
the metric in this Chapter shows that reliance on merits briefs has decreased over time.
75
Figure 3.3: Box Plots of Language Overlap for All Justices: 1946-2014
Note: Justices listed chronologically by date of appointment.
Note: In this plot the boxes include the interquartile range with the median line in the
middle. The lines extend to 1.5 interquartile ranges and all outliers are marked with
dots.
To derive a clearer sense of the Justices' average overlap values and the
standard deviations for these values, Table 3.5 presents these gures for all
Justices included in the section.
Variation in these mean values is not a function of the internal machinations
of the Justices and clerks alone. They are confronted with varying levels
of the macro-level factors that aect the relationship between merits brief
and opinions based on the cases they are assigned (or to which they assign
themselves as is often the case with the Chief Justice and senior associate
76
Table 3.5: Mean Overlap and Standard Deviation by Justice
Majority Opinion Writer
Organized Chronologically
By Date of Appointment
Mean
Overlap
Standard
Deviation
Black 8.100 6.083
Reed 9.493 6.135
Frankfurter 7.967 6.359
Douglas 12.649 8.778
Murphy 13.854 8.985
Jackson 6.455 5.126
Rutledge 7.647 7.781
Burton 10.500 5.802
Vinson 9.173 6.949
Clark 9.736 6.146
Minton 12.978 7.698
Warren 9.563 7.969
Harlan 9.234 6.689
Brennan 8.638 6.147
Whittaker 13.597 11.297
Stewart 10.429 7.367
White 8.541 5.138
Goldberg 12.304 9.050
Fortas 9.274 6.366
Marshall 9.594 5.685
Burger 10.115 6.551
Blackmun 9.905 7.560
Powell 9.082 5.868
Rehnquist 10.622 6.458
Stevens 8.907 4.721
O'Connor 9.502 5.441
Scalia 8.740 4.747
Kennedy 8.321 4.428
Souter 8.437 4.625
Thomas 11.185 5.441
Ginsburg 10.546 5.532
Breyer 7.473 3.234
Roberts 9.189 4.508
Alito 9.481 4.756
Sotomayor 11.044 3.945
Kagan 7.130 3.410
Overall Average 9.537 6.262
77
justice) which may account for at least some of the dierences in values
between Justices.
Another way to look at the Justices comparatively is through their maxi-
mum overlap values. This shows their potential willingness to incorporate
a high level of language found in a merits brief. The maximum value has
decreased appreciably over time. The last justice (by date of appointment)
with a maximum value of 40% or greater was Justice Powell. Thirteen of
the Justices have maximum values of 40% or greater including and all were
on the Court prior to the appointment of Justice Powell.
25
Since Justice
Powell's appointment, the two highest overlap values in a case are 37% each
for Justices Rehnquist and O'Connor. Justice Ginsburg is the last justice
appointed with an instance where the percentage of overlapping language
between a brief and opinion exceeded 30%. The two Justices with the low-
est maximum overlap values in a case are Justices Breyer and Kagan with
values of 16% and 15% respectively.
Next I examine the Justices' ideological sensitivity to the parties' merits
briefs. Given the predictive value of the Justices' ideal point scores (Martin
and Quinn 2002), ideology may be expected to play a role in understanding
the Justices' choices of opinion language and choices of which briefs they
and their clerks use as resources in drafting the opinions. To measure this
relationship, I looked at the decision direction of the lower court's ruling
which denes the direction of the parties' briefs at the Supreme Court level.
I then took the dierence in average overlap values for each justice when the
lower court ruled liberally and when the lower court ruled conservatively.
The average dierence across Justices in their ideological sensitivity to briefs
based on the lower court's ruling is not particularly large at .41%. Several
of the Justices, however, display strong sensitivities to the ideological direc-
tion of the briefs in the dierence in language they incorporate from briefs
25
This may also be a relic of shorter opinions in the past.
78
depending on the ideological position of the party. Table 3.6 displays the re-
sults for all Justices that have a 2% or greater dierence in language overlap
values for their opinions based on the ideological direction of the briefs.
Table 3.6: Justice Ideological Sensitivity (of 2% or Greater)
Justice Cons Lib % Dierence Direction
Whittaker 9.50 17.44 7.94 lib
Rutledge 3.86 10.30 6.44 lib
Fortas 8.46 12.07 3.61 lib
Vinson 7.87 10.77 2.90 lib
Murphy 13.33 15.80 2.47 lib
Warren 9.08 11.16 2.08 lib
Goldberg 11.95 14.00 2.05 lib
Thomas 12.33 10.31 2.02 cons
Note: Cons=Conservative Brief, Lib=Liberal Brief, % Dierence=the absolute value of
the dierence between Cons and Lib, and Direction=the ideological direction of this dif-
ference.
Almost all of the Justices listed with a strong propensity to share brief
language depending on the ideological direction of the brief sat on the Court
prior to the 1980's. All of those Justices favored language from liberal
briefs. Justice Rutledge and Whittaker's dierences between overlap values
for liberal and conservative briefs at 7.94% and 6.44% respectively are almost
double that of the justice with next highest value: Justice Fortas at 3.61%.
The remainder of the Justices in Table 3.6 t into the 2-3% dierence range.
Justice Thomas is the only sitting justice with a dierence value of over 2%.
26
Still, other Justices whose votes on the merits are strongly associated with
their ideological preferences do not t this pattern. Only seven other Justices
have dierences in overlap values depending on the ideological direction of
the brief of greater than 1% and all have values under 2%. This indicates
that in some instances the Justices' are sensitive to the ideological direction
26
Since he is often touted as a staunch conservative justice (Smith 1996), his preference
towards conservative briefs may not be surprising.
79
of the briefs, but that for the majority of the Justices the ideological direction
of the briefs does not have a large impact on the Justices and clerks' choices
of which merits brief to focus on in a case.
3.5 Findings by Attorneys and Party Type
When attorneys are listed as counsel of record on multiple briefs over time,
this signals the credibility of their arguments to the Justices and clerks
as these attorneys gain greater exposure to the Justices. To examine the
signicance of attorney experience and of support from the OSG, Figure 3.4
below plots linear ts plots for increasing attorney experience and language
overlap values both including and excluding briefs on which the SG is counsel
of record.
This gure supports the notion that increased experience is related to in-
creases in overlap values between opinions and briefs. Not accounting for
OSG briefs, the least experienced attorneys' overlap value on the plot is
just over 8% while the most experienced attorneys' (which includes attor-
neys leaving the OSG to return to private practice) values on the plot are
between 11% and 12%. The unique relationship between the OSG and the
Court is evident as well. Accounting for similar levels of experience, the plot
reaches almost 16% for the most experienced attorneys when OSG attorneys
are included with all other attorneys.
Another way to view the OSG's advantage is to examine the government's
success relative to other party types. The OSG's relative superiority in
overlap value is evident in Figure 3.4, but the gure does not convey how
well the US Government fares relative to other party types. Figure 3.5
presents box plots with this alternative way of looking at the government's
merits briefs success. The boxes in this gure are based on instances when
the party types listed are petitioners in the case.
80
Figure 3.4: Fit Plots of Language Overlap Based on Logged Attorney
Experience
Note: The top line includes instances involving the OSG while the lower line is based on
instances excluding the OSG's briefs but including previous OSG attorneys while in
private practice.
Once again the federal government's success relative to all other party types
is clear. The federal government's median overlap value (indicated by the
vertical line in the US Government's box) is the only value noticeably dier-
entiated from those for the other party types. But how do we quantify this
advantage? Figure 3.6 presents a way of doing so by providing the relative
dierences in the US Government's mean overlap value as petitioner and
those for the other petitioning party types.
Not surprisingly, the two party types with the lowest mean overlap values
(those with the greatest separation from the federal government) are state
81
Figure 3.5: Box Plots of Language Overlap Between Petitioner's Briefs
by Party Type
governments with more than a 3% mean dierence and criminals.
Beyond separating parties by type, there are also advantages that the parties
have in a case due to their procedural posture. Theoretically, petitioning
parties have an advantage in winning Supreme Court cases (Epstein and
Knight 1998). Based on the theory of aggressive grants and on available
evidence, there is reason to believe this trend should hold across time. Figure
3.7 looks at the dierences between overlapping language in opinions with
petitioners' and respondents' briefs for the 1946 through 2013 terms.
As predicted, the line representing the overlap values for petitioners' briefs
82
Figure 3.6: Percent Dierence in Overlap Value Between US Govern-
ment and Other Party Types (Petitioners' Briefs)
consistently lies above the line for respondents' briefs.
27
The median value
for petitioners' briefs is 9% with a mean value of 10.43% while the me-
dian value for respondents' briefs is 7% with a mean of 8.64%. The greatest
amount of dierence was in 1963 when opinions shared 4.45% more language
with petitioners' briefs on average. Interestingly, there is noticeable varia-
tion in the dierence between the two values over the course of the terms in
the dataset. While petitioners' briefs averaged 1.86% more language over-
lap with opinions than respondents' briefs across all of the terms, there were
anomalies to trend as is evident in Figure 3.7. Opinions shared more lan-
guage on average with respondents' briefs during six non-consecutive Court
27
A t-test also showed that the dierence in means between overlap values for petition-
ers' and respondents' briefs is signicant at the .001 level.
83
Figure 3.7: Comparison of Language Overlap Between Briefs of Peti-
tioner and Respondents and Supreme Court Opinions (1946-2013)
Terms between 1946 and 2013.
3.6 Case Context
Figure 3.8 presents a ne-grained way of looking at overlap values by specic
issue area. The gure uses the Supreme Court Database's coding for issue
area to split the brief/opinion relationships into eleven groupings.
There are two main takeaways from these box plots. First, overlap values
for briefs in the civil liberties issue areas are relatively consistent and low
with the possible exception of privacy.
28
Second, the two highest median
overlaps values are in the issue areas of Unions and Federal Tax. These
are both statutorily dominated compared to the primarily constitutional
28
The top ve boxes.
84
Figure 3.8: Comparison of Language Overlap by Issue Area
cases dealing with civil liberties. Federal Tax and Union briefs and opinions
often have lengthy overlapping statutory sections, which may account for the
greater overlap values in these two areas. Federal tax cases are also often
more complex than other case types in the gure and this added complexity
could add to greater reliance on the merits briefs.
One additional way to look at the relationship between the language in
briefs and opinions is to examine this relationship over time. If additional
resources work to crowd-out the centrality of merits briefs then we might
expect a decrease in the average overlap values over time. Figure 3.9 plots
the mean overlap values on a term-by-term bases across the 68 terms within
this dataset.
While there is a decrease in language overlap over time, the decrease is quite
minimal. This data makes any inference about the change in overlap values
over time at best inconclusive. Another interesting aspect is that there is
85
Figure 3.9: Mean Brief/Opinion Overlap Values (1946-2013)
a decrease in variability in overlap values over time. While there was large
variation between the 1946 and 1976 terms, this variation diminishes signif-
icantly after the late 1970's. This is apparent in the decrease in standard
deviation of overlap values from 6.88% prior to 1979 to 5.68% from 1979
through 2013.
3.7 What Matters?
This Chapter shows there are four primary sets of factors that aect the
Justices and clerks' utility of merits briefs as they draft Supreme Court
opinions. The rst has to do with the brief's writing quality. The analyses
in this Chapter show that the Court is more likely to adopt language from
higher quality briefs than it is from lower quality briefs.
86
The next set of factors have to do with the Justices and their clerks. The
Justices each have distinct patterns of incorporating language from briefs in
their opinions. Some Justices are more swayed by the ideological direction of
the brief although this is not the case for even some Justices most noted for
their overt voting preferences. Interestingly, while variation exists between
the Justices' overlap values, the use of merits brief language remains fairly
consistent across time and across Justice. This provides evidence that briefs
have remained a stable resource for the Justices and clerks and a resource
they tend to use in fairly similar amounts.
The third set of factors relate to the attorneys. Experience writing briefs
for the Court generally leads to more overlapping language between briefs
and opinions. The strongest factor that aects the amount of language the
Justices and clerks incorporate from a brief in an opinion is whether the brief
was submitted by the OSG. This implies the importance of the institutional
advantage of the SG and the advantage provided by the Oce's relationship
with the Court. It also shows that experience writing briefs for the Court
makes a dierence. The SG historically submits the most merits briefs to the
Court on an annual basis giving the SG an inherent advantage in terms of
brief writing experience at this level. Experience also matters for non-OSG
attorneys indicating that there is likely a synergy between the experience
attorneys gain as they write more briefs for the Court with the credibility
attorneys gain with additional exposure to the Justices.
Finally the fourth set of factors have to do with the case type. The Jus-
tices tend to share less language with briefs in highly salient cases. This
may show that the Justices and clerks comb through additional materials
outside of the merits briefs in such cases. The decrease in language shared
between merits briefs and opinions as amicus lings increase also presents
that as Justices and clerks either seek out or are provided with additional
sources of information, they tend to share less language with merits briefs.
Similarly the Justices and their clerks may seek additional information in
87
highly complex cases leading to a decrease in shared language between briefs
and opinions as case complexity rises.
By developing original ways to test the factors in briefs that lead to higher
language incorporation in the Court's majority opinions, this Chapter pro-
vides greater insight into the Justices and clerks' opinion construction pro-
cess.
Chapter 4
The Greatest Impact of
Briefs on Opinion Language
Imitation may be the most sincere form of
attery, but in the language
of the law, the demarcation between imitation and wholesale borrowing
remains opaque (Colton 1837). This is especially true in the Supreme Court
opinion construction process. This Chapter examines the implications of the
Court's use of substantive language directly from merits briefs. To do this,
the Chapter critically analyzes the relationship between merits briefs and
opinions, focusing on instances where there is strong evidence that briefs
played a substantial role in the Court's choice of majority opinion language.
The analysis compares phrasing from briefs and opinions and locates case
type where briefs have played an especially in
uential role.
Long ago, Judge Cardozo warned of an \agglunative" style of judicial writ-
ing where whole sections are cut from other documents and reassembled
together in courts' opinions (Lasky 1961). Critics of this practice label it
88
89
judicial plagiarism while others in the legal community see this as an ac-
ceptable, common practice. While there is an expectation among those ex-
perienced in Supreme Court practice that language from briefs will seep into
the Court's opinions, there is considerable variation in the extent and types
of overlapping language (Peppers and Ward 2012; Weiden and Ward 2006).
Maybe due to this expectation, there is minimal scholarship dedicated to ex-
amining instances of high levels of language sharing between Supreme Court
opinions and the sources of opinion language such as merits briefs, as well
the normative consequences of this language sharing practice. The idea of
judicial plagiarism is a recent phenomenon (Liptak 2015). To help explain
the variation in shared language from case to case, this Chapter contains a
typology of relationships between briefs and opinions and presents examples
of the dierent types before focusing on the brief-opinion relationship where
briefs play the greatest role.
The manner in which Justices and their clerks utilize language from merits
in Supreme Court opinions shows distinct practices across Justices' cham-
bers and across time. By inquiring into the inputs for the information and
wording in Supreme Court opinions, we may begin to better understand the
opinion construction process in a fashion that interviews with clerks and
Justices alone cannot convey (Woodward and Armstrong 2011).
This Dissertation shows that from 1946 through 2013, Supreme Court opin-
ions shared 9.55% of their language on average with individual merits briefs
from their respective cases (infra Chapter 3). That analysis of almost 10,000
briefs, elucidates several patterns in the ways that merits brief language
makes its way into Supreme Court opinions. The main foci of this Chapter
are cases where the briefs have the greatest impact on the Court's opinions
or, put another way, cases where the language from a merits brief is most
fundamental to the language of the opinion. Surprisingly, even these opin-
ions with large sections uncited language that overlaps with the merits briefs
90
are not wholly devoid of citations to the briefs, which I elaborate on in the
conclusion.
In many of these instances, the Court uses language from the briefs as its
own without attribution. With these examples, this Chapter confronts the
normative question about when the Court should note that the opinion
language is not spawned originally from the Justices and their clerks (for
instance some might view overlapping language in the Court's statements
of case facts as inconsequential, while overlapping language in the Court's
assessment of the law as more worthy of attribution if unoriginal). Even
with cited material there is often strong evidence of the impact of briefs on
opinions through the sharing of citations to less common reference material
such as treatises and law reviews as well as common string cites including
multiple citations.
In addition to examining specic brief-opinion relationships, this Chapter
also looks at the factors that tie the set of cases together. In this sample
of cases there are common threads in terms of the attributes of the cases,
Justices, and litigants. These shared features show that cases with high
levels of language overlap are not entirely at-random.
The initial premise in this Chapter is that the language in Supreme Court
opinions matters. It begins by looking at why opinion language matters and
by developing a hypothesis for why we might expect the language in briefs
to lter into Supreme Court opinions; sometimes in especially high doses.
After this and a short discussion of how the section tracks the linguistic
similarities between briefs and opinions, I look at the cases of interests com-
paratively with their respective briefs. To conclude, this Chapter examines
the normative implications of this language sharing practice by focusing on
how bringing this to light may aect the perception of the way the Court
conducts its business.
91
4.1 Why Focus On Language
The strength of the Supreme Court's precedents are only as iron-clad as
opinion language permits. Time and again, issues left unsettled by Supreme
Court lead to new test cases at the lower court level and percolate back
up to the Supreme Court. This can be illustrated by the Court's sinuous
precedent in Fourth Amendment cases that dene the exclusionary rule.
One strand in Fourth Amendment jurisprudence that begins with Mapp
v. Ohio,
1
examines the exclusionary rule. Although in Mapp v. Ohio,
2
the Court broadly held that evidence obtained in violation of the Fourth
Amendment should be excluded from trial, subsequent cases either chipped
away at this holding or added teeth to it. In Schember v. California,
3
for
example, the Court upheld the Mapp ruling allowing physical evidence at
trial taken without the subject's consent, with Camera v. Municipal Court,
4
the Court held that housing inspectors need a warrant to enter and search
an apartment building, and in Immigration and Naturalization Service v.
Lopez-Mendoza,
5
the Court held that the body of the suspect is not subject
to the exclusionary rule.
The various trajectories of the exclusionary rule is the focus of much schol-
arly debate and analysis. Examples include Stephen Hessler's (2000) Estab-
lishing Inevitability without Active Pursuit, which looks at the lack of clarity
surrounding the reach of the exclusionary rule and James Kainen's (1992)
The Impeachment Exception to the Exclusionary Rules, which examines how
illegally obtained evidence still may be used to impeach a witness at trial.
As language from past cases is repeated in subsequent cases, the specic
1
367 U.S. 643 (1961)
2
Id.
3
384 U.S. 757 (1966)
4
387 U.S. 523 (1967)
5
468 U.S. 1032 (1984)
92
wording of the opinions clearly aects future decisions of the Supreme and
lower courts.
When interviewed on the subject of the Supreme Court opinions, Justice
Antonin Scalia underscored the importance of opinion language:
\the only important part about an appellate case is not who
wins or loses; it's not, you know, armed or reversed. The
important part is the opinion. And if you arm or reverse for
the wrong reason, you've done everything wrong...if you haven't
made clear what your holding is, instead of reducing litigation,
instead of making life simpler for courts and lawyers below you,
you've complicated it" (Scalia in Garner (2010d)).
Similarly, Justice Kennedy noted, \...the law lives through language and
we must be very careful about the language we use" (Kennedy in Garner
(2010c, 97)).
With this understanding, many scholars take seriously the downstream ef-
fects of the Court's opinion language in guiding and constraining future
Supreme Court and lower court decisions (Richards and Kritzer 2002; Hath-
away 2001). In The Court as an Idea, Not a Building (or a Game), Howard
Gillman (1999, 80) writes, \There is evidence that most Justices act in accor-
dance with the Court's formal responsibility to decide actual legal disputes
based on their best understanding of the law." In a similar vein Gordon
Silverstein (2009) describes that the way judges reasoned through their de-
cisions in prior cases constrains judges' later decisions.
The source of Supreme Court opinion language is relevant to the legitimacy
of the Court as an institution (Carter and Burke 1998). The \logic of the
triad" provides theoretical framework for judges' perceived role as neutral
party in legal disputes (Shapiro 1981; Sweet 1999) Based on this logic, there
93
is an expectation for the Court to use language that comes from the machi-
nations of the Justices based on their understandings of the Constitution as
well as from citations to existing law and other secondary sources (Hath-
away 2001). One of the main sources that provides potential language for
Supreme Court opinions are parties' merits briefs. Briefs organize and syn-
thesize cases for the Justices and clerks, minimizing their need to turn to
other sources (Prettyman 1953). They also distill the information so the
Justices and clerks can focus on other case attributes (Kravitz 2009). Jus-
tice Rehnquist (1999, 4) underscored this point when he wrote, \The brief
writer must...bring order to [the case materials] by organizing-and I cannot
stress that term enough-by organizing, organizing, and organizing, so that
the brief is a coherent presentation of the arguments in favor of the writer's
client." This view on judges' utility for briefs is further highlighted by legal
professionals (Tate 1978; Posner 1998).
Given that the Justices' own statements describe the brief's importance in
organizing all of the case material for their digestion, it may not be surprising
that there is often directly overlapping language between the opinion and the
brief. The extent of this relationship, however, varies immensely and studies
are, to this point, generally devoid of such analysis as qualitative assessments
of the language shared between Supreme Court briefs and opinions like this
one are sparse. Judge Posner (1993, 111) describes, \[j]udges opinions do
not acknowledge the borrowing of ideas, even language, from the parties'
briefs, so the evaluation of a judge's creativity requires comparison between
the opinion and the briefs in each case.
The methodology in this section cuts between quantitative and qualitative
ends of the spectrum. As I explain in the following section, I use quantitative
methods to locate the specic cases and briefs of interest. I then analyze the
content of the shared language to generate inferences surrounding these re-
lationships in cases with the greatest percentage of shared language between
94
opinions and briefs.
6
4.2 Measuring Language Overlap
Text analysis software allows for precise measurements of shared language.
A drawback, however, is that the language from the two sources must exactly
match and as a consequence it does not pick up on shared meaning when
the words dier. For example it would not pick up on the instances when
Cardozo used briefs' language but changed the actual words for his opinions
(Posner 1993, 109-15). In this sense software to measure language similarity
is generally underinclusive of the actual relationship between the language
in briefs and opinions. In this section I use software called WCopynd for
the analyses (Bloomeld 2014). This program compares two documents
and reports the similarities in their words both in terms of percentage of
overlapping language from one document to the other and number of shared
words. The user inputs certain settings to calibrate the requirements of
language similarity necessary for the program to note an instance of overlap.
4.2.1 Types of Language Shared Generally
The software does not discriminate between types of overlap so the matches
contain language that was likely meaningful in the brief/opinion relationship
as well as language that is most likely in the brief and opinion by pure chance
or because of its importance to the case. I use the opinion and the SG's brief
from the case Federal Energy Regulatory Commission v. Mississippi,
7
to
illustrate the spectrum of language overlap from presumably inconsequential
to likely meaningful.
6
The percentage is based on the amount of shared language relative to the total opinion
and not to the brief.
7
456 U.S. 742 (1982)
95
To dierentiate the instances of language overlap there are two main facets
that I focus on in each phrase. The rst is whether the language is orig-
inal or quoted. This includes whether the opinion cites to the brief or if
either cite to an extrinsic source. The second is the type of phrase focus-
ing on whether it is facts, argument, a citation from a statute or case, etc.
The FERC case examines the constitutionality of the Public Utility Regu-
latory Policies Act (PURPA). To describe the statute, the opinion written
by Justice Blackmun satates, \The Act was part of a package of legislation
approved the same day, designed to combat the nationwide energy crisis at
the time."
8
The government's brief written by Solicitor General Rex E. Lee
contains the language, \(PURPA), was enacted by Congress as part of a
package of legislation designed to combat the nationwide energy crisis."
9
In
this example the shared language consists of \part of a package of legislation
designed to combat the nationwide energy crisis." This descriptive, factual
language is not quoted and so it may be assumed as original in both the
brief and opinion. From this shared description it is clear that the Court
agrees with the SG's characterization of the legislation and that the opinion
does not attribute this characterization to the SG.
In this case there are many instances where the language is similar, but where
the opinion does not mention the SG's brief as a source. An example of a
statement from the recitation of facts in the Court's opinion that overlaps
with the SG's brief is the shared phrasing (In two column comparisons in
this section the brief is always on the left and the opinion is on the right):
electric utilities were plagued with
skyrocketing fuel costs and decreas-
ing eciency in the use of their gen-
erating capacity; both of these fac-
tors had an adverse eect on rates to
8
Id. at 745
9
Brief for Appellants at 2, Federal Energy Regulatory Commission v. Mississippi, 456
U.S. 742 (80-1749) (1982)
10
Brief for Appellants, supra n. 31, at 3.
96
consumers and on the economy.
10
...electricity utilities were plagued
with increasing costs and decreasing
eciency in the use of their gener-
ating capacities; each of these fac-
tors had an adverse eect on rates
to consumers and on the economy as
a whole.
11
Similar to the phrase previously mentioned, this opinion does not cite to
the SG's brief or to an extrinsic source, yet the language is an almost exact
match. Importantly, in this example the dierent word choices do not change
the meaning of the text.
Another type of shared language occurs when the brief and the opinion both
share a citation. An example of this as it appears in the opinion is, \[the
statutory section] directs FERC, in consultation with state regulatory au-
thorities, to promulgate 'such rules as it determines necessary to encourage
cogeneration and small power production...'"
12
In such instances the shared
language provides evidence of a shared understanding of the importance of
the statutory language. In isolation one can't infer that the brief writer's
choice of statutory language led to the adoption of the same language in the
opinion. Cumulatively with other instances of shared language though, this
underscores the possibility of, at a minimum, coinciding views between the
brief and opinion writer.
Opinions may also share language with the brief's argument. This is perhaps
the best evidence of the persuasive power of the brief, especially when the
phrase is not attributed to the brief's author. An example from the opinion
in the same case of shared argumentative language is the sentence, \It is
sucient that Congress was not irrational in concluding that limited federal
regulation of retail sales of electricity and natural gas, and of relationships
11
456 U.S. at 745-46 (1982).
12
456 U.S. at 751 (1982). Language not shared between brief and opinion is italicized
97
between cogenerators and electric utilities, was essential."
13
In the brief the
word \sucient" is replaced with \clear enough"
14
which does little to alter
the meaning of the phrase. The opinion uses this sentence to support its
subsequent declaration, \That is enough to place the challenged portions
of PURPA within Congress' power under the Commerce Clause."
15
While
this does not provide conclusive evidence that the wording in the brief led
to the phrasing in the opinion, this along with other phrases that articulate
viewpoints that support the SG's position in the case lend support to the
proposition that the Justices and/or clerks read through and analyzed the
brief, found the brief persuasive, and potentially that they used the brief's
language in the opinion itself (Garrow 1998).
4.3 The Cases
There are shared features of the cases that are worth note. The similarities in
the set of cases provide insight into the factors or mechanisms that play a role
in leading to extremely high levels of language overlap between an opinion
and a merits brief. Still, in this section the factors linking cases merely show
correlations that may play a substantial role in generating a deeper causal
understanding of the role briefs play in in
uencing opinion language. This
Dissertation shows that the Court tends to share more or less language with
briefs based on specic factors such as the presence of the Solicitor General
as a counsel of record. There is further support from previous works that
highlight the Solicitor General's success before the Supreme Court from
multiple dimensions including shared language with the Court's opinions
(Black and Owens 2012; Corley 2008).
13
Id. at 758
14
Brief for Appellants at 24
15
456 U.S. at 751 (1982)
98
I use the same dataset for the analyses in this Chapter as I used in the
previous Chapter. Along with the percentage of overlapping language for
each case, I also aggregated the number of overlapping words in each brief-
opinion pair. Upon analysis I found one subset of cases where there is an
especially high probability that merits briefs will strongly impact Supreme
Court opinion content. Figure 4.1 presents the typical impact of merits
briefs on opinions depending on the combination of the percentage of the
opinion that shares language with the brief and the number of words in the
opinion that overlap with words from the brief (both calculations are from
the same overlapping language).
Figure 4.1: Typology of Impact of Brief on Opinion Language
Later in the Chapter I present examples from all four types of relationships.
The Lifted relationship is the one of primary interest. It presents some of
the strongest examples that specic briefs made meaningful contributions
to the language in the Court's opinions and that the Court used substantive
language from the briefs in the opinions. To bracket the cases, I set a
threshold of 33% overlap between opinion and brief as the point of distinction
between high and low and 1,000 shared words between opinion and brief as
the demarcation between high and low word overlap. There are 22 instances
99
of Type I relationships in the dataset. Although there are a few exceptions,
most of these opinions use language from a merits briefs in the cases as
templates for the wording in the opinion. This often includes the Court
usurping the argument from the brief. In Lifted opinions in particular the
percentage overlap and overlapping word metrics do not provide the entire
picture. As the examples below show, instances of language overlap in these
cases also point to sections in the opinion which almost assuredly derive from
language in the brief but due to the software's capabilities do not come up
as overlapping matches.
The last case that ts the Lifted criteria is from 1993. This implies that
the relationship between specic merits briefs and the Court's opinions has
evolved in the last several decades away from Lifted relationships. This is
underscored by the fact that the length of opinions has generally grown in
recent years (Black and Spriggs 2008). The Lifted opinion from 1993 was
authored by the Justice with the majority of Lifted opinions: Justice Black-
mun. As I show below, Justice Blackmun was the majority opinion writer
for 12 of the 23 total Lifted opinions and each of these Blackmun opin-
ions shares a substantial amount of language with briefs from the Solicitor
General.
Finally there is a strong distinction in the cases between the amount of lan-
guage each Lifted opinion shares with the non-highly similar merits brief,
when compared to the highly-similar merits brief. This is evident from the
fact that no Lifted opinion shares the threshold number of words and per-
centage overlap with both briefs in the same case. The high level of dierence
between the amounts of language the opinions share with one merits brief
relative to the other is telling in that it is not merely uncontested statements
and quotations that form the basis of the shared language. Although it is
beyond the scope of this Chapter to dissect the psychology of the Justices
in an attempt to uncover whether the briefs cause the Justices to utilize the
same language or if the Justices simply share the opinion of the brief writer,
100
the examples of Lifted relationships show the extent of the role briefs can
play in opinion construction.
4.3.1 Common Source Relationships (High Percentage Over-
lap, Low Words Overlap)
First I examine Common Source relationships, relationships of Shared Un-
derstanding and Traditional Impact brief-opinion relationships and then jux-
tapose these with the case type of most interest: the Lifted. To begin, the
case with the highest percentage overlap between a brief and opinion in the
dataset is Ashton v. Kentucky.
16
The case concerns a libel law in the state
of Kentucky. This opinion authored by Justice Douglas shares 59% of its
language with the brief for the Petitioner which equates to 850 words. The
total opinion is 1,430 words. This is a Type III opinion because although
the percentage overlap is high, the number of overlapping words is below
the 1,000 word threshold. The opinion in this case is typical of a Type III
as the overlapping language is primarily based on shared citations. At one
point in this opinion, for example, 434 continuously shared words or more
than 50% of the shared language between the brief and opinion derive from
a pamphlet, the source of the libel claim, lengthily quoted in both.
Aside from this long shared citation, the Court and the petitioner's brief
also rely on many of the same cases to support their positions. The Court
analogizes the nature of the Kentucky case to a Connecticut statute in a
related case Cantwell v. Connecticut.
17
The Ashton opinion and the pe-
titioner's brief both convey that in Cantwell the Court stated that, \[the]
oense known as breach of the peace embraces a great variety of conduct
destroying or menacing public order and tranquility. It includes not only
16
384 U.S. 195 (1966)
17
310 U.S. 296 (1940)
101
violent acts but acts and words likely to produce violence in others."
18
The
Ashton opinion and the petitioner's brief both quote Cantwell for what was
problematic in the law, \Here we have a situation analogous to a convic-
tion under a statute sweeping in a great variety of conduct under a general
and indenite characterization, and leaving to the executive and judicial
branches too wide a discretion..."
19
As this analogy is not in the State's
favor it is excluded from Kentucky's brief in the case. Comparisons in such
cases may intimate that a party thought along the same lines as the Justices
and clerks involved in drafting the opinion although it sheds little light on
whether the brief made a signicant impact on the opinion writer(s).
Common Source opinions can, however, mirror more closely to Type I opin-
ions although they generally still do not to display the same overall similitude
between brief and opinion as with Type I instances. An example of this kind
of relationship can be found with Bulova Watch v. United States.
20
Justice
Whittaker's opinion in Bulova shares 48% of its language with the SG's brief
or 861 words from the opinion. The brief in this instance was written by
Solicitor General Archibald Cox and the SG's presence is often an indicator
that the Court will rely on language in the brief; especially when the views
expressed in it accord with the Justices' positions on the issues. In one in-
stance the opinion clearly relies on the SG's brief (the respondent's brief)
for the depiction of the petitioner's argument:
18
384 U.S. at 199; 310 U.S. at 308.
19
Id.; See also 310 U.S. At 308.
20
365 U.S. 753 (1960)
102
Even if petitioner were correct in
stating that Section 2411(a) is to be
regarded as the later enactment, it
would not follow that it would take
precedence, for it has been frequently
held that a specic statute over-rides
a general statute without regard to
priority of enactment. Townsend v.
Little, 109 U.S. 504, 512. More
than that, however, the premise of
the argument is erroneous. At the
time Section 3771(e) was enacted (in
1942), a predecessor provision of Sec-
tion 2411(a) had long been on the
books. Save for one word of no possi-
ble signicance, this predecessor pro-
vision (Section 177(b) of the Judicial
Code, 28 U.S.C., 1940 ed., Section
284(b)) was identical in terms with
the present Section 2411(a).
Petitioner further contends that
x2411(a) is a later enactment than
x3771(e) and, for that reason, should
take precedence over it. We do not
believe that x2411(a) can fairly be
regarded as a later enactment than
x3771(e), for at the time x3771(e)
was enacted, in 1942, a predecessor
provision of x2411(a) had long been
on the books. Save for the word
hereby - of no possible signicance
that predecessor provision (x177(b)
of the Judicial Code, 28 U.S.C. (1940
ed.) x284(b) was identical with the
present x2411(a). But even if pe-
titioner were correct in concluding
that x2411(a) is to be regarded as the
later enactment, it would not neces-
sarily take precedence over x3711(e),
for it is familiar law that a specic
statute controls over a general one
`without regard to priority of enact-
ment.' Townsend v. Little, 109 U.S.
504, 512.
21
In the above example, the Court uses the same reasoning as the SG to
explain both why the petitioner was incorrect in that instance as well as
why x2411(a) is not necessarily a later enactment than x3771(e).
In another portion of the opinion, the Court shares language with the SG's
brief in interpreting Congress' intention with x3771(e). Here they both rely
on the same report from the Senate Finance Committee:
21
Id. at 758; Brief for the United States at 10, Bulova Watch v. United States, 365
U.S. 753 (No. 241) (1960).
103
From this statement, it is apparent
that Congress proposed (1) to deny
interest up to the date that a carry-
back could be determined and (2) to
prevent, through delay in the presen-
tation of claims, the accumulation of
interest after that date.
This surely shows Congress' purpose
to deny interest on carry-back re-
funds for any period prior to the time
they could be determined, and also to
prevent, through delay in the presen-
tation of claims, the accumulation of
interest after that date and prior to
the ling of the claim.
22
There are several other instances of extensive language sharing between
the SG's brief and the Court's opinion in Bulova Watch,
23
mainly dealing
with the interpretation of relevant precedent. With its instances of shared
substantive language between brief and opinion, Bulova Watch presents an
example of how a Type III opinions may share Type I opinion characteristics
on a smaller scale.
4.3.2 Shared Understanding (Low Percentage Overlap, High
Words Overlap) and Typical Impact (Low Percentage
Overlap, Low Words Overlap) Relationships
Shared Understanding opinions are more frequent occurrences as of late.
Due to lengthier opinions, we see brief-opinion relationships where the over-
all number of shared words between brief and opinion is on the higher end
of the spectrum although the percentage of overlapping language is nowhere
near the 33% threshold for the Lifted relationship. I include four cases from
the Roberts Court years as examples of Shared Understanding relationships.
These cases have the highest level of word overlap during this period and
two of the cases are on the higher end of percentage language overlap for this
period as well. The rst of these is the case Graham v. Florida,
24
with an
22
Id. at 760; Brief for the United States at 16, Bulova Watch v. United States, 365
U.S. 753 (No. 241) (1960).
23
Id.
24
130 S. Ct. 2011 (2009)
104
opinion authored by Justice Kennedy. Graham deals with an Eighth Amend-
ment analysis of the constitutionality of an imposition of a life sentence on
a juvenile. 12% of the Graham opinion overlaps with the petitioner's brief
which is equivalent to 1,327 words. The shared language between the opin-
ion and the petitioner's brief is often salient to the Court's argument but it
does not tend to originate with the petitioner's brief as they are based on
common citations. Some such instances include:
First, juveniles possess less maturity
and an underdeveloped sense of re-
sponsibility, which often results in
impetuous and ill-considered actions
and decisions. Second, juveniles are
more vulnerable and susceptible to
negative in
uences and outside pres-
sures, including peer pressure.
As compared to adults, juveniles
have a \lack of maturity and an un-
derdeveloped sense of responsibility;
they are more vulnerable or suscep-
tible to negative in
uences and out-
side pressures, including peer pres-
sure; and their characters are \not
as well formed." Id., at 569-570, 125
S. Ct. 1183, 161 L. Ed. 2d 1.
25
In fact, even expert psychologists
cannot reliably \dierentiate be-
tween the juvenile oender whose
crime re
ects unfortunate yet tran-
sient immaturity, and the rare juve-
nile oender whose crime re
ects ir-
reparable corruption." Id. at 573.
These salient characteristics mean
that `[i]t is dicult even for ex-
pert psychologists to dierentiate be-
tween the juvenile oender whose
crime re
ects unfortunate yet tran-
sient immaturity, and the rare juve-
nile oender whose crime re
ects ir-
reparable corruption.' Id., at 573.
26
In Graham, the shared language is integral to the Court's decision. It sup-
ports the notion that the imposition of such a sentence violates the Eighth
Amendment prohibition against cruel and unusual punishment which was
the Court's ultimate conclusion. Still, the overlapping language primarily
stems from a common source so it cannot be determined if the Justice and
25
Id. at 2026; Petitioner's Brief at 26, Graham v. Florida, 130 S. Ct. 2011 (08-7412)
(2009).
26
Id.; Petitioner's Brief at 49, Graham v. Florida, 130 S. Ct. 2011 (08-7412) (2009).
105
clerks utilized the language due to the brief's persuasive powers or due to a
common acknowledgement of the importance of the wording from the shared
precedent.
Similarly in Holder v. Humanitarian Law Project,
27
an opinion authored by
Chief Justice Roberts the opinion shares 12% of its language or 1,359 words
with the petitioner's brief (from the SG in this instance). As in Graham,
there are several instances of meaningful shared language between the brief
and opinion from shared sources that support the Court's ultimate conclu-
sion in the case. In this case the Court examines whether a federal statute
is unconstitutionally vague. Examples of such shared language include:
This Court has repeatedly observed
that \perfect clarity and precise
guidance have never been required
even of regulations that restrict ex-
pressive activity." Ward v. Rock
Against Racism, 491 U.S. 781, 794
(1989)
But `perfect clarity and precise guid-
ance have never been required even
of regulations that restrict expressive
activity.' Williams, supra, at 304,
128 S. Ct. 1830, 170 L. Ed. 2d
650 (quoting Ward v. Rock Against
Racism, 491 U.S. 781, 794, 109 S. Ct.
2746, 105 L. Ed. 2d 661 (1989)).
28
The statute does not prohibit inde-
pendent advocacy or expression of
any kind.
As the Government states: \The
statute does not prohibit indepen-
dent advocacy or expression of any
kind." Brief for Government.
29
In both Graham and Humanitarian Law Project, there is no question of
the source of the language. Even with the same propositions supporting
the arguments in the opinion and the petitioner's brief, there is attribution
to the proper sources. The role of the petitioner's brief in fashioning the
27
130 S. Ct. 2705 (2009)
28
Id. at 2719; Petitioner's Brief at 18, Holder v. Humanitarian Law Project, 130 S. Ct.
2705 (08-1498) (2009).
29
Id. at 2723; Petitioner's Brief at 13, Holder v. Humanitarian Law Project, 130 S. Ct.
2705 (08-1498) (2009).
106
language for the opinion is not entirely clear outside of the instance where
the opinion cites the brief.
The next case I examine is Pepper v. United States,
30
with an opinion
authored by Justice Sotomayor. The case deals with the constitutionality of
resentencing a defendant. With 1,784 overlapping words and 22% overlap
between opinion and the respondent's brief, this case ranks high on the list of
overall word overlap between brief and opinion. Like the cases before it and
as is typical for Shared Understanding cases, however, the shared language
in this case primarily revolves around shared citations to precedent. Some
examples of shared language that is meaningful to the Court's conclusion
include:
It has been a \uniform and constant"
principle of the federal sentencing
tradition that the sentencing court
will \consider every convicted person
as an individual and every case as a
unique study in the human failings
that sometimes mitigate, sometimes
magnify, the crime and the punish-
ment to ensue." Koon v. United
States, 518 U.S. 81, 113 (1996)
It has been uniform and constant
in the federal judicial tradition for
the sentencing judge to \consider ev-
ery convicted person as an individ-
ual and every case as a unique study
in the human failings that some-
times mitigate, sometimes magnify,
the crime and the punishment to en-
sue." Koon v. United States, 518
U.S. 81, 113.
31
Section 3577 permitted a sentencing
judge in determining the appropriate
punishment to \conduct an inquiry
broad in scope, largely unlimited ei-
ther as to the kind of information
he may consider, or the source from
which it may come."
Both Congress and the Sentencing
Commission thus expressly preserved
the traditional discretion of sentenc-
ing courts to `conduct an inquiry
broad in scope, largely unlimited ei-
ther as to the kind of information
[they] may consider, or the source
from which it may come.'
32
30
131 S. Ct. 1229 (2011)
31
Id. at 1239-40; Brief for the United States at 30, Pepper v. United States, 131 S. Ct.
1229 (09-6822) (2011).
32
Id. at 1240; Brief for the United States at 32, Pepper v. United States, 131 S. Ct.
1229 (09-6822) (2011).
107
Even with the high level of overlapping words in Pepper the overlapping
language is dispersed about the opinion and the sections of overlapping
language are not as dense as is typical in Lifted scenarios.
One recent Shared Understanding case in particular shares features with
Lifted cases in terms of the substance of the overlapping language. This
case, Marx v. General Revenue Corporation,
33
authored by Justice Thomas,
includes 1198 overlapping words or 25% overall overlap between opinion and
the respondent's brief. Although the majority of the shared language in
this case tracks similarly to the previously mentioned Type II examples,
two instances of shared language in Marx are quite similar to the type of
overlap that is typically particular to Type I cases. The rst is an instance
where the opinion shares language from the respondent's brief to strengthen
a legal argument:
By adding \and costs" to the second
sentence, Congress foreclosed the ar-
gument that, under expressio unius
est exclusio alterius, the expression of
costs in the rst sentence and exclu-
sion of the same term in the second
meant that defendants could recover
only attorney's fees, and not costs.
If Congress had excluded `and costs'
in the second sentence, plaintis
might have argued that the expres-
sion of costs in the rst sentence and
the exclusion of costs in the second
meant that defendants could only re-
cover attorney's fees when plaintis
bring an action in bad faith.
34
In this instance both the opinion and brief use the same tools of statutory
interpretation to gain leverage on Congress' intent for recovering costs and
attorney fees. The matching language presents a likely situation where the
argument from the brief was used as a template for that in the opinion;
something that is prevalent in Type I opinions. Comparing the brief and
opinion is instructive here to ascertain the likely source of the Justices and
33
133 S. Ct. 1166 (2012)
34
Id. at 1176; Respondent's Brief at 15-16, Marx v. General Revenue Corp., 133 S. Ct.
1166 (11-1175) (2012).
108
clerks' choice of language. The next example from Marx is something atyp-
ical in non-Type I cases. The example is an extensive passage where the
respondent's brief was almost assuredly used as a basis for the language in
the opinion. Not only is there considerable overlapping language through-
out the passage, but even the non-overlapping language expresses the same
points in the same sequence:
109
The canon against super
uity should
particularly give way when \excess
language" is \hardly unusual." Mi-
crosoft, 131 S. Ct. at 2249. A myriad
of statutes overlap with Rule 54(d).
E.g., 12 U.S.C. x2607(d)(5) (Real
Estate Settlement Procedures Act
of 1974) (\the court may award to
the prevailing party the court costs
of the action"); 12 U.S.C. x5565(b)
(Consumer Financial Protection Act
of 2010) (\the [Consumer Financial
Protection] Bureau...may recover its
costs in connection with prosecuting
such action if [it] is the prevailing
party in the action").
Second, redundancy is \hardly un-
usual" in statutes addressing costs.
See id., at... 131 S. Ct. 2238, 180
L. Ed. 2d 131. Numerous statutes
overlap with Rule 54(d)(1). See,
e.g., 12 U.S.C x2607(d)(5) ('[T]he
court may award to the prevailing
party the court costs of the action');
x5565(b) (2006 ed., Supp. V) (\the
[Consumer Financial Protection] Bu-
reau...may recover its costs in con-
nection with prosecuting such action
if [it]...is the prevailing party in the
action); 15 U.S.C x6104(d) (2006 ed.)
(`The court...may award costs of suit
and reasonable fees for attorneys and
expert witnesses to the prevailing
party').
35
The example above presents an instance where the brief provided a template
for the Court that becomes clear when examining how the opinion tracks
the lengthy phrase in the brief both in meaning and with the same specic
references. Much of the language is through shared citations but the sim-
ilarity between the passage in the brief and in the opinion make it highly
unlikely that the brief's wording did not highly in
uence the language in
the opinion.
Traditional Impact opinions look somewhat like typical Type II opinions
except with less overall shared language between brief and opinion. These
are the most common cases in the dataset. Although, they also may occa-
sionally share characteristics of Type I opinions, the shared language tends
to be sparse and so it is more dicult to decipher the extent of impact (if
any) the brief made on the opinion's content.
35
Id. at 1177-78; Respondent's Brief at 24-25, Marx v. General Revenue Corp., 133 S.
Ct. 1166 (11-1175) (2012).
110
One way we know that the brief aected the opinion's content is when the
opinion cites to the brief as in these instances from the case, Kloeckner v.
Solis:
36
Section 7702(a)(3) denes for the
most part which MSPB decisions
qualify as \judicially reviewable ac-
tion[s]," providing that \[a]ny deci-
sion of the Board under paragraph
(1) of this subsection shall be a ju-
dicially reviewable action as of" the
date of the decision...
That provision, the Government
states, \denes for the most part
which MSPB decisions qualify as
`judicially reviewable actions[s]'" by
\providing that `[a]ny decision of the
Board under paragraph (1) of this
subsection shall be a judicially re-
viewable action as of' the date of the
decision." Brief for Respondent.
37
Properly read, Section 7702(e)(1)(B)
applies only to cases over which the
Board continues to exert jurisdiction,
not to cases in which the Board has
already issued a decision. The pur-
pose of the provision is to save em-
ployees from being held in perpetual
uncertainty by Board inaction.
Another section of the statute
x7702(e)(1)(B) \ That provision, as
the Government notes, is designed
\to save employees from being held
in perpetual uncertainty by Board in-
action." Brief for Respondent.
38
In the more recent cases, we tend to see the largest impact of a brief on an
opinion when the brief is cited in the opinion. Although shared citations to
precedent are common, instances of shared language similar to the lengthy
passage in Marx are not due to the lower percentage of overlapping language
and fewer overlapping words in Traditional Impact relationships. Also with
Kloekner, as with many other cases from the last several decades, opinions
tend to attribute unoriginal language to a source internal or external to the
case itself. With Lifted cases, such attribution typically is not present.
36
133 S. Ct. 596 (2012)
37
Id. at 605; Respondent's Brief at 21, Kloeckner v. Solis, 133 S. Ct. 596 (11-184)
(2012).
38
Id. at 606; Respondent's Brief at 27-28, Kloeckner v. Solis, 133 S. Ct. 596 (11-184)
(2012).
111
4.3.3 Lifted Relationships (High Percentage Overlap and High
Word Overlap)
The briefs in Lifted cases do something more than merely persuade the
Justices to vote a certain way. These are cases where the briefs ultimately
aect the Court's arguments, assessments of the facts, and the outcome of
the law that will be applied in future cases. Table 1 below breaks each
opinion down into a facts and a law section and measures the percentage
of overlapping opinion language from each section with the merits brief of
interest. This dichotomy points to opinions where the briefs impact the
recitation of facts versus where the briefs are an aid in the construction of
law and in the Court's legal reasoning.
It should not be surprising that one of the most trusted litigants before the
Court, the S.G., was the author of all but two of the Lifted case briefs.
The examples below not only show high levels of similarity due to shared
citations, but also due to large quantities of text shared between the brief
and opinion where the opinions lacks any citation.
The chronological chain of Supreme Court decision-making strengthens the
probability that the briefs played a decisive role in the opinion language.
The briefs predate the opinions and are circulated among the Justices. The
Justices and clerks gain valuable information about the case through the
briefs. As the cases show, the voice and rhetorical devices in the language
in the opinions are suciently similar to those in the brief. It is dicult
to conceive of another source that could provide both the opinion writer
and the brief writer with the same material. The construction of the facts
and arguments in these opinions share the tone of the briefs with the highly
overlapping language leading the opinions to echo arguments made by the
brief writers. This goes beyond the neutral language we might expect to nd
from a shared source and is often instead pointed language that parallels the
arguments made in the brief.
112
There tends to be little in the way of attribution in these opinions to any
source although the similarity between the language in the briefs and opin-
ions makes it nearly impossible that the language in the opinions did not
derive from the briefs. As Table 1 shows, there is generally a large portion
of both facts and law shared between the opinion and brief. The briefs in
these cases impact the Court's reasoning as well as recitation of elements
likely derived from the case record.
Table 4.1: Typology of Impact of Brief on Opinion Language
Justice Term Case Party Facts Overlap (%) Law Overlap (%) Words Overlap % Overlap SG Brief
Warren 1955 Steiner Resp 71 14 1,015 53 Yes
Goldberg 1963 Tilton Pet 43 44 1,187 44 Yes
Warren 1963 Foti Resp 43 30 1,119 34 Yes
Marshall 1975 Train Pet 29 33 1,466 32 Yes
Blackmun 1977 MacDonald Pet 30 46 1,395 42 Yes
Blackmun 1979 Lewis Resp 35 38 1,036 40 Yes
Blackmun 1980 DiFrancesco Pet 41 28 1,921 33 Yes
Marshall 1980 Hodel Pet 56 37 1,675 43 Yes
Stewart 1980 Lehman Pet 22 50 1,141 43 Yes
Stewart 1980 Carter Pet 48 32 1,427 41 No
White 1980 Valencia Pet 50 30 1,153 35 No
Blackmun 1982 New Banner Pet 55 38 1,975 43 Yes
Blackmun 1983 Russello Resp 33 55 1821 51 Yes
Burger 1983 89 Firearms Pet 28 38 1,178 38 Yes
Blackmun 1984 Nat'l Bank Pet 43 51 2,216 50 Yes
Blackmun 1985 Hughes Pet 37 28 1,108 33 Yes
Powell 1986 Yuckert Pet 39 49 1,446 43 Yes
Blackmun 1987 Egan Pet 60 47 1,955 52 Yes
Blackmun 1989 Doe Agency Pet 44 38 1,097 42 Yes
Blackmun 1989 U-Penn Resp 53 25 1,555 33 Yes
Blackmun 1992 Keystone Pet 39 37 1,010 39 Yes
Blackmun 1993 Posters N' Things Resp 54 38 1,178 41 Yes
As the Table shows, only two cases fall below 20% overlap in the area of the
opinion's legal reasoning. This indicates that in these instances the Court
typically tends to rely on the briefs' presentations of both law and facts.
To examine each case in this list in detail would require hundreds of pages.
In this Chapter, I attempt to distill the most salient and relevant shared
passages between the briefs and opinions in these cases.
While comparing phrases between briefs and opinions, often even the non-
overlapping language in the opinions is so closely related to the language in
the briefs that it would be dicult to make a compelling argument that the
author of the opinion did not take the language and argument in the brief
and exchange a few words for their synonyms in the process.
113
Justice Blackmun, more so than any other Supreme Court Justice, shows
a high level of willingness to engage the SG's arguments and to parallel
the SG's reasoning and language throughout his opinions. This relationship
is underscored by its continuity through Justice Blackmun's career on the
Court. While other Justices authored two Lifted opinions during the period
of this dataset, Justice Blackmun authored twelve from 1977 through 1993.
Due to the high percentage of Lifted opinions authored by Justice Blackmun,
the Lifted cases are divided in those he authored and those authored by other
Justices.
4.3.3.1 Blackmun Cases
The rst Lifted case authored by Justice Blackmun is United States v. Mac-
Donald.
39
The Solicitor General who was counsel of record on the peti-
tioner's brief in this case was Wade McCree. As Table 1 shows, almost
50% of the discussion of the law in the opinion overlaps with the SG's brief.
The Court in MacDonald examines the time limit to appeal a federal dis-
trict court's order denying a motion to dismiss an indictment because of an
alleged violation of a Sixth Amendment right to a speedy trial. The opin-
ion clearly follows the petitioner's brief for the assessment of the applicable
case-law as is evident below:
This Court has twice departed in
criminal cases from the general pro-
hibition against piecemeal appellate
review, invoking on both occasions
the so-called \collateral order" ex-
ception to the nal judgment rule,
rst announced in Cohen.
This Court in criminal cases has
twice departed from the general pro-
hibition against piecemeal appellate
review... In each instance, the Court
relied on the nal-judgment rule's
\collateral order" exception articu-
lated in Cohen..
40
39
435 U.S. 850 (1977)
40
Id. at 854; Brief for the United States at 27, United States v. MacDonald, 435 U.S.
850 (75-1892) (1977).
114
The opinion continues to follow the reasoning in the petitioner's brief in the
following two passages:
Like the denial of a motion to dis-
miss an indictment on double jeop-
ardy grounds, a pretrial order reject-
ing a defendant's speedy trial claim
plainly `lacks the nality tradition-
ally considered indispensable to ap-
pellate review' (Abney v. United
States, supra, slip op. 7). Hence, if
such orders may be appealed prior to
trial, it must be because they satisfy
the restrictive qualications identi-
ed in Cohen and Abney as sucient
to justify dispensing with the normal
rules against piecemeal review before
nal judgment. The same cannot be
said of the denial of a pretrial motion
to dismiss an indictment on speedy
trial grounds.
Like the denial of a motion to dis-
miss an indictment on double jeop-
ardy grounds, a pretrial order reject-
ing a defendant's speedy trial claim
plainly `lacks the nality tradition-
ally considered indispensable to ap-
pellate review,' Abney v. United
States, 431 U.S., at 659, 97 S.Ct.,
at 2040, that is, such an order obvi-
ously is not nal in the sense of ter-
minating the criminal proceedings in
the trial court. Thus, if such an or-
der may be appealed before trial, it is
because it satises the criteria identi-
ed in Cohen and Abney as sucient
to warrant suspension of the estab-
lished rules against piecemeal review
before nal judgment. We believe it
clear that an order denying a motion
to dismiss an indictment on speedy
trial grounds does not satisfy those
criteria.
41
41
Id. at 857; Brief for the United States at 30-31, United States v. MacDonald, 435
U.S. 850 (75-1892) (1977).
115
We are aware of only two other fed-
eral cases in which a defendant has
sought pretrial appellate review of an
order denying his motion to dismiss
the indictment because of an alleged
violation of the Sixth Amendment
right to a speedy trial, and in both
instances the court of appeals held
that it lacked jurisdiction to consider
the claim prior to conviction. See
United States v. Bailey, 512 F. 2d
833 (C.A. 5), certiorari dismissed,
423 U.S. 1039; Kyle v. United States,
211 F. 2d 912 (C.A. 9).
In keeping with what appear to be
the only two other federal cases in
which a defendant has sought pretrial
review of an order denying his motion
to dismiss an indictment on speedy
trial grounds, we hold that the Court
of Appeals lacked jurisdiction to en-
tertain respondent's speedy trial ap-
peal. United States v. Bailey, 512
F.2d 833 (CA5), cert. dism'd, 423
U.S. 1039, 96 S.Ct. 578, 46 L.Ed.2d
415 (1975); Kyle v. United States,
211 F.2d 912 (CA9 1954).
42
The opinion's reasoning parallels that in the petitioner's brief throughout
and the opinion's conclusion is no dierent:
Even were this conclusion less clear,
the important policy considerations
that underlie both the Speedy Trial
Clause and the statutory bar to
piecemeal appeals in criminal cases
strongly suggest that speedy trial
motions are the least appropriate
subject for interlocutory appellate
review.
Our conclusion, however, is rein-
forced by the important policy con-
siderations that underlie both the
Speedy Trial Clause and 28 U.S.C.
x1291.
43
42
Id.; Brief for the United States at 23-24, United States v. MacDonald, 435 U.S. 850
(75-1892) (1977).
43
Id. at 861; Brief for the United States at 43, United States v. MacDonald, 435 U.S.
850 (75-1892) (1977).
116
This Court has recognized that one
of the principal reasons for its strict
adherence to the doctrine of nality
in criminal cases is that `[t]he Sixth
Amendment guarantees a speedy
trial.' DiBella v. United States,
supra, 369 U.S. at 126. The Speedy
Trial Clause fosters the compelling
societal interest in the swift punish-
ment of the guilty and the prompt ex-
oneration of the innocent, an interest
that would be severely compromised
if every contested legal question aris-
ing in the course of a criminal pro-
ceeding could be resolved in a sepa-
rate appeal before trial of the general
issue.
Signicantly, this Court has empha-
sized that one of the principal rea-
sons for its strict adherence to the
doctrine of nality in criminal cases
is that `[t]he Sixth Amendment guar-
antees a speedy trial.' Di Bella v.
United States, 369 U.S., at 126, 82
S. Ct., at 658. Fulllment of this
guarantee would be impossible if ev-
ery pretrial order were appealable.
44
In Lewis v. United States,
45
Justice Blackmun's second Lifted opinion, the
Court examines the use of a defendant`s prior conviction that was obtained
without representation of counsel as evidence. Lewis is a case where the
Court uses the SG's construction of the statute in question as the basis for
the resolution of the case and for the reasoning to support it. The following
passage presents an example of such parallel argumentation:
44
Id.; Brief for the United States at 43-44, United States v. MacDonald, 435 U.S. 850
(75-1892) (1977).
45
445 U.S. 55 (1979)
117
In our view, this particular omis-
sion is especially indicative of con-
gressional intent, since other federal
statutes involving prior convictions
explicitly permit the accused to chal-
lenge the validity or constitutional-
ity of the predicate felony as a de-
fense. See Section 411 (c) (2) of the
Comprehensive Drug Abuse Preven-
tion and Control Act of 1970,
It thus stands in contrast with other
federal statutes that explicitly per-
mit a defendant to challenge, by way
of defense, the validity or constitu-
tionality of the predicate felony. See,
e. g., 18 U. S. C. / x3575 (e) (dan-
gerous special oender) and 21 U. S.
C. x851 (c)(2) (recidivism under the
Comprehensive Drug Abuse Preven-
tion and Control Act of 1970).
46
In other instances the Court relies on the SG's statutory construction:
The structure of Title IV of the Om-
nibus Act, which was enacted si-
multaneously with Title VII, rein-
forces that conclusion. Like Title
VII, Title IV prohibits various cat-
egories of presumptively dangerous
persons from transporting and re-
ceiving rearms. 18 U.S.C. 922 (g)
and 922(h)...Thus, with regard to
the statutory question at issue here,
there is no signicant dierence be-
tween title IV and Title VII. Both
statutes seek to keep rearms away
from \any person who has been con-
victed" of a felony.
The very structure of the Omnibus
Act's Title IV, enacted simultane-
ously with Title VII, reinforces this
conclusion. Each Title prohibits cat-
egories of presumptively dangerous
persons from transporting or receiv-
ing rearms. See 18 U. S. C. x922 (g)
and (h). Actually, with regard to the
statutory question at issue here, we
detect little signicant dierence be-
tween Title IV and Title VII. Each
seeks to keep a rearm away from
\any person . . . who has been con-
victed" of a felony"
47
46
Id. at 62; Brief for the United States at 17-18, 445 U.S. 55 (78-1595) (1979).
47
Id. at 64; Brief for the United States at 24-25, Lewis v. United States, 445 U.S. 55
(78-1595) (1979).
118
Congress rationally concluded that
any felony conviction {even an al-
legedly invalid one { is a sucient
basis on which to prohibit the pos-
session of rearms. See, e.g., United
States v. Samson, supra, 533 F.2d at
722; United States v. Ransom, 515
F.2d 885, 891-892 (5th Cir. 1975);
Congress could rationally conclude
that any felony conviction, even an
allegedly invalid one, is a sucient
basis on which to prohibit the posses-
sion of a rearm. See, e. g., United
States v. Ransom, 515 F.2d 885, 891-
892
48
The express congressional purpose in
enacting Title VII is set forth in the
statute itself. [T]he receipt, posses-
sion, or transportation of a rearm
by felons constitutes...a threat to the
continued and eective operation of
the Government of the United States
and of the government of each State
guaranteed by article IV of the Con-
stitution.
18 U.S.C. App. 1201. Similarly,
the legislative history of the gun con-
trol laws \evidences Congress' deep
concern about the easy availabil-
ity of rearms, especially to those
who Congress has reason to believe
pose a greater threat to community
peace than does the public gener-
ally." United States v. Liles, supra,
432 F.2d at 20. See pages 19-22,
supra. and in particular, Congress
focused on the substantial nexus be-
tween violent crimes and the posses-
sion of rearms by \any person who
has a criminal record." 114 Cong.
Rec. 13220 (1968) (remarks of Sen.
Tydings); see, e.g., id. at 16298 (re-
marks of Rep. Pollock).
Congress, as its expressed purpose in
enacting Title VII reveals, 18 U. S. C.
App. x1201, was concerned that the
receipt and possession of a rearm by
a felon constitutes a threat, among
other things, to the continued and ef-
fective operation of the Government
of the United States. The legisla-
tive history of the gun control laws
discloses Congress' worry about the
easy availability of rearms, espe-
cially to those persons who pose a
threat to community peace. And
Congress focused on the nexus be-
tween violent crime and the posses-
sion of a rearm by any person with
a criminal record. 114 Cong. Rec.
13220 (1968) (remarks of Sen. Tyd-
ings); id., at 16298 (remarks of Rep.
Pollock).
49
48
Id.; Brief for the United States at 41, Lewis v. United States, 445 U.S. 55 (78-1595)
(1979).
49
Id. at 66; Brief for the United States at 40-41, Lewis v. United States, 445 U.S. 55
(78-1595) (1979).
119
The Court also follows the SG's rationale for the rejection of petitioner's
construction of the relevant statute:
\[R]esort to an alternative construc-
tion to avoid deciding a constitu-
tional question is appropriate only
when such a course is 'fairly possible'
or when the statute provides a 'fair
alternative' construction." Swain v.
Pressley, 430 U.S. 372, 378 n.11
(1977)
With the face of the statute and
the legislative history so clear, pe-
titioner's argument that the statute
nevertheless should be construed so
as to avoid a constitutional issue is
inapposite. That course is appro-
priate only when the statute pro-
vides a fair alternative construction.
This statute could not be more plain.
Swain v. Pressley, 430 U.S. 372, 378,
and n. 11.
50
Finally the opinion adopts, nearly verbatim, the SG's assessment of the
Court's own precedent:
And this Court has repeatedly recog-
nized that a legislature may consti-
tutionally prohibit convicted felons
from engaging in activities far more
fundamental than the right to pos-
sess rearms at issue here. See
Richardson v. Ramirez, 418 U.S.
24 (1974) (disenfranchisement of
felons); DeVeau v. Braisted, 363
U.S. 144, 157-160 (1960) (felons
barred from waterfront employ-
ment); Hawker v. New York, 170
U.S. 189 (1898) [58] (prohibition on
medical practice by a felon).
This Court has recognized repeat-
edly that a legislature constitution-
ally may prohibit a convicted felon
from engaging in activities far more
fundamental than the possession of a
rearm. See Richardson v. Ramirez,
418 U.S. 24 (1974) (disenfranchise-
ment); DeVeau v. Braisted, 363
U.S. 144 (1960) (proscription against
holding oce in a waterfront labor
organization); Hawker v. New York,
170 U.S. 189 (1898) (prohibition
against the practice of medicine).
51
50
Id. at 65; Brief for the United States at 36, Lewis v. United States, 445 U.S. 55
(78-1595).
51
Id. at 66; Brief for the United States at 42, Lewis v. United States, 445 U.S. 55
(78-1595) (1979).
120
To be sure, the Court has made
clear that an outstanding uncoun-
selled felony conviction cannot reli-
ably be used for certain purposes.
See Burgett v. Texas, 389 U.S. 109
(1967);United States v. Tucker, 404
U.S. 443 (1972);Loper v. Beto, 405
U.S. 473 (1972). But the Court has
never suggested that an uncounseled
conviction is invalid for all purposes
(see, e.g., Scott v. Illinois, supra
We recognize, of course, that under
the Sixth Amendment an uncoun-
seled felony conviction cannot be
used for certain purposes. See Bur-
gett, Tucker, and Loper, all supra.
The Court, however, has never sug-
gested that an uncounseled convic-
tion is invalid for all purposes. See
Scott v. Illinois.
52
In each of those cases this Court
found that the conviction or sen-
tence in question violated the Sixth
Amendment because it depended
upon the reliability of a particular
uncounseled conviction in the past.
The federal gun laws, however, focus
on the mere fact of conviction, re-
gardless of its reliability, in order to
keep rearms away from potentially
dangerous people.
In each of those cases, this Court
found that the subsequent convic-
tion or sentence violated the Sixth
Amendment because it depended
upon the reliability of a past uncoun-
seled conviction. The federal gun
laws, however, focus not on reliabil-
ity, but on the mere fact of convic-
tion, or even indictment, in order to
keep rearms away from potentially
dangerous persons.
53
The high level of similarity between the SG's brief and the opinion in Lewis
reinforces the trust and faith that Justice Blackmun and his clerks placed in
the SG. This pattern continues through the remainder of Justice Blackmun's
Lifted opinions, which all rely heavily on the language in the SG's briefs.
DiFrancesco v. United States,
54
is yet another opinion authored by Justice
Blackmun concerning sentencing under a federal statute where the opinion
language parallels that in the SG's brief. The Court in DiFrancesco con-
sidered whether an increased sentence under the Organized Crime Control
52
Id. at 66-67; Brief for the United States at 45, Lewis v. United States, 445 U.S. 55
(78-1595) (1979).
53
Id. at 67; Brief for the United States at 48, Lewis v. United States, 445 U.S. 55
(78-1595) (1979).
54
449 U.S. 117 (1980)
121
Act of 1970 violated the Fifth Amendment's Double Jeopardy Clause. As
with Lewis, the DiFrancesco opinion utilizes the same linguistic framework
as the SG's brief to interpret the Court's precedent. This is evident when
the opinion describes the policy rationale for preventing a retrial after an
acquittal:
This rule has been characterized
as attaching \particular signicance
to an acquittal" (United States v.
Scott, supra, 437 U.S. at 91), and it
has been justied on the basis that
\[t]o permit a second trial after an
acquittal, however mistaken the ac-
quittal may have been, would present
an unacceptably high risk that the
Government, with its vastly superior
resources, might wear down the de-
fendant so that 'even though inno-
cent he may be found guilty.'" Ibid.
This is justied on the ground
that, however mistaken the acquit-
tal may have been, there would be
an unacceptably high risk that the
Government, with its superior re-
sources, would wear down a defen-
dant, thereby \enhancing the pos-
sibility that even though innocent
he may be found guilty." Green v.
United States, 355 U.S., at 188.
55
The decisions of this Court in the
sentencing area have also established
that a sentence does not have quali-
ties of constitutional nality compa-
rable to an acquittal. In Bozza v.
United States, 330 U.S. 160 (1947).
This Court's decisions in the sentenc-
ing area clearly establish that a sen-
tence does not have the qualities of
constitutional nality that attend an
acquittal. In Bozza v. United States,
330 U.S. 160 (1947)
56
55
Id. at 130; Brief for the United States at 21, DiFrancesco v. United States, 449 U.S.
117 (79-567) (1980).
56
Id. at 134; Brief for the United States at 28, DiFrancesco v. United States, 449 U.S.
117 (79-567) (1980).
122
The multiple punishment guarantee
that has evolved in the holdings of
this Court, apart from the Benz
dictum, clearly is not involved in
this case. That guarantee concerns
the amount of punishment that law-
fully can be imposed on a defendant.
As in Ex Parte Lange, a defendant
may not receive a higher sentence
than that authorized by the legis-
lature...Clearly, no double jeopardy
problem would have been presented
in Ex Parte Lange if Congress had
established that the oense was pun-
ishable by ne and imprisonment,
even though those are multiple pun-
ishments. See Whalen v. United
States, supra, slip op. 4. 20 There
is no question what punishment was
authorized by Congress under 18
U.S.C. 3575 and 18 U.S.C. 3576, or
that Congress had the power to es-
tablish that punishment. Accord-
ingly, 18 U.S.C. 3576 does not vio-
late the guarantee against multiple
punishment that is enunciated in Ex
parte Lange.
The guarantee against multiple pun-
ishment that has evolved in the
holdings of this Court plainly is
not involved in this case. As Ex
parte Lange demonstrates, a defen-
dant may not receive a greater sen-
tence than the legislature has au-
thorized. No double jeopardy prob-
lem would have been presented in
Ex parte Lange if Congress had pro-
vided that the oense there was pun-
ishable by both ne and imprison-
ment, even though that is multiple
punishment. See Whalen v. United
States, 445 U.S., at 688-689; id., at
697-698 (concurring opinion). The
punishment authorized by Congress
underxx3575 and 3576 is clear and
specic and, accordingly, does not vi-
olate the guarantee against multiple
punishment expounded by Ex parte
Lange.
57
The opinion also tracks the reasoning and language in the SG's brief to
explain its decision, namely that the Double Jeopardy Clause is not violated
by allowing the government to appeal a sentence under the statute:
57
Id. at 139; Brief for the United States at 46-47, DiFrancesco v. United States, 449
U.S. 117 (79-567) (1980).
123
Since it is not a prosecution ap-
peal itself that can fall afoul of the
Double Jeopardy Clause, but rather
the relief requested by the appeal, it
must next be considered whether a
criminal sentence, once pronounced,
must be accorded constitutional -
nality similar to that attaching to a
jury's verdict of not guilty. Neither
the history of sentencing practices,
the pertinent holdings of this Court,
nor considerations of double jeop-
ardy policy supports such an equa-
tion.
The double jeopardy focus, thus, is
not on the appeal but on the re-
lief that is requested, and our task
is to determine whether a criminal
sentence, once pronounced, is to be
accorded constitutional nality and
conclusiveness similar to that which
attaches to a jury's verdict of acquit-
tal. We conclude that neither the his-
tory of sentencing practices, nor the
pertinent rulings of this Court, nor
even considerations of double jeop-
ardy policy support such an equa-
tion.
58
Thus, appeal of a sentence would
seem to be a violation of double jeop-
ardy only if the original pronounce-
ment of sentence is to be treated in
the same way as an acquittal and the
appeal is to be treated as a retrial.
Essentially, the court of appeals' the-
ory is that the imposition of a sen-
tence should be treated, for double
jeopardy nality purposes, as an \im-
plied acquittal" of a greater sentence.
See Van Alstyne, In Gideon's Wake:
Harsher Penalties and the \Success-
ful" Criminal Appellant, 74 Yale L.
J. 606, 634-635 (1965).
Appeal of a sentence, therefore,
would seem to be a violation of dou-
ble jeopardy only if the original sen-
tence, as pronounced, is to be treated
in the same way as an acquittal
is treated and the appeal is to be
treated in the same way as a re-
trial. Put another way, the argument
would be that, for double jeopardy
nality purposes, the imposition of
the sentence is an \implied acquit-
tal" of any greater sentence. See Van
Alstyne, In Gideon's Wake: Harsher
Penalties and the \Successful" Crim-
inal Appellant, 74 Yale L. J. 606,
634-635 (1965).
59
In a later section, the language in the opinion mirrors the certainty in the
SG's brief regarding the common law tradition of allowing such an appeal:
58
Id. at 132; Brief for the United States at 20, DiFrancesco v. United States, 449 U.S.
117 (79-567) (1980).
59
Id. at 133; Brief for the United States at 21-22, DiFrancesco v. United States, 449
U.S. 117 (79-567) (1980).
124
While there is little American ex-
perience with appellate review of
sentences, this history demonstrates
that the common law has never as-
cribed such nality to a sentence as
would prevent a legislature from au-
thorizing its appeal by the prosecu-
tion. Indeed, several countries that
trace their legal systems to the En-
glish common law permit such ap-
peals. See Canada: Can. Rev. Stat.
xx605 (1)(b) and 748(b)(ii) (Martin's
Annual Criminal Code (1979)); Aus-
tralia (New South Wales): Crimi-
nal Appeal Act of 1912, as amended
in 1924, 3 Pub. Acts N.S.W.x5D
(1959); New Zealand: Crimes Act
1961, as amended by the Crimes
Amendment Act of 1966, 1 Repr.
Stat. N.Z. x383 (a) (1979).
Thus it may be said with certainty
that history demonstrates that the
common law never ascribed such -
nality to a sentence as would pre-
vent a legislative body from autho-
rizing its appeal by the prosecution.
Indeed, countries that trace their
legal systems to the English com-
mon law permit such appeals. See
Can. Rev. Stat. xx605 (1)(b)
and 748 (b)(ii) (1970), Martin's An-
nual Criminal Code 523, 636 (E.
Greenspan ed. 1979); New Zealand
Crimes Act 1961, as amended by the
Crimes Amendment Act of 1966, 1
Repr. Stat. N. Z. x383 (2) (1979).
60
Similar to Lewis, Justice Blackmun's next Type I opinion Dickerson v. New
Banner Institute,
61
deals with sentencing under a federal rearms statute.
The Court actually adopts the SG's interpretation of the precedent from
Lewis and why the Court should dierentiate its reasoning in this case:
60
Id. at 134; Brief for the United States at 27, DiFrancesco v. United States, 449 U.S.
117 (79-567) (1980).
61
460 U.S. 103 (1983)
125
In Lewis, this Court recognized an
obvious, \commonsense" exception
to the statutory language for per-
sons whose convictions have been re-
versed or vacated on appeal or on
collateral attack (445 U.S. at 60-61
n.5), but that exception is not ap-
plicable to persons whose convictions
have been expunged...A conviction
that has been expunged, however,
stands on an entirely dierent foot-
ing. Expunction does not call into
question the legality of the previous
conviction, and it does not signify
that the defendant was innocent of
the crime for which he was convicted.
As explained below [28] (pages 30-
35, infra), expunction merely means
that the responsible jurisdiction has
decided not to accord the conviction
certain continuing eects.
In Lewis, it is true, we recognized an
obvious exception to the literal lan-
guage of the statute for one whose
predicate conviction had been va-
cated or reversed on direct appeal.
445 U.S., at 61, n. 5; see Note,
Prior Convictions and the Gun Con-
trol Act of 1968, 76 Colum. L. Rev.
326, 334, n. 42 (1976). But, in
contrast, expunction does not alter
the legality of the previous convic-
tion and does not signify that the de-
fendant was innocent of the crime to
which he pleaded guilty. Expunction
in Iowa means no more than that the
State has provided a means for the
trial court not to accord a conviction
certain continuing eects under state
law.
62
There seems little doubt that
rearms disabilities may constitu-
tionally be attached to an expunged
conviction (see Lewis v. United
States, supra, 445 U.S. at 65-68),
and an exception for such convic-
tions, unlike convictions reversed or
vacated due to legal error, is far from
obvious.
Clearly, rearms disabilities may be
attached constitutionally to an ex-
punged conviction, see Lewis v.
United States, 445 U.S., at 65-68,
and an exception for such a convic-
tion, unlike one reversed or vacated
due to trial error, is far from obvi-
ous.
63
Much of the shared reasoning in this case involves the parallel interpreta-
tions of the statute at issue and so the majority of the overlapping language
focuses on Congress' intent with the crime bill in question. There is a bevy
62
Id. at 115; Petitioner's Brief at 19; 27-28, Dickerson v. New Banner Institute, 460
U.S. 103 (81-1180) (1983).
63
Id.
126
of examples of this relationship in this case. Below is one such extensive in-
stance where the Court adopts the SG's interpretation of Congress' purpose
and in the process, much of the brief's original language:
We have found nothing in the leg-
islative history of the Gun Control
Act or related federal rearms laws
that even faintly suggests that state
expunctions were intended automat-
ically to remove the disabilities im-
posed by 18 U.S.C. x922(g)(1) and
(h)(1), 12 and neither the court be-
low nor respondent has cited any
such proof. This lack of evidence is
highly signicant for several reasons.
First, the purpose of the Gun Control
Act will be frustrated by the decision
of the court of appeals. That decision
would require the Secretary to grant
dealer and manufacturer licenses to
organizations directed by individuals
convicted of serious criminal oenses
(or to such individuals themselves)
whenever the conviction in question
has been expunged under state law.
This would result even though state
expunctions typically do not focus
upon the question with which the
Gun Control Act is concerned, i.e.,
whether the convicted person is t
to engage in the rearms business
or to possess, ship, transport, or re-
ceive rearms. See pages 28-29, 32-
33 notes 30, 31, infra.
Although we have searched dili-
gently, we have found nothing in
the legislative history of Title IV
or related federal rearms statutes
that suggests, even remotely, that a
state expunction was intended auto-
matically to remove the disabilities
imposed by xx922(g)(1) and (h)(1).
See, e. g., S. Rep. No. 1501, 90th
Cong., 2d Sess. (1968); S. Rep. No.
1097, 90th Cong., 2d Sess. (1968);
H. R. Rep. No. 1577, 90th Cong., 2d
Sess. (1968); H. R. Conf. Rep. No.
1956, 90th Cong., 2d Sess. (1968);
H. R. Rep. No. 488, 90th Cong., 1st
Sess. (1967). This lack of evidence is
signicant for several reasons. First,
the purpose of the statute would be
frustrated by a ruling that gave ef-
fect to state expunctions; a state
expunction typically does not focus
upon the question with which Title
IV is concerned, namely, whether the
convicted person is t to engage in
the rearms business or to possess a
rearm.
64
64
Id. at 119; Petitioner's Brief at 47-48, Dickerson v. New Banner Institute, 460 U.S.
103 (81-1180) (1983).
127
Second, the lack of any proof that
Congress intended that state expunc-
tions be credited is signicant be-
cause \`[i]n the absence of a plain
indication to the contrary it is to
be assumed when Congress enacts a
statute that it does not intend to
make its application dependent on
state law.'" NLRB v. Natural Gas
Utility District, 402 U.S. 600, 603
(1971), quoting NLRB v. Randolph
Electric Membership Corp., 343 F.2d
60 (4th Cir. 1965). Accord, United
States v. Nardello, 393 U.S. 286,
289-296 (1969). In Jerome v. United
States, 318 U.S. 101, 104 (1943), the
Court explained the reason for this
rule: \That assumption is based on
the fact that the application of fed-
eral legislation is nationwide and at
times on the fact that the federal pro-
gram would be impaired if state law
were to control" (citations omitted).
We explain below (see pages 39-41,
infra) how application of state law
in the present context would impair
the federal gun control scheme. Fur-
thermore, the legislative history re-
veals that Congress believed a uni-
form national program was necessary
to assist in curbing the illegal use of
rearms.
Second, \`[in] the absence of a plain
indication to the contrary... it is
to be assumed when Congress en-
acts a statute that it does not intend
to make its application dependent
on state law.'" NLRB v. Natural
Gas Utility Dist., 402 U.S. 600, 603
(1971), quoting NLRB v. Randolph
Electric Membership Corp., 343 F.2d
60, 62-63 (CA4 1965). This is be-
cause the application of federal leg-
islation is nationwide and at times
the federal program would be im-
paired if state law were to control.
Jerome v. United States, 318 U.S.
101, 104 (1943). The legislative his-
tory reveals that Congress believed
a uniform national program was nec-
essary to assist in curbing the ille-
gal use of rearms. See S. Rep. No.
1097, 90th Cong., 2d Sess., 28, 76-
77 (1968).Third, Title IV \is a care-
fully constructed package of gun con-
trol legislation.' 'Congress knew the
signicance and meaning of the lan-
guage it employed.'"
65
65
Id. at 119-20; Petitioner's Brief at 48-49, Dickerson v. New Banner Institute, 460
U.S. 103 (81-1180) (1983).
128
As noted above, the Gun Control Act
\is a carefully constructed package
of gun control legislation," (Scarbor-
ough v. United States, supra, 431
U.S. at 570) and in framing its terms
Congress took pains to avoid the very
sort of problems that the decision be-
low will inevitably create. The pro-
visions of the Act demonstrate that
Congress endeavored to prevent any
uncertainty concerning those persons
subject to disabilities by virtue of
prior convictions. Congress used un-
ambiguous language in attaching gun
control disabilities to \any person
who has been convicted" of a qualify-
ing oense (18 U.S.C. 922(g)(1) and
(h)(1)).
Congress, in framing it, took pains
to avoid the very problems that the
Court of Appeals' decision inevitably
would create, such as individualized
federal treatment of every expunc-
tion law. Congress used unambigu-
ous language in attaching gun con-
trol disabilities to any person \who
has been convicted" of a qualifying
oense.
66
The Court also tracks the SG's reasoning almost identically regarding similar
state statutes:
66
Id. at 122a; Petitioner's Brief at 67, Dickerson v. New Banner Institute, 460 U.S.
103 (81-1180) (1983).
129
More than half the states have en-
acted one or more laws that may
be broadly classied as expunction
statutes, i.e., statutes that attempt
to conceal prior convictions or to re-
move some of the convictions' col-
lateral or residual eects. The var-
ious statutes dier, however, in al-
most every particular. While some
are applicable only to young oend-
ers, others may be invoked by adults.
Some are available only to persons
convicted of certain oenses, but oth-
ers permit the expunction of a con-
viction for any crime, including mur-
der. Some are conned to rst of-
fenders, but others permit relief to
recidivists. Some apply only to per-
sons given certain sentences or sen-
tenced under certain laws, while oth-
ers contain no such restriction. Some
but not all require a waiting period
following conviction. Some are dis-
cretionary, while others provide for
automatic expunction under certain
circumstances.
Over half the States have enacted one
or more statutes that may be clas-
sied as expunction provisions that
attempt to conceal prior convictions
or to remove some of their collateral
or residual eects. These statutes
dier, however, in almost every par-
ticular. Some are applicable only
to young oenders, e. g., Mich.
Comp. Laws x780.621 and .622
(1982). Some are available only to
persons convicted of certain oenses,
e. g., N. J. Stat. Ann. x2C:52-2(b)
(West 1982); others, however, per-
mit expunction of a conviction for
any crime including murder, e. g.,
Mass. Gen. Laws Ann., ch. 276,
x100A (West Supp. 1982-1983).Some
are conned to rst oenders, e. g.,
Okla. Stat., Tit. 22, x991c (Supp.
1982-1983). Some are discretionary,
e. g., Minn. Stat. x638.02(2) (Supp.
1982), while others provide for auto-
matic expunction under certain cir-
cumstances, e. g., Ariz. Rev. Stat.
Ann. x13-912 (1978).
67
67
Id. at 121; Petitioner's Brief at 54-55, Dickerson v. New Banner Institute, 460 U.S.
103 (81-1180) (1983).
130
The statutes also vary widely in the
language employed to describe what
they are supposed to do. Various
statutes are said to \expunge" the
conviction, guilty verdict, or guilty
plea; \seal" the le or record; \limit
access" to the convicted person's
\criminal history"...Thus, only a mi-
nority of the statutes address the fol-
lowing questions: whether the ex-
punged conviction may be consid-
ered in sentencing for a subsequent
oense, in setting bail on subse-
quent charges, or in parole proceed-
ings; whether the expunged convic-
tion may be used to impeach testi-
mony; what is to be done with the
records pertaining to the conviction;
whether any eort is to be made to
retrieve records previously furnished
to other jurisdictions; and whether,
and if so under what circumstances,
the convicted person may deny the
fact of conviction.
The statutes vary in the language
employed to describe what they do.
Some speak of expunging the convic-
tion, others of \sealing" the le or of
causing the dismissal of the charge.
The statutes also dier in their ac-
tual eect. Some are absolute; oth-
ers are limited. Only a minority ad-
dress questions such as whether the
expunged conviction may be consid-
ered in sentencing for a subsequent
oense or in setting bail on a later
charge, or whether the expunged con-
viction may be used for impeachment
purposes, or whether the convict may
deny the fact of his conviction.
68
By sharing the SG's reasoning in Dickerson, the Court buttresses its decision
from multiple angles. In doing so and with the high level of language overlap
with the SG's brief, the respondent's arguments have very little in
uence on
the Court's opinion language.
This pattern continues in Russello v. United States.
69
Russello is another
case with an opinion authored by Justice Blackmun concerning the inter-
pretation of a federal crime bill. The opinion frames the facts in the same
manner as laid out by the SG in a non-favorable light to the defendant. The
68
Id.; Petitioner's Brief at 56-57, Dickerson v. New Banner Institute, 460 U.S. 103
(81-1180) (1983).
69
464 U.S. 16 (1983)
131
Court, for example, adopts the SG's strong language in dening the am-
biguous term \interest," which is consequential in interpreting the meaning
of the statute in question:
In sum, whatever else is denoted by
the term \interest," it undoubtedly
comprehends all forms of real and
personal property, including prof-
its and proceeds. This Court has
repeatedly relied upon the term
\interest" in dening the meaning
of \property" in the Due Process
Clause. For example in Perry v. Sin-
dermann, 408 U.S. 593, 601 (1972)...
It is thus apparent that the term
\interest" comprehends all forms of
real and personal property, includ-
ing prots and proceeds. This Court
repeatedly has relied upon the term
\interest" in dening the meaning
of \property" in the Due Process
Clause of the Fourteenth Amend-
ment of the Constitution. See Perry
v. Sindermann, 408 U.S. 593, 601
(1972).
70
It was undoubtedly because Congress
did not want the RICO forfeiture
provision to be limited by \rigid,
technical" ( Perry v. Sindermann,
supra, 408 U.S. at 601) denitions
drawn from other areas of law that it
selected the broad term \interest" to
describe those things subject to for-
feiture under Section 1963(a)(1)....
Congress therefore selected the term
\interest." This choice of language
was fully consistent with the pattern
of the RICO statute
It undoubtedly was because Congress
did not wish the forfeiture provision
of Sec. 1963(a) to be limited by rigid
and technical denitions drawn from
other areas of the law that it se-
lected the broad term \interest" to
describe those things that are sub-
ject to forfeiture under the statute.
Congress selected this general term
apparently because it was fully con-
sistent with the pattern of the RICO
statute in utilizing terms and con-
cepts of breadth. Among these are
\enterprise" in Sec. 1961(4); \racke-
teering [22] activity" in Sec. 1961(1)
(1982 ed.); and \participate" in Sec.
1962(c).
71
70
Id. at 21; Brief for the United States at 26-27, Russello v. United States, 464 U.S.
16 (82-472) (1983).
71
Id. at 21-22; Brief for the United States at 28-29, Russello v. United States, 464 U.S.
16 (82-472) (1983).
132
If Congress had intended to restrict
subsection (a)(1) to interests in an
enterprise, as petitioner argues, it
presumably would have done so ex-
pressly, as it did in subsection (a)(2).
Had Congress intended to restrict
Sec. 1963(a)(1) to an interest in
an enterprise, it presumably would
have done so expressly as it did in
the immediately following subsection
(a)(2).
72
As is the case with many other Lifted opinions, the Court in Russello also
utilizes similar reasoning to that in the SG's brief for rejecting the opposing
party's contentions. This feature pervades the Russello opinion to a much
greater extent than in other Type I cases, as is evident in the passages
compared below:
Petitioner has not attempted to de-
ne the term \interest." 4 Petitioner
insists (Br. 9), however, that the
term does not reach prots and pro-
ceeds because \`[i]nterest', [31] by
denition, includes of necessity an in-
terest in something." Petitioner then
asserts (ibid.) that the interests in-
cluded within the scope of Section
1963(a)(1) must consequently be in-
terests in an enterprise. This argu-
ment is plainly invalid. Every prop-
erty interest, including the ownership
of or right to receive prots or pro-
ceeds, may be described as an inter-
est in something. Before the prots
of an illegal enterprise are divided,
each participant may be said to own
an \interest" in the ill-gotten gain.
After distribution, each participant
will have a possessory or ownership
interest in currency,
Petitioner himself has not attempted
to dene the term \interest" as used
in 1963(a)(1). He insists, however,
that the term does not reach money
or prots because, Sec. e says: \`In-
terest,' by denition, includes of ne-
cessity an interest in something."
Brief for Petitioner 9...We do not
agree. HN4 Every property interest,
including a right to prots or pro-
ceeds, may be described as an inter-
est in something. Before prots of
an illegal enterprise are divided, each
participant may be said to own an
\interest" in the ill-gotten gains. Af-
ter distribution, each will have a pos-
sessory interest in currency or other
items so distributed.
73
72
Id. at 23; Brief for the United States at 18, Russello v. United States, 464 U.S. 16
(82-472) (1983).
73
Id. at 22; Brief for the United States at 30-31, Russello v. United States, 464 U.S.
16 (82-472) (1983).
133
Petitioner argues (Br. 17-18) that
if the term \`interest' were as all
encompassing as suggested by the
en banc decision below, 18 U.S.C.
Sec. 1963(a)(2) would have no mean-
ing independent of 18 U.S.C. Sec.
1963(a)(1)." This argument is plainly
incorrect. Section 1963(a)(1) reaches
\any interest," whether or not in
an enterprise, provided that the in-
terest was \acquired or maintained
in violation of section 1962." Sec-
tion 1963(a)(2), on the other hand,
is restricted to interests in an enter-
prise, but the interest itself need not
have been illegally acquired or main-
tained.
It is no answer to say, as peti-
tioner does, Brief for Petitioner 17-
18, that if the term \interest" were
as all-encompassing as suggested by
the majority opinion of the Court
of Appeals, Sec. 1963(a)(2) would
have no meaning independent of Sec.
1963(a)(1), and would be mere sur-
plusage. This argument is plainly
incorrect. Subsection (a)(1) reaches
\any interest," whether or not in an
enterprise, provided it was \acquired
. . . in violation of section 1962."
Subsection (a)(2), on the other hand,
is restricted to an interest in an en-
terprise, but that interest itself need
not have been illegally acquired.
74
74
Id. at 24; Brief for the United States at 38-39, Russello v. United States, 464 U.S.
16 (82-472) (1983).
134
Petitioner also suggests (Br. 29-33)
that subsequent proposed legislation
demonstrates that the 1970 RICO
forfeiture statute excludes prots.
This conclusion is wholly unjustied.
The bills in question were introduced
to rectify marubeni and similar dis-
trict court cases. Their introduction
hardly suggests that their sponsors
viewed those decisions as correct in-
terpretations of 18 U.S.C. 1963(a)(1)
as it currently stands. 10 See United
States v. Gordon, 638 F.2d 886, 888
n.5 (5th Cir. 1981). And, in any
event, it is settled that \`the views of
a subsequent Congress form a haz-
ardous basis for inferring the intent
of an earlier one.'" Jeerson County
Pharmaceutical Association v. Ab-
bott Laboratories
The bills to which petitioner refers,
however, were introduced in order
to [26] overcome the decisions in
Marubeni, Meyers, and Thevis. See,
e. g., S. 2320, 97th Cong., 2d
Sess. (1982). The introduction
of these bills hardly suggests that
their sponsors viewed those deci-
sions as correct interpretations of
Sec. 1963(a)(1).See United States
v. Gordon, 638 F.2d 886, 888, n.
5 (CA5), cert. denied, 452 U.S.
909 (1981). In any event, it is well
settled that \`the views of a subse-
quent Congress form a hazardous ba-
sis for inferring the intent of an ear-
lier one.'" Jeerson County Pharma-
ceutical Assn. v. Abbott Laborato-
ries.
75
75
Id. at 25-26; Brief for the United States at 49, Russello v. United States, 464 U.S.
16 (82-472) (1983).
135
For the same reasons, petitioner's ar-
gument draws no support from the
fact that certain state racketeering
laws provide expressly for the for-
feiture of \prots," \money," or \all
property, real or personal," acquired
from racketeering (see Pet. Br. 8-9).
With one exception, all of the state
provisions upon which petitioner re-
lies postdate federal court decisions
barring the forfeiture of racketeer-
ing prots under the federal law.
See United States v. Meyers, 432
F. Supp. 456 (W.D. Pa. 1977);
United States v. Thevis, 474 F.
Supp. 134 (N.D. Ga. 1979). 11
Undoubtedly aware of the problems
created by such decisions, the legisla-
tures of these states presumably em-
ployed language dierent from that
in 18 U.S.C. 1963(a)(1) in order to
avoid similar interpretations of their
new racketeering laws.
Neither are we persuaded by peti-
tioner's argument that his position
is supported by the fact that certain
state racketeering statutes expressly
provide for the forfeiture of \prots,"
\money," \interest or property," or
\all property, real or personal," ac-
quired from racketeering. Brief for
Petitioner 8-9. Nearly all of the
state statutes postdate the Meyers
and Thevis District Court decisions.
See, e. g., Colo. Rev. Stat. Sec.
18-17-106 (Supp. 1982) (enacted in
1981); R.I. Gen. Laws Sec. 7-15-
3 (Supp. 1982) (enacted in 1979).
The legislatures of those States pre-
sumably employed language dier-
ent from that of Sec. 1963(a)(1) so
as to avoid narrow interpretations of
their laws along the lines of the nar-
row interpretations given the federal
statute by the courts in Meyers and
Thevis.
76
76
Id. at 25-26; Brief for the United States at 49, Russello v. United States, 464 U.S.
16 (82-472) (1983).
136
As evidence that Congress did not
intend to reach racketeering prots,
petitioner points (Br. 14-15) to a
1969 letter from then Deputy At-
torney General Kleindienst to Sen-
ator McClellan... The court below
correctly concluded that this letter
did not indicate a congressional in-
tent to preclude forfeiture of racke-
teering prots. The sentence at issue
did not refer to Section 1963(a) as
nally enacted but to an earlier ver-
sion in which forfeiture was expressly
limited to interests in an enterprise.
Thus, by stating that forfeiture un-
der Section 1963(a) was \limited to
one's interest in the enterprise," the
letter was merely following the lan-
guage of the bill then pending. More-
over, the purpose of this sentence was
not to explain what the statutory
provision meant but to explain why
the Department of Justice believed it
was constitutional.
We are not persuaded otherwise by
the presence of a 1969 letter from
the then Deputy Attorney General
to Senator McClellan. See Measures
Relating to Organized Crime: Hear-
ings before the Subcommittee on
Criminal Laws and Procedures of the
Senate Committee on the Judiciary,
91st [29] Cong., 1st Sess., 407 (1969).
That letter, with its reference to
\one's interest in the enterprise" does
not indicate, for us, any congres-
sional intent to preclude forfeiture of
racketeering prots. The reference,
indeed, is not to Sec. 1963(a) as -
nally enacted but to an earlier ver-
sion in which forfeiture was to be ex-
pressly limited to an interest in an
enterprise. The letter was merely
following the language of the then
pending bill. Furthermore, the real
purpose of the sentence was not to
explain what the statutory provision
meant, but to explain why the De-
partment of Justice believed it was
constitutional.
77
Without delving into the psychology of the Justices and the clerks, it is im-
possible to determine if they use independent reasoning to reject the peti-
tioner's points in Russello. But, due to the high similarity between language
in the opinion and the SG's brief, it is clear, at very least, that the Court
found the SG made a compelling argument.
77
Id. at 28-29; Brief for the United States at 68-69, Russello v. United States, 464 U.S.
16 (82-472) (1983).
137
As with Russello, Justice Blackmun's opinion in United States v. National
Bank of Commerce,
78
almost entirely follows the linguistic template set forth
in the SG's brief. The case dealt with taxation rights under joint bank
accounts. In the instance below, the opinion uses highly similar language
to the reasoning provided by the SG's brief regarding Congress' statutory
intent:
In holding that Roy did not possess
\property [or] rights to property" on
which the IRS could levy, the court
of appeals relied heavily on Arkansas
creditors' rights law...This reasoning
seriously misconceives the role prop-
erly played by state law in federal tax
collection matters.
The Court of Appeals' conclusion
that Roy did not possess \property
[or] rights to property" on which the
IRS could levy rested heavily on its
understanding of the Arkansas law of
creditors' rights, particularly those in
garnishment. Id., at 1295-1296.See
Hayden v. Gardner, 238 Ark. 351,
381 S. W. 2d 752 (1964). As we have
suggested, this misconceives the role
properly played by state law in fed-
eral tax-collection matters.
79
78
472 U.S. 713 (1985)
79
Id. at 727; Brief for the United States at 41, United States v. National Bank of
Commerce, 472 U.S. 713 (84-498) (1985).
138
Contrary to the court of appeals'
conclusion, therefore, the facts that
under Arkansas law Roy's creditors
(unlike Roy himself) could not exer-
cise his right of withdrawal in their
favor (Pet. App. 7a) and would have
to join his co-depositors in a garnish-
ment proceeding (ibid.) are irrele-
vant in answering the question pre-
sented here. The federal statute, af-
ter all, refers to the taxpayer's prop-
erty and rights to property, not to
his creditor's rights. Yet the court of
appeals has eectively deprived the
federal levy statute of all indepen-
dent force, by remitting the IRS to
only the rights that an ordinary cred-
itor of the taxpayer would have un-
der state law. That result is plainly
wrong, for it is to \compare the gov-
ernment to a class of creditors to
which it is superior" (Randall, 542
F.2d at 274 n.8).
Thus, the facts that under Arkansas
law Roy's creditors, unlike Roy him-
self, could not exercise his right of
withdrawal in their favor and in a
garnishment proceeding would have
to join his codepositors are irrele-
vant. The federal statute relates
to the taxpayer's rights to prop-
erty and not to his creditors' rights.
The Court of Appeals would remit
the IRS to the rights only an ordi-
nary creditor would have under state
law. That result \[compares] the
government to a class of creditors to
which it is superior." Randall v. H.
Nakashima & Co., 542 F.2d 270, 274,
n. 8 (CA5 1976).
80
In its solicitude for the potential
claims of Roy's co-depositors, the
court of appeals has ignored the
statutory scheme that Congress es-
tablished.
In its understandable concern for
Ruby's and Neva's property inter-
ests, the Court of Appeals has ig-
nored the statutory scheme estab-
lished by Congress to protect those
rights.
81
80
Id.; Brief for the United States at 44, United States v. National Bank of Commerce,
472 U.S. 713 (84-498) (1985).
81
Id. at 728; Brief for the United States at 48, United States v. National Bank of
Commerce, 472 U.S. 713 (84-498) (1985).
139
As a nal justication for refusing
to impose personal liability on the
bank, the court of appeals theorized
that an IRS levy \is not normally in-
tended for use as against property in
which third parties have an interest"
or \as against property bearing on
its face the names of third parties."
(Pet. App. 17a). The court ap-
peared to recognize that Congress's
enactment of Section 7426 { which
[53] permits wrongful-levy actions
by \any person who claims an in-
terest in" seized property { tended
to undermine this theory. But the
court suggested that the Section 7426
remedy is designed to protect only
those third parties \whose property
has been seized 'inadvertently'" (Pet.
App. 17a).
The Court of Appeals' nal justi-
cation for its holding was its belief
that an IRS levy \is not normally
intended for use as against property
in which third parties have an inter-
est" or \as against property bearing
on its face the names of third par-
ties, and in which those third parties
likely have a property interest." 726
F.2d, at 1300. The court acknowl-
edged the existence of Sec. 7426 but
felt that that statute was designed
to protect only those third parties
\whose property has been seized 'in-
advertently.'" 726 F.2d, at 1300.
82
While Justice Blackmun's opinion in National Bank of Commerce utilizes
the brief's language throughout the entire opinion and especially with legal
reasoning, in a second tax case, United States v. Hughes Properties,
83
the
opinion utilizes language and reasoning from the SG's brief to interpret and
apply the Court's precedent in the area. Some examples include:
82
Id. at 729-30; Brief for the United States at 52-53, United States v. National Bank
of Commerce, 472 U.S. 713 (84-498) (1985).
83
476 U.S. 593 (1986)
140
This Court has consistently held that
a liability does not accrue for pur-
poses of the \all events" test as long
as it remains contingent. Brown v.
Helvering, 291 U.S. 193, 200 (1934).
To be deductible for tax purposes,
\the obligation to pay [must] ha[ve]
become nal and denite." Security
Flour Mills Co. v. Commissioner,
321 U.S. 281, 287 (1944). It must
be \unconditional." Lucas v. North
Texas Lumber Co., 281 U.S. 11, 13
(1930).
The Court's cases have emphasized
that \a liability does not accrue
as long as it remains contingent."
Brown v. Helvering, 291 U.S. 193,
200 (1934); accord, Dixie Pine Prod-
ucts Co. v. Commissioner, 320 U.S.
516, 519 (1944). Thus, to satisfy
the all-events test, a liability must
be \nal and denite in amount,"
Security Flour Mills Co. v. Com-
missioner, 321 U.S. 281, 287 (1944),
must be \xed and absolute," Brown
v. Helvering, 291 U.S., at 201, and
must be \unconditional," Lucas v.
North Texas Lumber Co., 281 U.S.
11, 13 (1930).
84
Rather, \the tax law requires that a
deduction be deferred until 'all the
events' have occurred that will make
it xed and certain" (Thor Power
Tool Co., 439 U.S. at 543).
And one may say that \the tax law
requires that a deduction be deferred
until 'all the events' have occurred
that will make it xed and certain."
Thor Power Tool Co. v. Commis-
sioner, 439 U.S. 522, 543(1979).
85
84
Id. at 600; Brief for the United States at 13, United States v. Hughes Properties, 476
U.S. 593 (85-554) (1986).
85
Id. at 600-01; Brief for the United States at 16, United States v. Hughes Properties,
476 U.S. 593 (85-554) (1986).
141
The Commissioner's \broad pow-
ers" to depart from the taxpayer's
usual accounting practice in comput-
ing taxable income owes in part to
\the vastly dierent objectives that
nancial and tax accounting have."
Thor Power Tool Co., 439 U.S. at
542. \The primary goal of nan-
cial accounting is to provide useful
information to management, share-
holders, [and] creditors" and \to pro-
tect these parties from being misled"
(ibid.).... \[T]he major responsibil-
ity of the Internal Revenue Service,"
by contrast, \is to protect the public
sc,"
The Court has long recognized \the
vastly dierent objectives that -
nancial and tax accounting have."
Thor Power Tool Co. v. Commis-
sioner, 439 U.S., at 542. The goal
of nancial accounting is to provide
useful and pertinent information to
management, shareholders, and cred-
itors. On the other hand, the ma-
jor responsibility of the Internal Rev-
enue Service is to protect the public
sc. Ibid.
86
In a case concerning the rights of a laborer upon losing his job at a Naval
Facility, Department of Navy v. Egan,
87
, Justice Blackmun authored an
opinion that adopts the case law and statutory analysis provided in the SG's
brief as the basis for the written opinion. There are several instances where
the opinion utilizes the brief's language in interpreting and in applying the
Court's precedents:
86
Id. at 603 Brief for the United States at 37-38, United States v. Hughes Properties,
476 U.S. 593 (85-554) (1986).
87
484 U.S. 518 (1988)
142
This Court has, on numerous occa-
sions, recognized the government's
\compelling interest" in withholding
national security information from
unauthorized persons in the course of
executive business (Snepp v. United
States, 444 U.S. 507, 509 n.3 (1980)
(per curiam)). See, e.g.,United
States v. Robel, 389 U.S. 258, 267
(1967); United States v. Reynolds,
345 U.S. 1, 10 (1953)...The authority
and the solemn obligation to protect
such information fall on the Presi-
dent as head of the Executive Branch
and Commander in Chief.
This Court has recognized the Gov-
ernment's \compelling interest" in
withholding national security infor-
mation from unauthorized persons
in the course of executive business.
Snepp v. United States, 444 U.S.
507, 509, n. 3 (1980). See also
United States v. Robel, 389 U.S.
258, 267 (1967); United States v.
Reynolds, 345 U.S. 1, 10... The au-
thority to protect such information
falls on the President as head of the
Executive Branch and as Comman-
der in Chief.
88
National security matters, as this
Court has recognized, are \the
province and responsibility of the ex-
ecutive." Haig v. Agee, 453 U.S. 280,
293-294, 304 (1981). \As to these
areas of Art. II duties the courts
have traditionally shown the utmost
deference to Presidential responsi-
bilities." United States v. Nixon,
418 U.S. 683, 710 (1974). Absent
an unambiguous grant of jurisdic-
tion by Congress, courts have tra-
ditionally been reluctant to intrude
upon the authority of the executive
in military and national security af-
fairs...The court of appeals' contrary
holding...puts the matter backwards.
The Court also has recognized \the
generally accepted view that foreign
policy was the province and respon-
sibility of the Executive." Haig v.
Agee, 453 U.S. 280, 293-294 (1981).
\As to these areas of [530] Art.
II duties the courts have tradition-
ally shown the utmost deference to
Presidential responsibilities." United
States v. Nixon, 418 U.S. 683, 710
(1974). Thus, unless Congress specif-
ically has provided otherwise, courts
traditionally have been reluctant to
intrude upon the authority of the Ex-
ecutive in military and national secu-
rity aairs...We feel that the contrary
conclusion of the Court of Appeals'
majority is not in line with this au-
thority.
89
88
Id. at 527; Petitioner's Brief at 30-31, Department of Navy v. Egan, 484 U.S. 518
(86-1552) (1988).
89
Id. at 529-30; Petitioner's Brief at 39-40, Department of Navy v. Egan, 484 U.S. 518
(86-1552) (1988).
143
The Court's language also tracks that from the brief's regarding the policy
rationale and implications in the case:
No individual has a \right" to a
security clearance. Under long es-
tablished principles, the grant of a
security clearance requires an af-
rmative act of discretion on the
part of the granting ocial based
on a high degree of condence in
the grantee...The general standard
therefore is that a clearance may be
granted only when \clearly consis-
tent with the interests of the na-
tional security." See, e.g., Exec. Or-
der No. 10,450, Sec. Sec. 2, 7, 3
C.F.R. 936, 938 (1949-1953 Comp.);
OPNAVINST 5510.1F, para. 16-
100(1); 10 C.F.R. 710.10(a) (De-
partment of Energy regulation); 32
C.F.R. 156.3(a) (Department of De-
fense regulation); Department of De-
fense Regulation 5200.2-R, para. 6-
100(a) (Dec. 1979).
It should be obvious that no one has
a \right" to a security clearance. The
grant of a clearance requires an ar-
mative act of discretion on the part
of the granting ocial. The general
standard is that a clearance may be
granted only when \clearly consis-
tent with the interests of the national
security." See, e. g., Exec. Order No.
10450, Sec. Sec. 2 and 7, 3 CFR 936,
938 (1949-1953 Comp.); 10 CFR Sec.
710.10(a) (1987) (Department of En-
ergy); 32 CFR Sec. 156.3(a) (1987)
(Department of Defense).
90
90
Id. at 528; Petitioner's Brief at 33-34, Department of Navy v. Egan, 484 U.S. 518
(86-1552) (1988).
144
A clearance determination, it should
be stressed, is not a judgment of
an individual or his past conduct.
It is an attempt to predict his fu-
ture behavior, to assess whether he
might...under the compulsion of cir-
cumstances beyond his control, com-
promise sensitive information. The
prediction may be based upon the
individual's past or present conduct;
but it may also be based upon con-
cerns unrelated to an individual's
conduct, such as whether he has close
relatives residing in a country that is
hostile to the United States
A clearance does not equate with
passing judgment upon an individ-
ual's character. Instead, it is only an
attempt to predict his possible future
behavior and to assess whether, un-
der compulsion of circumstances or
for other reasons, he might compro-
mise sensitive information. It may
be based, to be sure, upon past or
present conduct, but it also may be
based upon concerns completely un-
related to conduct, [529] such as hav-
ing close relatives residing in a coun-
try hostile to the United States.
91
Such predictive judgments must be
made by those with the necessary
expertise in protecting classied in-
formation. For \reasons too obvi-
ous to call for enlarged discussion"
(CIA v. Sims, 471 U.S. 159, 170
(1985)), the protection of classied
information must be committed to
the broad discretion of the agencies
responsible for such information, and
this must include broad discretion to
judge who may have access to it. It
is not reasonably possible for an out-
side, nonexpert body to review the
substance of such a judgment and de-
cide whether, under the \clearly con-
sistent" standard, the agency should
have been able to make the necessary
armative prediction with the neces-
sary condence.
Predictive judgment of this kind
must be made by those with the nec-
essary expertise in protecting classi-
ed information. For \reasons . .
. too obvious to call for enlarged
discussion," CIA v. Sims, 471 U.S.
159, 170 (1985), the protection of
classied information must be com-
mitted to the broad discretion of the
agency responsible, and this must in-
clude broad discretion to determine
who may have access to it. Certainly,
it is not reasonably possible for an
outside nonexpert body to review the
substance of such a judgment and
to decide whether the agency should
have been able to make the neces-
sary armative prediction with con-
dence.
92
91
Id. at 528-29; Petitioner's Brief at 34-35, Department of Navy v. Egan, 484 U.S. 518
(86-1552) (1988).
92
Id. at 529; Petitioner's Brief at 36-37, Department of Navy v. Egan, 484 U.S. 518
(86-1552) (1988).
145
Nor can such a body determine
what constitutes an acceptable mar-
gin of error in assessing the poten-
tial risk that condential informa-
tion will be compromised. Accord-
ingly, this Court has acknowledged
that with respect to employees in
sensitive positions \there is a reason-
able basis for the view that an agency
head who must bear the responsibil-
ity for the protection of classied in-
formation committed to his custody
should have the nal say in deciding
whether to repose his trust in an em-
ployee who has access to such infor-
mation." Cole v. Young, 351 U.S.
536, 546 (1956)
Nor can such a body determine what
constitutes an acceptable margin of
error in assessing the potential risk.
The Court accordingly has acknowl-
edged that with respect to employ-
ees in sensitive positions \there is a
reasonable basis for the view that an
agency head who must bear the re-
sponsibility for the protection of clas-
sied information committed to his
custody should have the nal say in
deciding whether to repose his trust
in an employee who has access to
such information." Cole v. Young,
351 U.S. 536, 546 (1956).
93
With John Doe Agency v. John Doe Corp.,
94
the Court examines an exemp-
tion under the federal Freedom of Information Act (FOIA). This Blackmun
opinion relies on the SG's brief primarily for the interpretation of the rele-
vant statutory provisions:
93
Id.; Petitioner's Brief at 37, Department of Navy v. Egan, 484 U.S. 518 (86-1552)
(1988).
94
493 U.S. 146 (1989)
146
The Legislative History Of Exemp-
tion 7, As Enacted And As Amended
In 1974, Conrms The Plain Mean-
ing Of Exemption 7...This Court
thoroughly discussed the legislative
history of Exemption 7 in NLRB
v. Robbins Tire Rubber Co., 437
U.S. 214, 224-236 (1978). As orig-
inally enacted, Exemption 7 per-
mitted nondisclosure of \investiga-
tory les compiled for law enforce-
ment purposes except to the extent
available by law to a private party."
80 Stat. 251. By that exemp-
tion, \Congress recognized that law
enforcement agencies had legitimate
needs to keep certain records con-
dential, lest the agencies be hindered
in their investigations or placed at a
disadvantage when it came time to
present their cases." Robbins Tire,
437 U.S. at 224.
If, despite what we regard as the
plain meaning of the statutory lan-
guage, it were necessary or advisable
to examine the legislative history of
Exemption 7, as originally enacted
and as amended in 1974, we would
reach the same conclusion. Justice
MARSHALL, writing for the Court
in Robbins Tire, 437 U.S., at 224-
236, discussed this legislative his-
tory in detail. In its original 1966
form, Exemption 7 permitted nondis-
closure of \investigatory les com-
piled for law enforcement purposes
except to the extent available by law
to a private party." Pub. L. 89-
487, x3(e)(7), 80 Stat. 251. But
the Court in Robbins Tire observed:
\Congress recognized that law en-
forcement agencies had legitimate
needs to keep certain records con-
dential, lest the agencies be hindered
in their investigations or placed at
a disadvantage when it came time
to present their cases." 437 U.S., at
224.
95
The legislative history of the 1974
amendments says nothing about lim-
iting Exemption 7 to those docu-
ments originating as law enforcement
records.
The legislative history of the 1974
amendments says nothing about lim-
iting Exemption 7 to those docu-
ments originating as law enforcement
records.
96
95
Id. at 155-56; Petitioner's Brief at 39, John Doe Agency v. John Doe Corp., (88-1083)
493 U.S. 146 (1989).
96
Id. at 157; Petitioner's Brief at 46, John Doe Agency v. John Doe Corp., (88-1083)
493 U.S. 146 (1989).
147
This Court has consistently taken
a practical approach to interpreting
FOIA in an eort to apply a work-
able statutory balance between the
interests of the public in greater ac-
cess to information and the needs
of the government in protecting cer-
tain kinds of information from dis-
closure. See, e.g., EPA v. Mink,
410 U.S. 73, 80 (1973);Department of
the Air Force v. Rose, 425 U.S. 352,
361-362 (1976);Baldrige v. Shapiro,
455 U.S. 345, 352 (1982). In ac-
cord [54] with that approach, the
Court looks to the reasons for ex-
emption from FOIA's mandatory dis-
closure requirements in determining
whether the government has properly
invoked a particular exemption. See,
e.g., NLRB v. Sears, Roebuck Co.,
421 U.S. 132, 148-154 (1975); FBI v.
Abramson, 456 U.S. at 630. In ap-
plying Exemption 7, this Court has
looked carefully at the eect that dis-
closure would have on the interests
that exemption seeks to protect. In
NLRB v. Robbins Tire Rubber Co.,
437 U.S. at 242-243.
This Court consistently has taken a
practical approach when it has been
confronted with an issue of interpre-
tation of the Act. It has endeavored
to apply a workable balance between
the interests of the public in greater
access to information and the needs
of the Government to protect cer-
tain kinds of information from dis-
closure. The Court looks to the
reasons for exemption from the dis-
closure requirements in determining
whether the Government has prop-
erly invoked a particular exemption.
See e.g., NLRB v. Sears, Roebuck
Co., 421 U.S. 132, 148-154 (1975).
In applying Exemption 7, the Court
carefully has examined the eect that
disclosure would have on the interest
the exemption seeks to protect. Rob-
bins Tire, 437 U.S., at 242-243.
97
The opinion also relies on the SG's interpretation of the lower court opinion
and of the opposing parties' arguments:
97
Id.; Petitioner's Brief at 53-54, John Doe Agency v. John Doe Corp., (88-1083) 493
U.S. 146 (1989).
148
The plain meaning of the word \com-
pile" does not permit such a distinc-
tion [as made by the Court of Ap-
peals]. It is the pulling together of
materials that constitutes the com-
piling. It does not matter whence
the documents were obtained...This
Court itself has used the word \com-
piled," quite naturally, to refer to
the process of pulling together at one
time records and information that
were generated (or even compiled) at
an earlier time and for dierent pur-
poses. In FBI v. Abramson, 456 U.S.
at 622 n.5
We disagree with that interpretation
for, in our view, the plain meaning
of the word \compile," or, for that
matter, of its adjectival form \com-
piled," does not permit such rene-
ment. This Court itself has used
the word \compile" naturally to re-
fer even to the process of gathering
at one time records and information
that were generated on an earlier oc-
casion and for a dierent purpose.
See FBI v. Abramson, 456 U.S., at
622, n. 5.
98
Even respondent has used the noun
form of the word \compiled" in its
ordinary sense to refer to the gather-
ing together of documents, whether
or not they were generated or com-
piled at an earlier time for a dier-
ent purpose. In its FOIA requests
of September 30, 1986, and Febru-
ary 3, 1987, respondent \ask[ed] that
copies of the requested materials be
furnished to us as soon as individ-
ual items are available, and that your
response to this request not await a
compilation of all the materials re-
quested." J.A. 21, 47-48. Thus re-
spondent, unlike the court of ap-
peals, obviously and quite properly
recognized that the documents' hav-
ing been \compiled" once for the
purpose of routine audits would in
no way prevent their being compiled
again later for a dierent purpose.
Respondent, too, has used the word
\compile" in its ordinary sense to re-
fer to the assembling of documents,
even though those documents were
put together at an earlier time for a
dierent purpose. In its FOIA re-
quests of September 30, 1986, and
February 3, 1987, respondent asked
that the requested materials be fur-
nished as soon as they were available,
and that the response to the request
\not await a compilation of all the
materials requested." App. 21, 47-
48. This was a recognition, twice re-
peated, that the documents having
been compiled once for the purpose
of routine audits were not disqual-
ied from being \compiled" again
later for a dierent purpose.
99
98
Id. at 154; Petitioner's Brief at 32-33, John Doe Agency v. John Doe Corp., (88-1083)
493 U.S. 146 (1989).
99
Id. at 154-55; Petitioner's Brief at 36, John Doe Agency v. John Doe Corp., (88-1083)
493 U.S. 146 (1989).
149
In sum, the ruling of the court of
appeals that Exemption 7(A) can-
not be invoked in this case is [38]
at odds with the plain meaning of
the exemption. The articial dis-
tinction that the court of appeals
has drawn between documents that
were originally obtained for law en-
forcement purposes and those that
were not so originally obtained, but
later gathered together for law en-
forcement purposes, nds no support
in the plain language of Exemption 7.
We thus do not accept the distinc-
tion the Court of Appeals drew be-
tween documents that originally were
assembled for law enforcement pur-
poses and those that were not so orig-
inally assembled but were gathered
later for such purposes. The plain
language of Exemption 7 does not
permit such a distinction.
100
By adopting the SG's position both regarding the lower court decision and
the opposing parties' contentions, the Blackmun opinion clearly aligns itself
with the SG. An opinion with shared reasoning in this fashion leaves little
room for one to speculate on the Court's conclusions well before the Court's
judgment is set forth.
In the constitutional case of University of Pennsylvania v. EEOC,
101
the
Court examines whether universities deserve First Amendment privileges
against disclosing peer reviewed material when claims of racial or sexual
discrimination are raised in tenure decisions. The opinion parallels the SG's
analysis of Title VII provisions and the SG's explanation for why the peti-
tioner interpreted these provisions incorrectly:
100
Id. at 155; Petitioner's Brief at 37-38, John Doe Agency v. John Doe Corp., (88-1083)
493 U.S. 146 (1989).
101
493 U.S. 182 (1990)
150
The eect of the elimination of Title
VII's exemption for educational in-
stitutions was to expose tenure deter-
minations to the same enforcement
procedures applicable to other em-
ployment decisions. As this Court
has noted, Title VII creates \`an in-
tegrated, multistep enforcement pro-
cedure' that enables the Commission
to detect and remedy instances of
discrimination." EEOC v. Shell Oil
Co., 466 U.S. 54, 62 (1984) (cita-
tion omitted). 10 The ecacy of
each step of that procedure depends
directly on the Commission's unen-
cumbered access to information rele-
vant to alleged discrimination. The
Commission's enforcement responsi-
bilities are triggered by the ling of a
specic, sworn charge of discrimina-
tion. The Act obligates the Commis-
sion to investigate charges of discrim-
ination to determine whether \there
is reasonable cause to believe that
the charge is true." Section 706(b) of
Title VII, 42 U.S.C. 2000e-5(b).
The eect of the elimination of this
exemption was to expose tenure de-
terminations to the same enforce-
ment procedures applicable to other
employment decisions. This Court
previously has observed that Title
VII \sets forth 'an integrated, multi-
step enforcement procedure' that en-
ables the Commission to detect and
remedy instances of discrimination."
EEOC v. Shell Oil Co., 466 U.S.
54, 62 (1984), quoting Occidental
Life Ins. Co. v. EEOC, 432 U.S.
355, 359 (1977). The Commission's
enforcement responsibilities are trig-
gered by the ling of a specic sworn
charge of discrimination. The Act
obligates the Commission to inves-
tigate a charge of discrimination to
determine whether there is \reason-
able cause to believe that the charge
is true." 42 U.S.C. x2000e-5(b) (1982
ed.).
102
102
Id. at 190-91; Brief for the Respondent at 30-32, University of Pennsylvania v.
EEOC, 493 U.S. 182 (88-493) (1990).
151
The Commission's enforcement re-
sponsibilities are triggered by the l-
ing of a specic, sworn charge of dis-
crimination. The Act obligates the
Commission to investigate charges of
discrimination to determine whether
\there is reasonable cause to be-
lieve that the charge is true." Sec-
tion 706(b) of Title VII, 42 U.S.C.
2000e-5(b). If it nds no reason-
able cause, the Commission is obli-
gated to dismiss the charge. Ibid.
If it does nd reasonable cause, the
Commission \endeavor[s] to elimi-
nate [the] alleged unlawful employ-
ment practice by informal methods of
conference, conciliation, and persua-
sion." Ibid. This requirement re
ects
Congress's wish \that violations of
the statute could be remedied with-
out resort to the courts." EEOC v.
Shell Oil Co., 466 U.S. at 78; see id.
at 90 (O'Connor, J., concurring in
part and dissenting in part). See also
Occidental Life Ins. Co. v. EEOC,
432 U.S. 355, 368 (1977). If attempts
at voluntary resolution fail, the Com-
mission may bring an action against
the employer in accordance with Sec-
tion 706(f)(1) of the Act, 42 U.S.C.
2000e-5(f)(1).
The Commission's enforcement re-
sponsibilities are triggered by the
ling of a specic sworn charge
of discrimination. The Act obli-
gates the Commission to investigate
a charge of discrimination to deter-
mine whether there is \reasonable
cause to believe that the charge is
true." 42 U.S.C. x2000e-5(b) (1982
ed.). If it nds no such reasonable
cause, the Commission is directed to
dismiss the charge. If it does nd rea-
sonable cause, the Commission shall
\endeavor to eliminate [the] alleged
unlawful employment practice by in-
formal methods of conference, concil-
iation, and persuasion." Ibid. If at-
tempts at voluntary resolution fail,
the Commission may bring an ac-
tion against the employer. 2000e-
5(f)(1).
103
103
Id.
152
To enable the Commission to make
informed decisions at each stage
of the enforcement process,Section
709(a) of Title VII, 42 U.S.C. 2000e-
8(a), confers a broad right of access
to relevant evidence: [T]he Commis-
sion or its designated representative
shall have access to, for the purposes
of examination, and the right to copy
any evidence of any person being
investigated that relates to unlaw-
ful employment practices covered by
[the Act] and is relevant to the charge
under investigation. If employers
refuse to provide information volun-
tarily, the Act authorizes the Com-
mission to issue subpoenas and to
seek orders enforcing them. Section
710 of Title VII, 42 U.S.C. 2000e-9
(incorporating 29 U.S.C. 161).
To enable the Commission to make
informed decisions at each stage of
the enforcement process, x2000e-8(a)
confers a broad right of access to rel-
evant evidence: \The Commission or
its designated representative shall at
all reasonable times have access to,
for the purposes of examination, and
the right to copy any evidence of any
person being investigated . . . that
relates to unlawful employment prac-
tices covered by [the Act] and is rel-
evant to the charge under investiga-
tion." If an employer refuses to pro-
vide this information voluntarily, the
Act authorizes the Commission to is-
sue a subpoena and to seek an order
enforcing it. x2000e-9 (incorporating
29 U.S.C. x161).
104
104
Id.
153
Petitioner argues, nevertheless, that
Title VII leaves courts with discre-
tion to provide additional protection
for tenure review documents. Al-
though petitioner recognizes that Ti-
tle VII gives the Commission broad
\power to seek access to all evidence
that may be 'relevant to the charge
under investigation'" (Pet. Br. 38),
it nevertheless contends that Title
VII's subpoena enforcement provi-
sions do not give the Commission an
unqualied right to acquire such ev-
idence. See Pet. Br. 38-41. That
interpretation is untenable. First,
the plain language of Section 709(a)
of Title VII, 42 U.S.C.2000e-8(a),
states that the Commission \shall
have access" to relevant evidence;
this can only be read as giving the
Commission a right to that evidence,
not a mere \power to seek" it.
Petitioner argues, nevertheless, that
Title VII armatively grants courts
the discretion to require more than
relevance in order to protect tenure-
review documents. Although pe-
titioner recognizes that Title VII
gives the Commission broad \power
to seek access to all evidence that
may be 'relevant to the charge un-
der investigation,'" Brief for Peti-
tioner 38 (emphasis added), it con-
tends that Title VII's subpoena en-
forcement provisions do not give the
Commission an unqualied right to
acquire such evidence. Id., at 38-
41. This interpretation simply can-
not be reconciled with the plain lan-
guage of the text of x2000e-8(a),
which states that the Commission
\shall...have access" to \relevant" ev-
idence (emphasis added). The provi-
sion can be read only as giving the
Commission a right to obtain that
evidence, not a mere license to seek
it.
105
105
Id. at 192; Brief for the Respondent at 44-45, University of Pennsylvania v. EEOC,
493 U.S. 182 (88-493) (1990).
154
Signicantly, Title VII anticipates
and addresses situations in which an
employer may have an interest in the
condentiality of its records. The
same Section that gives the Commis-
sion access to any evidence relevant
to its investigations also makes it
\unlawful for any ocer or employee
of the Commission to make public
in any manner whatever any infor-
mation obtained by the Commission
pursuant to its authority under this
section prior to the institution of any
proceeding" under the Act. Section
709(e) of Title VII, 42 U.S.C. 2000e-
8(e). Any violation of this provision
subjects the Commission's employees
to criminal penalties. Ibid.
Congress did address situations in
which an employer may have an in-
terest in the condentiality of its
records. The same x2000e-8 which
gives the Commission access to any
evidence relevant to its investigation
also makes it \unlawful for any o-
cer or employee of the Commission
to make public in any manner what-
ever any information obtained by the
Commission pursuant to its author-
ity under this section prior to the
institution of any proceeding" under
the Act. A violation of this provi-
sion subjects the employee to crimi-
nal penalties. Ibid.
106
This deference to the SG's interpretation of Congress' statutory scheme is
also evident in the tax case Commissioner of Internal Revenue v. Keystone
Consolidated Industries.
107
The opinion uses the same or similar language
to the SG's in its interpretation of the tax statute and uses the same canons
of interpretation. A few examples of this include:
106
Id.; Brief for the Respondent at 43-44, University of Pennsylvania v. EEOC, 493
U.S. 182 (88-493) (1990).
107
508 U.S. 152 (1993)
155
But even if \sale or exchange" had
not had a settled meaning under the
Internal Revenue Code, it would be
clear that Section 4975(c)(1)(A) pro-
hibits the transfer of property in sat-
isfaction of a debt. Congress did not
merely prohibit a \sale or exchange,"
it barred \any direct or indirect \sale
or exchange" between employers and
the pension plans they sponsor. At
the least, the contribution of prop-
erty in satisfaction of a funding obli-
gation is a type of sale of the prop-
erty. It is equally surely a form of
exchange, since the property is ex-
changed for diminution of the em-
ployer's funding obligation.
Even if this phrase had not possessed
a settled meaning, it still would be
clear that x4975(c)(1)(A) prohibits
the transfer of property in satisfac-
tion of a debt. Congress barred
not merely a \sale or exchange." It
prohibited something more, namely,
\any direct or indirect \sale or ex-
change." The contribution of prop-
erty in satisfaction of a funding obli-
gation is at least both an indirect
type of sale and a form of exchange,
since the property is exchanged for
diminution of the employer's funding
obligation.
108
Congress's goal in enacting the pro-
vision, which was to bar categori-
cally transactions that are likely to
injure pension plans. S. Rep. No.
383, supra, at 95-96. The transfer
of property to a pension plan in sat-
isfaction of a funding obligation can
jeopardize the ability of the plan to
pay promised benets, and Congress
sought to eliminate that possibility.
Congress' goal was to bar categor-
ically a transaction that was likely
to injure the pension plan. S. Rep.
No. 93-383, pp. 95-96 (1973). The
transfer of encumbered property may
jeopardize the ability of the plan to
pay promised benets.
109
108
Id. at 159; Brief for the Petitioner at 30, Commissioner of Internal Revenue v.
Keystone Consolidated Industries, 508 U.S. 152 (91-1677) (1993).
109
Id. at 160; Brief for the Petitioner at 30-31, Commissioner of Internal Revenue v.
Keystone Consolidated Industries, 508 U.S. 152 (91-1677) (1993).
156
The statutory text at issue { pro-
viding that a transfer of encum-
bered property \shall be treated as"
a sale or exchange { supports the
Fourth Circuit's view that Congress
intended Section 4975(f)(3) to ex-
pand the scope of the prohibited
transaction provision... Thus, Sec-
tion 4975(f)(3) amplies and extends
the reach of \sale or exchange" in
Section 4975(c)(1)(A) to include con-
tributions of encumbered property
that do not satisfy funding obliga-
tions. The legislative history con-
rms that Congress understood Sec-
tion 4975(f)(3) to enlarge, rather
than restrict, the reach of the pro-
hibited transaction provision...thus,
Congress intended Section 4975(f)(3)
to provide additional protection, not
to limit the protection provided by
Section 4975(c)(1)(A).
We feel that by this language
Congress intended x4975(f)(3) to ex-
pand, not limit, the scope of the
prohibited-transaction provision. It
extends the reach of \sale or ex-
change" in x4975(c)(1)(A) to include
contributions of encumbered prop-
erty that do not satisfy funding obli-
gations. See H.R. Conf. Rep. No.
93-1280, p. 307 (1974). Congress
intended by x4975(f)(3) to provide
additional protection, not to limit
the protection already provided by
x4975(c)(1)(A).
110
Also as with the Doe case, the Court in Keystone Properties adopts the
SG's language to reject the rationale for the lower court's decision:
110
Id. at 161; Brief for the Petitioner at 38; 40, Commissioner of Internal Revenue v.
Keystone Consolidated Industries, 508 U.S. 152 (91-1677) (1993).
157
The court of appeals interpreted
\sale or exchange" in Section
4975(c)(1)(A) contrary to its ordi-
nary, settled meaning primarily as a
result of its erroneous construction of
Section 4975(f)(3). That provision
states, in pertinent part, that \[a]
transfer [of] real or personal prop-
erty by a disqualied person to a plan
shall be treated as a sale or exchange
if the property is subject to a mort-
gage or similar lien." The court of ap-
peals called Section 4975(f)(3) [37] a
\denition" and read it as \implying
that unless [property] is encumbered
by a mortgage or lien, a transfer of
property is not to be treated as if it
were a sale or exchange."
We do not agree with the Court of
Appeals' conclusion that x4975(f)(3)
limits the meaning of \sale or ex-
change," as that phrase appears in
x4975(c)(1)(A). Section 4975(f)(3)
states that a transfer of property \by
a disqualied person to a plan shall
be treated as a sale or exchange if the
property is subject to a mortgage or
similar lien." The Court of Appeals
read this language as implying that
unless property \is encumbered by a
mortgage or lien, a transfer of prop-
erty is not to be treated as if it were a
sale or exchange." 951 F.2d at 78.
111
The Blackmun opinion in Posters N' Things v. United States,
112
follows a
similar form to the previously described decisions. In the same manner as in
those decisions, the opinion adopts an abundance of language from the SG's
brief regarding Congress' statutory intent, the Supreme Court's precedent,
and the opposing parties' arguments. The case itself deals with the consti-
tutionality of a federal drug paraphernalia statute's scienter requirements.
Looking at Congress' intent, the Court and the SG's brief state respectively:
111
Id.; Brief for the Petitioner at 36-37, Commissioner of Internal Revenue v. Keystone
Consolidated Industries, 508 U.S. 152 (91-1677) (1993).
112
511 U.S. 513 (1994)
158
Congress omitted a factor that would
have borne much more directly on
the question of subjective intent
{ the defendant's own statements
about his intent. That omission
is particularly striking when Section
857 is compared to the Model Drug
Paraphernalia Act. The Model Act
includes among the relevant factors
\statements by an owner concerning
[the] use" of the object and \direct or
circumstantial evidence of the intent
of an owner to deliver it to persons
whom he knows, or should reason-
ably know, intend to use the object
to facilitate a violation of this Act." 8
App., infra, 6a-7a. Congress's omis-
sion of both factors in Section 857
indicates that it did not intend to
dene drug paraphernalia in terms
of the subjective intent of the defen-
dant.
Congress did not include among the
listed factors a defendant's state-
ments about his intent or other fac-
tors directly establishing subjective
intent. This omission is signicant in
light of the fact that the parallel list
contained in the Drug Enforcement
Administration's Model Drug Para-
phernalia Act, on which x857 was
based, 8 includes among the relevant
factors \statements by an owner . .
. concerning [the object's] use" and
\direct or circumstantial evidence of
the intent of an owner . . . to de-
liver it to persons whom he knows,
or should reasonably know, intend to
use the object to facilitate a violation
of this Act." 9 An objective construc-
tion of the denitional provision also
nds support in x857(f), which estab-
lishes an exemption for items \tra-
ditionally intended for use with to-
bacco products." An item's \tradi-
tional" use is not based on the sub-
jective intent of a particular defen-
dant.
113
113
Id. at 520-21; Brief for the United States at 28, Posters N' Things v. United States,
511 U.S. 513 (92-903) (1994).
159
In 1988, Congress replaced \primar-
ily" with \traditionally" in order to
\clarify" the meaning of the exemp-
tion. See Pub. L. No. 100-690,
Tit. IV, x6485, 102 Stat. 4384. If
Congress had meant to shift from a
subjective to an objective concept of
intent, it is unlikely that it would
have characterized the amendment
as merely \clarifying" the law.
An objective construction of the def-
initional provision also nds support
in HN11 x857(f), which establishes
an exemption for items \traditionally
intended for use with tobacco prod-
ucts." 10 An item's \traditional" use
is not based on the subjective [521]
intent of a particular defendant. In
1988, Congress added the word \tra-
ditionally" in place of \primarily" in
the x857(f) exemption in order to
\clarify" the meaning of the exemp-
tion. Pub. L. 100-690, Tit. VI,
x6485, 102 Stat. 4384. Congress'
characterization of the amendment
as merely \clarifying" the law sug-
gests that the original phrase { \pri-
marily intended" { was not a refer-
ence to the fundamentally dierent
concept of a defendant's subjective
intent.
114
114
Id. at 520-21; Brief for the United States at 29, Posters N' Things v. United States,
511 U.S. 513 (92-903) (1994).
160
\The failure of Congress explic-
itly and unambiguously to indicate
whether mens rea is required does
not signal a departure from this
background assumption of our crim-
inal law." Liparota, 471 U.S. at 426.
Instead, \far more than the simple
omission of the appropriate phrase
from the statutory denition is nec-
essary to justify dispensing with an
intent requirement." United States
Gypsum Co., 438 U.S. at 438. De-
spite this presumption, \courts obvi-
ously must follow Congress' intent as
to the required level of mental cul-
pability for any particular oense."
United States v. Bailey, 444 U.S.
394, 406 (1980). In this instance,
however, we do not believe that the
language or legislative history of Sec-
tion 857 demonstrates that Congress
intended to dispense with a mens rea
requirement.
Neither our conclusion that Congress
intended an objective construction of
the \primarily intended" language in
x857(d), nor the fact that Congress
did not include the word \know-
ingly" in the text of x857, justies the
conclusion that Congress intended to
dispense entirely with a scienter re-
quirement. This Court stated in
United States v. United States Gyp-
sum Co., 438 U.S. 422, 438, 57 L. Ed.
2d 854, 98 S. Ct. 2864 (1978): \Cer-
tainly far more than the simple omis-
sion of the appropriate phrase from
the statutory denition is necessary
to justify dispensing with an intent
requirement."
115
Responding to the arguments raised by petitioner, the opinion and the brief
state:
115
Id. at 522; Brief for the United States at 32, Posters N' Things v. United States, 511
U.S. 513 (92-903) (1994).
161
Petitioners contend (Pet. Br. 15-18)
that Section 857 is unconstitution-
ally vague as applied in this case...
Whatever its standing in the ab-
stract, Section 857 is not unconstitu-
tionally vague as applied to petition-
ers. The void-for-vagueness doctrine
\requires that a penal statute dene
the criminal oense with sucient
deniteness that ordinary people can
understand what conduct is prohib-
ited and in a manner that does not
encourage arbitrary and discrimina-
tory enforcement." Kolender v. Law-
son, 461...Many of the items at issue
in this case { including bongs, roach
clips, and pipes designed for use with
illegal drugs { are listed in Section
857(d). There is no plausible basis
for arguing that the statute is uncon-
stitutionally vague concerning those
items.
Petitioners argue that x857 is un-
constitutionally vague as applied to
them in this case...Whatever its sta-
tus as a general matter, we can-
not say that x857 is unconstitution-
ally vague as applied in this case.
First, the list of items in x857(d)
constituting per se drug parapher-
nalia provides individuals and law
enforcement ocers with relatively
clear guidelines as to prohibited con-
duct. With respect to the listed
items, there can be little [526] doubt
that the statute is suciently de-
terminate to meet constitutional re-
quirements. Many items involved in
this case { including bongs, roach
clips, and pipes designed for use with
illegal drugs { are among the items
specically listed in x857(d).
116
116
Id. at 525-26; Brief for the United States at 45-46; 48, Posters N' Things v. United
States, 511 U.S. 513 (92-903) (1994).
162
Petitioner Acty contends (Pet. Br.
16-17) that she was improperly con-
victed of aiding and abetting the
manufacture and distribution of co-
caine, in violation of 21 U.S.C.
841(a)(1) (Count 4). She asserts (Br.
16-17) that \the government's wit-
nesses testied that mannitol, man-
nite and inositol were 'objectively'
drug paraphernalia," and that un-
der the district court's instructions
the jury was \required" to nd
that the substances were intended
for manufacturing with a controlled
substance." Petitioner argues that
the district court's instructions \cre-
ate[d] 'a presumption that relieve[d]
the [government] of its burden of
persuasion on an element of the of-
fense,'" in violation of the Due Pro-
cess Clause of the Fifth Amendment.
Pet. Br. 17, quoting Francis v.
Franklin, 471 U.S. 307, 315 (1985).
Petitioner did not raise that argu-
ment in the court of appeals, and
that court did not address it.
Petitioner Acty's other contentions
are not properly before the Court.
First, she argues that she was im-
properly convicted of aiding and
abetting the manufacture and distri-
bution of cocaine because the jury
instructions created a \presumption"
that certain items of drug parapher-
nalia \were intended for manufac-
turing with a controlled substance."
Brief for Petitioners. This argument
was neither raised in nor addressed
by the Court of Appeals.
117
In another section of the opinion, the Court establishes the SG's interpre-
tation of the phrase, \primarily intended" in the statute as the denition
governing the phrase:
117
Id. at 527; Brief for the United States at 51, Posters N' Things v. United States, 511
U.S. 513 (92-903) (1994).
163
Finally, our objective construction
of the \primarily intended for use"
language avoids the creation of an
unusual mens rea standard and
is consistent with the meaning of
similar language in other federal
criminal statutes. See 18 U.S.C.
921(a)(17)(B) (\armor piercing am-
munition" excludes any projectile
that the Secretary of the Trea-
sury nds is \primarily intended"
to be used for sporting purposes);
21 U.S.C. 860(d)(2) (\youth cen-
ter" means a recreational facility \in-
tended primarily for use by persons
under 18 years of age").
Finally, an objective construction of
the phrase \primarily intended" is
consistent with the natural read-
ing of similar language in deni-
tional provisions of other federal
criminal statutes. See 18 U.S.C.
x921(a)(17)(B) (\armor piercing am-
munition" excludes any projectile
that is \primarily intended" to be
used for sporting purposes, as found
by the Secretary of the Treasury); 21
U.S.C. x860(d)(2) (1988 ed., Supp.
V) (\youth center" means a recre-
ational facility \intended primarily
for use by persons under 18 years of
age").
118
4.3.4 Non-Blackmun Cases
Justice Blackmun's Type I opinions present some of the most elaborate
examples of opinions that adopt language from the briefs in all of their
facets including legal reasoning. They also represent the majority of Type
I opinions. Other Type I opinions show greater variation in the material
they include from the briefs, but they all tend to take the focal brief as the
template for the language in the opinion. Many of these, like the Blackmun
opinions, rely on the SG's briefs from beginning to the end.
The earliest Type I relationship in the dataset comes from a 1955 decision
authored by Chief Justice Earl Warren, Steiner v. Mitchell.
119
This case is
atypical for this group of cases both because the legal reasoning section of
the opinion is quite small compared to the discussion of facts and because
the percent of overlapping language in the facts is higher than in any other
118
Id. at 521; Brief for the United States at 30, Posters N' Things v. United States, 511
U.S. 513 (92-903) (1994).
119
350 U.S. 247 (1956)
164
Lifted case at 71%. The Court's reliance on the government's (respondent)
brief is apparent at the outset when the question posed in the opinion is
compared to the question in both briefs (The Petitioner's Brief, which is not
highly similar to the opinion is in the far left column. The highly similar
Respondent's Brief is in the center column and the opinion is on the right):
Under the provisions of
the Fair Labor Stan-
dards Act, as amended
by the Portal-to-Portal
Act, does time spent by
petitioner's employees
at its battery manufac-
turing plant in chang-
ing from street clothes
into work clothes prior
to punching the time
clock at the beginning
of the work day, and
in taking shower baths
and changing from work
clothes to street clothes
after punching out the
time clock at the end
of the work day, con-
stitute compensable
\time worked" under
the amended Act?
Where workers in a bat-
tery plant must make
extensive use of dan-
gerously caustic and
toxic materials and are
compelled by circum-
stances, including vi-
tal considerations of
health and hygiene, to
change clothes and ...
to shower in facilities
which State law re-
quires their employer
to provide, are these
\principal," rather
than \preliminary" or
\postliminary," activi-
ties within the meaning
of the Portal-to-Portal
Act.
The precise question is
whether workers in a
battery plant must be
paid as a part of their
\principal" activities
for the time incident to
changing clothes at the
beginning of the shift
and showering at the
end, where they must
make extensive use of
dangerously caustic and
toxic materials, and are
compelled by circum-
stances, including vital
considerations of health
and hygiene, to change
clothes and to shower
in facilities which state
law requires their em-
ployer to provide, or
whether these activities
are \preliminary" or
\postliminary" within
the meaning of the
Portal-to-Portal Act.
120
120
Id. at 248; Petitioner's Brief at 2-3 Steiner v. Mitchell, 350 U.S. 247 (No. 22) (1956);
Brief for the Respondent at 2 Steiner v. Mitchell, 350 U.S. 247 (No. 22) (1956).
165
The opinion in Steiner follows the facts as asserted in the SG's brief and
deviates considerably from the facts as conveyed by the petitioner. These
facts are often drawn from the record in the case, but as part of the opinion,
they are damaging to the petitioner's contention that the working conditions
in the battery plant are not hazardous and should not require the workers to
take additional precautions to maintain their safety. The following examples
exemplify how the pictures painted by the SG and the Court are quite
dierent from those conveyed by the petitioner.
166
The manufacturing pro-
cess for storage batter-
ies involves the han-
dling of toxic matter,
such as sulphuric acid
and lead oxide. This
matter damages clothes
if it is spilled on them.
All of the company's
production employees
customarily work with
or near the various
chemicals used in the
plant (fdg. 4, R. 213).
These include lead
metal, lead oxide, lead
sulphate, lead perox-
ide, and sulphuric acid
(ibid.). Some of these
are in liquid form, some
are in powder form, and
some are solid (R. 102,
103, 118, 154, 181-183).
In the manufacturing
process some of the ma-
terials go through var-
ious changes (R. 26)
and give o dangerous
fumes (R. 47-48, 60, 70,
98-99, 126-127). Some
are spilled or dropped
and thus become a part
of the dust in the air
(R. 27, 71, 108, 117,
154). In general, the
chemicals permeate the
entire plant and every-
thing and everyone in
it.
All of the produc-
tion employees, such
as those with whom
we are here concerned,
customarily work with
or near the various
chemicals used in the
plant. These include
lead metal, lead ox-
ide, lead sulphate, lead
peroxide, and sulphuric
acid. Some of these are
in liquid form; some are
in powder form, and
some are solid. In the
manufacturing process,
some of the materials go
through various changes
and give o dangerous
fumes. Some are spilled
or dropped, and thus
become a part of the
dust in the air. In gen-
eral, the chemicals per-
meate the entire plant
and everything and ev-
eryone in it.
121
In another instance the opinion directly adopts the SG's reasoning concern-
ing the petitioner's treatment of its workers.
121
Id. at 249-50; Petitioner's Brief at 4, Steiner v. Mitchell, 350 U.S. 247 (No. 22)
(1956); Brief for the Respondent, at 3-4 Steiner v. Mitchell, 350 U.S. 247 (No. 22) (1956).
167
Petitioners concededly do not record
or pay for the time which their em-
ployees spend in clothes changing
and showering, which was found to
amount to thirty minutes a day (ten
minutes in the morning and twenty
minutes in the afternoon) for each
employee (R. 221)... Petitioners do
not challenge the concurrent nding
of the courts below that the clothes-
changing and showering activities of
the battery plant employees (men
who work with or near dangerously
toxic materials) are indispensable to
the performance of productive work
and integrally related thereto. See
Pet. Br., p. 33.
Petitioners do not record or pay
for the time which their employees
spend in these activities, which was
found to amount to thirty minutes
a day, ten minutes in the morn-
ing and twenty minutes in the af-
ternoon, for each employee. They
do not challenge the concurrent nd-
ings of the courts below that the
clothes-changing and showering ac-
tivities of the employees are indis-
pensable to the performance of their
productive work and integrally re-
lated thereto.
122
While the opinion in Steiner is very fact intensive, most Lifted opinions,
as is apparent from the Blackmun examples, are an even mix of fact and
legal reasoning. The next Lifted opinion, Tilton v. Missouri Pacic Rail-
road,
123
authored by Justice Goldberg is more typical of the Lifted type.
This opinion language parallels the reasoning found in the SG's brief as well
as the SG's construction of the facts. The case concerns the reemployment
rights of veterans under federal legislation. The certainty present in the
petitioner's argument makes its way into the Court's opinion as is apparent
in the following:
122
Id. at 251; Brief for the Respondent at 7 Steiner v. Mitchell, 350 U.S. 247 (No. 22)
(1956).
123
376 U.S. 169 (1964)
168
There is no room for doubt then,
that, had Tilton, Beck and McClearn
remained continuously on the job,
they would have been able to com-
plete the work period and qualify
as journeymen in advance of those
who pass them in seniority during
their absence. Each was entitled, un-
der the agreement, to do carman's
work ahead of any man upgraded af-
ter him. It was only because of peti-
tioners' military service that men up-
graded after them were able to work
more days as carmen and to qualify
as journeymen before them (Tilt. R.
Stip). But for their absence, peti-
tioners would have qualied as jour-
neymen carmen and achieved the se-
niority dates they now claim.
There is no room for doubt in this
case that...had petitioners remained
continuously on the job during the
period of their military service, they
would have completed the work pe-
riod and qualied as journeymen in
advance of those who passed them in
seniority during their absence. Each
petitioner was entitled, under the la-
bor agreement, to do carman's work
ahead of any upgraded after him. It
was only because of petitioners' mili-
tary service that men upgraded after
them were able to work more days
as provisional carmen and to qualify
as journeymen before them. But for
their absence, petitioners would have
qualied as journeymen carmen and
achieved the seniority dates they now
claim.
124
Shared phrases such as \there is no room for doubt" intimate the parallel
strength and assuredness of the arguments in both the SG's brief and the
Court's majority opinion. In other examples, the opinion continues to val-
idate the SG's, reasoning as it uses the contentions made in the brief as
the opinion's wording. These statements in the opinion, similar to those in
Steiner make the Court's position on the issue and in particular its agree-
ment with the SG's logic abundantly clear:
124
Id. at 177; Petitioner's Brief at 25-26 Tilton v. Missouri Pacic Railroad, 376 U.S.
169 (No. 49) (1964).
169
For it was apparent that McKinney
could never have predicted \with any
degree of certainty," when he left for
service, that (1) a group 1 position
would fall vacant in his absence; (2)
that he would elect to bid for it; (3)
that he would be in adequate health
to bid for it; and (4) that he would
not have already lost his lower po-
sition because of unsatisfactory per-
formance... A returning veteran can-
not claim a promotion that depends
solely upon completing a prerequisite
period of employment training unless
he rst works that period but, upon
completion of that period of training,
he can insist upon a seniority date
in the higher position which re
ects
the delay in completing the requisite
period of training caused by military
service.
It was apparent that McKinney,
when he left for service, could not
have predicted with absolute cer-
tainty that a group position would
fall vacant in his absence; that he
would be in adequate health to bid
for it; that he would elect to bid
for it; and that he would not have
lost his lower position because of un-
satisfactory performance. A return-
ing veteran cannot claim a promo-
tion that depends solely upon satis-
factory completion of a prerequisite
period of employment training unless
he rst works that period. But upon
satisfactorily completing that period,
as petitioners did here, he can insist
upon a seniority date re
ecting the
delay caused by military service.
125
In a similar vein as in Steiner, these samples of parallel construction in the
SG's brief and in the opinion in Tilton depict language that leaves little
doubt regarding the position of the litigant or of the Court. They also
show that the Court adopted the same reasoning as the SG for ruling in the
government's favor.
The second Lifted opinion authored by Justice Warren, Foti v. INS,
126
deals
with jurisdiction under the Immigration and Nationality Act. Much more
than in Steiner, the Court in Foti adopts similar reasoning to that found in
the brief submitted by the SG. In particular, the reasoning is similar in terms
of the constructions of the relevant legislation and practice in enforcing it.
125
Id. at 177; Petitioner's Brief at 22; 34 Tilton v. Missouri Pacic Railroad, 376 U.S.
169 (No. 49) (1964).
126
375 U.S. 217 (1963)
170
From the beginning, by regulations
having the force and eect of law,
it has been exercised as an integral
part of the administrative proceed-
ings which have led to the issuance
of a nal deportation order;
Thus, the administrative discretion
to grant a suspension of deportation
has historically been consistently ex-
ercised as an integral part of the
proceedings which have led to the
issuance of a nal deportation or-
der.
127
The fundamental purpose of Sec-
tion 106(a), its legislative history dis-
closes, was to abbreviate the pro-
cess of judicial review of deporta-
tion orders in order to frustrate cer-
tain practices, which had come to
Congress's attention, whereby per-
sons subject to deportation were
forestalling enforcement by dilatory
tactics in the courts.
The fundamental purpose behind
x106(a) was to abbreviate the pro-
cess of judicial review of deportation
orders in order to frustrate certain
practices which had come to the at-
tention of Congress, whereby persons
subject to deportation were fore-
stalling departure by dilatory tactics
in the courts.
128
The last-mentioned consideration
also refutes the majority's suggestion
that it is \incredible" that Congress
meant to burden courts of appeals
with review of orders denying volun-
tary departure.
And the suggestion of the court
below that it is \incredible" that
Congress meant to burden the
Courts of Appeals with review of all
orders denying discretionary relief in
deportation cases is unconvincing.
129
Justice Marshall's opinion in Hodel v. Indiana,
130
follows the language and
reasoning of the SG's brief in a similar manner to Foti and Tilton. Hodel
looks at the constitutionality of an Indiana mining statute. Signicant to
this case are the instances where the opinion and the brief both discuss the
importance of giving deference to Congressional decisions:
127
Id. at 223; Brief for Respondent at 19 Foti v. INS, 375 U.S. 217 (No. 28) (1963).
128
Id. at 224; Brief for Respondent at 24 Foti v. INS, 375 U.S. 217 (No. 28) (1963).
129
Id. at 230; Brief for Respondent at 12 Foti v. INS, 375 U.S. 217 (No. 28) (1963).
130
452 U.S. 314 (1981)
171
A court may not substitute its judg-
ment for that of Congress merely be-
cause it believes that Congress was
\unwise in not choosing a means
more precisely related to its primary
purpose."
\This court will certainly not sub-
stitute its judgment for that of
Congress unless the relation of the
subject to interstate commerce and
its eect upon it are clearly non-
existent."
131
All of the provisions invalidated by
the district court are reasonably cal-
culated to further the legitimate con-
gressional goals of preserving the pro-
ductive capacity of mined lands, min-
imizing the adverse environmental
consequences that can result from
surface mining or inadequate recla-
mation of mined lands, and protect-
ing the public from health and safety
hazards that may be created by sur-
face mining.
All the provisions invalidated by
the court below are reasonably
calculated to further these legit-
imate goals. For example, the
approximate-original-contour re-
quirement in x515 (b)(5) is designed
to avoid the environmental and other
harm that may result from unre-
claimed or improperly restored min-
ing cuts.
132
That is surely a rational distinction
for Congress to draw, and the fact
that a particular state has more of
one kind of mining operation than
another does not establish discrimi-
nation in violation of the Due Pro-
cess Clause of the Fifth Amendment.
Congress acted rationally in drawing
these distinctions, and the fact that a
particular State has more of one kind
of mining operation than another
does not establish impermissible dis-
crimination under the Fifth Amend-
ment's Due Process Clause.
133
131
Id. at 326 (Citing Staord v. Wallace, 258 U.S. 495, 521 (1922); Petitioner's Brief
at 16 Hodel v. Indiana, 452 U.S. 314 (80-231) (1981).
132
Id. at 327-28; Petitioner's Brief at 17 Hodel v. Indiana, 452 U.S. 314 (80-231) (1981).
133
Id. at 333; Petitioner's Brief at 28 Hodel v. Indiana, 452 U.S. 314 (80-231) (1981).
172
Moreover, Congress' determination
that federal intervention is necessary
in this area was based in part on
a desire to ensure that mine op-
erators in states adhering to high
performance and reclamation stan-
dards would not be disadvantaged
in competition with their counter-
parts in states with less rigorous reg-
ulatory programs. See 30 U.S.C.
(Supp. I) 1201(g). The statutory
provisions overturned by the district
court advance these legitimate goals
of Congress and thus are rationally
related to the protection of commerce
from the adverse impact of surface
mining operations.
the Act re
ects the congressional
goal of protecting mine operators in
States adhering to high performance
and reclamation standards from dis-
advantageous competition with oper-
ators in States with less rigorous reg-
ulatory programs. See 30 U. S. C.
x1201 (g) (1976 ed., Supp. III). The
statutory provisions invalidated by
the District Court advance these le-
gitimate goals, and we conclude that
Congress acted reasonably in adopt-
ing the regulatory scheme contained
in the Act.
134
As with several other Type I cases, the Court in Hodel uses the SG's rea-
soning for rejecting the opposing parties' arguments.
Moreover, even assuming arguendo
that the provisions in question im-
pose a greater burden on mine op-
erators in midwestern states, that is
no basis for striking them down as
unconstitutional. A claim of arbi-
trariness or irrationality cannot be
founded merely upon a statute's lack
of uniform geographic impact
More important, even were appellees
correct that the challenged provi-
sions impose a greater burden on
mine operators in the Midwest, that
is no basis for nding the provi-
sions unconstitutional. A claim of
arbitrariness cannot rest solely on a
statute's lack of uniform geographic
impact. Secretary of Agriculture v.
Central Roig Rening Co., 338 U.S.
604, 616-619 (1950)
135
134
Id. at 329; Petitioner's Brief at 20 Hodel v. Indiana, 452 U.S. 314 (80-231) (1981).
135
Id. at 332; Petitioner's Brief at 27 Hodel v. Indiana, 452 U.S. 314 (80-231) (1981).
173
As in Virginia Surface Mining, plain-
tis' taking claims did not focus on
any particular properties to which
the challenged provisions have been
applied, and the district court did
not base its ruling on the denial of
a surface mining permit for specic
prime farmland operations proposed
by plaintis
In this case as in Virginia Surface
Mining, appellees' takings claims do
not focus on any particular proper-
ties to which the challenged provi-
sions have been applied. Similarly,
the District Court's ruling did not
pertain to the taking of a particu-
lar piece of property or the denial
of a mining permit for specic prime
farmland operations proposed by ap-
pellees.
136
We add only the observation that
plaintis here, like their counter-
parts in Virginia Surface Mining,
have made no showing that they were
ever assessed civil penalties under
the Act, much less that the statu-
tory prepayment requirement was
ever applied to them or that it caused
them injury.
However, like their counterparts in
Virginia Surface Mining, appellees
have made no showing that they were
ever assessed civil penalties under
the Act, much less that the statu-
tory prepayment requirement was
ever applied to them or caused them
any injury.
137
These two links to between the majority opinion and the SG's brief: through
the interpretation of Congress' purpose in enacting a particular statute and
through the rejection of the opposing parties' arguments are also present
in Justice Stewart's opinion in Lehman v. Nakshian.
138
This case deals
with the right to a jury trial in cases concerning the Age Discrimination
Employment Act. In reviewing the purpose behind the Act, the Court
shares the SG's views:
136
Id. at 334; Petitioner's Brief at 30 Hodel v. Indiana, 452 U.S. 314 (80-231) (1981).
137
Id. at 335-36; Petitioner's Brief at 36 Hodel v. Indiana, 452 U.S. 314 (80-231) (1981).
138
453 U.S. 156 (1981)
174
The general experience has been that
when Congress waives the sovereign
immunity of the United States, it
does not provide for jury trials. Such
trials historically have not been avail-
able in the broad range of cases sub-
ject to the Court of Claims' juris-
diction under U.S.C. which includes
claims arising under \any Act of
Congress."
When Congress has waived the
sovereign immunity of the United
States, it has almost always condi-
tioned that waiver upon a plainti's
relinquishing any claim to a jury
trial.
139
In any event, Rule of the Federal
Rules of Civil Procedure provides
that the right to trial by jury \as
declared by the Seventh Amendment
to the Constitution of as given by
a statute of the United States shall
be preserved to the parties inviolate"
(emphasis added). This language
certainly does not state a general rule
that jury trials are to be presumed
whenever Congress provides for cases
to be brought in the district court.
To the contrary, Rule renders it nec-
essary to look elsewhere for a spe-
cic, armative grant of the right
where, as here, the Seventh Amend-
ment does not apply.
Moreover, Rule 38(a) of the Fed-
eral Rules of Civil Procedure pro-
vides that the right to a jury trial \as
declared by the Seventh Amendment
to the Constitution or as given by
a statute of the United States shall
be preserved to the parties inviolate"
(emphasis added). This language
hardly states a general rule that jury
trials are to be presumed whenever
Congress provides for cases to be
brought in federal district courts. In-
deed, Rule (a) requires an armative
statutory grant of the right where, as
in this case, the Seventh Amendment
does not apply.
140
139
Id. at 161; Petitioner's Brief at 12 Lehman v. Nakshian, 453 U.S. 156 (80-242) (1981)
140
Id. at 164-65; Petitioner's Brief at 49 Lehman v. Nakshian, 453 U.S. 156 (80-242)
(1981)
175
As is evident, neither the provision
for cases under Section to be brought
in district court nor the use of the
word \legal" in that section can be
thought to manifest a congressional
intent that the plainti in an ADEA
action against the federal govern-
ment have a right to trial by jury.
Neither the provision for federal em-
ployer cases to be brought in dis-
trict courts rather than the Court of
Claims, nor the use of the word \le-
gal" in that section, evinces a con-
gressional intent that ADEA plain-
tis who proceed to trial against the
Federal Government may do so be-
fore a jury.
141
In its analysis of whether jury trials should be allowed in such cases, the
Court follows the SG's reasoning regarding the dearth of evidence presented
by the respondent:
There is nothing in the legislative his-
tory to indicate that Congress did
not mean what it said in providing
for jury trials in cases under Section
but not in cases against the federal
government under Section 7(c). In-
deed, the legislative history contains
not a single reference to the issue of
jury trials in federal sector cases, and
any inferences that may be drawn
from the legislative history on this
question cut against the availability
of jury trials...Indeed, any inferences
that may be drawn from the legisla-
tive history cut strongly against re-
spondent's position.
The respondent cannot point to a
single reference in the legislative his-
tory to the subject of jury trials
in cases brought against the Fed-
eral Government. There is none.
And there is nothing to indicate
that Congress did not mean what it
plainly indicated when it expressly
provided for jury trials in x7 (c) cases
but not in x7 (c) cases. In fact, the
few inferences that may be drawn
from the legislative history are in-
consistent with the respondent's po-
sition.
142
In another case dealing with the defendant's rights in jury trials, Carter
v. Kentucky,
143
Justice Stewart utilizes the SG's constitutional analysis for
141
Id. at 168; Petitioner's Brief at 59 Lehman v. Nakshian, 453 U.S. 156 (80-242) (1981)
142
Id. at 166; Petitioner's Brief at 13-14 Lehman v. Nakshian, 453 U.S. 156 (80-242)
(1981).
143
450 U.S. 288 (1981)
176
requiring specic jury instructions under the Fifth Amendment:
There is no doubt that the Fifth
Amendment privilege, the presump-
tion of innocence, and the burden of
proof are closely aligned...However,
to say that these principles are
closely aligned is not to say that
they do not serve distinctive func-
tions. Petitioner's jury would have
certainly derived \signicant addi-
tional guidance" from an instruction
that \no inference" should be drawn
from his failure to testify.
Without question, the Fifth Amend-
ment privilege and the presumption
of innocence are closely aligned. But
these principles serve dierent func-
tions, and we cannot say that the
jury would not have derived \signi-
cant additional guidance," Taylor v.
Kentucky, 436 U.S. 478, 484, from
the instruction requested.
144
In this passage, the opinion shares the strength of the SG's argument with
words that leave little room for alternative interpretations like \no doubt"
in the SG's brief and \[w]ithout question" in the opinion.
Justice Burger's opinion in United States v. One Assortment of 89 Firearms,
145
has a similar tone of certainty in its adoption of language from the SG's
brief. This is especially apparent in the Court's statutory analysis regarding
whether Congress intended for the possibility of forfeiture proceedings after
a gun owner is acquitted of criminal charges:
144
Id. at 304; Petitioner's Brief at 39-40 Carter v. Kentucky, 450 U.S. 288 (80-5060)
(1981).
145
465 U.S. 354 (1984)
177
Congress' intent that Section 924(d)
be regarded as a civil rather than a
criminal penalty is most clearly ev-
idenced, however, by the procedu-
ral mechanisms it established for en-
forcing forfeitures under the statute.
Section 924(d) does not prescribe
the steps to be followed in eec-
tuating a forfeiture, but rather in-
corporates by reference the proce-
dures of the Internal Revenue Code
of 1954. The Internal Revenue Code
provides that proceedings to enforce
forfeitures \shall be in the nature of
a proceeding in rem in the United
States District Court for the dis-
trict where such seizure is made" (26
U.S.C. 7323). As outlined above, in
rem actions are, by their very na-
ture, civil proceedings, with juris-
diction dependent upon seizure of a
physical object, in contrast with the
in personam nature of criminal ac-
tions. See Calero-Toledo, 416 U.S.
at 684.
Applying the rst prong of the Ward
test to the facts of the instant case,
we conclude that Congress designed
forfeiture under x924(d) as a reme-
dial civil sanction. Congress' intent
in this regard is most clearly demon-
strated by the procedural mecha-
nisms it established for enforcing for-
feitures under the statute. Section
924(d) does not prescribe the steps
to be followed in eectuating a forfei-
ture, but rather incorporates by ref-
erence the procedures of the Internal
Revenue Code of 1954 (Code), 26 U.
S. C. x7321-7328. The Code in turn
provides that an action to enforce a
forfeiture \shall be in the nature of
a proceeding in rem in the United
States District Court for the district
where such seizure is made." 26 U. S.
C. x7323. In contrast to the in per-
sonam nature of criminal actions, ac-
tions in rem have traditionally been
viewed as civil proceedings, with ju-
risdiction dependent upon seizure of
a physical object. See Calero-Toledo
v. Pearson Yacht Leasing Co., 416
U.S. 663, 684 (1974).
146
146
Id. at 363; Petitioner's Brief at 34 United States v. One Assortment of 89 Firearms,
465 U.S. 354 (82-1047) (1984).
178
In addition to the in rem nature
of the action, the Internal Revenue
Code provides a summary, admin-
istrative proceeding for forfeiture of
seized goods valued at $2,500 or less.
See 26 U.S.C. 7325...That Congress
provided a distinctly civil procedure
for [forfeitures under 18 U.S.C. 924
[36] (d)] indicates clearly that it in-
tended a civil, not a criminal, sanc-
tion." Helvering, 303 U.S. at 402.
In addition to establishing the in rem
nature of the action, the Code autho-
rizes a summary administrative pro-
ceeding for forfeiture of items valued
at $2,500 or less, for which notice
of a seizure may be by publication.
See 26 U. S. C. x7325. By creating
such distinctly civil procedures for
forfeitures under x924(d), Congress
has \[indicated] clearly that it in-
tended a civil, not a criminal, sanc-
tion." Helvering v. Mitchell, supra,
at 402.
147
147
Id.; Petitioner's Brief at 35 United States v. One Assortment of 89 Firearms, 465
U.S. 354 (82-1047) (1984).
179
When Congress enacted the 1968
gun control legislation, \it was con-
cerned with the widespread trac
in rearms and with their general
availability to those whose posses-
sion thereof was contrary to the pub-
lic interest." Huddleston v. United
States, 415 U.S. 814, 824 (1974).
Congress concluded that \the ease
with which rearms could be ob-
tained contributed signicantly to
the prevalence of lawlessness and vi-
olent crime in the United States."
Ibid., citing S. Rep. No. 1097, 90th
Cong., 2d Sess. 108 (1968). The
Gun Control Act of 1968, in particu-
lar, was designed to \control the in-
discriminate
ow" of rearms across
state borders and to \assist and en-
courage States and local communi-
ties to adopt and enforce stricter gun
control laws." H.R. Rep. No. 1577,
90th Cong., 2d Sess. 8 (1968)...Sec-
tion 924(d) plays an important role
in furthering the \broad prophylac-
tic purpose" of the 1968 gun control
legislation (Dickerson v. New Banner
Institute, supra, slip op. 15) by elim-
inating stocks of rearms that have
been used or intended for use outside
regulated channels of commerce.
In enacting the 1968 gun con-
trol legislation, Congress \was con-
cerned with the widespread trac
in rearms and with their general
availability to those whose posses-
sion thereof was contrary to the pub-
lic interest." Huddleston v. United
States, 415 U.S. 814, 824 (1974).
Accordingly, Congress sought to
\control the indiscriminate
ow" of
rearms and to \assist and encour-
age States and local communities to
adopt and enforce stricter gun con-
trol laws." H. R. Rep. No. 1577, 90th
Cong., 2d Sess., 8 (1968). Section
924(d) plays an important role in fur-
thering the prophylactic purposes of
the 1968 gun control legislation by
discouraging unregulated commerce
in rearms and by removing from cir-
culation rearms that have been used
or intended for use outside regulated
channels of commerce.
148
By sharing the SG's reasoning in this case the Court shows its regard for
the SG's interpretation of Congress' legislative intent and purpose to such
a degree that it is willing to appropriate many aspects of SG's argument
wholesale in the opinion. This, however, is not an anomaly in this set of
148
Id. at 364; Petitioner's Brief at 37-38 United States v. One Assortment of 89
Firearms, 465 U.S. 354 (82-1047) (1984).
180
cases. Lifted cases, almost without exception convey the strength of trust
the Court puts in the Oce of the SG. On a separate level though it also
shows great deference on the part of the Court to the SG's contentions.
The SG has insight into the inner-workings of the government not available
to other parties although even this knowledge is fallible. Unfortunately,
comparisons of briefs and opinions do not provide evidence of the Court's
level of scrutiny into the SG's contentions.
4.4 Making Sense of This Relationship
These examples present a slice of the relationship between Supreme Court
briefs and opinions. They highlight cases where the Court borrows a large
amount of substantive language from merits briefs. While there is existing
evidence that Supreme Court opinions borrow language from briefs the ex-
tent of this borrowing on the case level is uncharted territory (Corley 2008).
There are important factors that these cases illuminate regarding the re-
lationship between briefs and opinions specic to cases with high levels of
overlapping language. On the other hand, these comparisons also leave us
with unanswered questions.
In terms of supporting existing theory, these cases provide even more con-
rmation that repeat players, and the Oce of the Solicitor General in
particular, have strong relationships with the Court that makes this ex-
tent of language sharing possible (McGuire 1993, 1995; Black and Owens
2012). They also help dierentiate the types of impact a brief can have
on an opinion. There are clear dierences between cases where the opin-
ions share citations and quoted language with the briefs and where opinions
share original language that derives from the brief. The former cases present
examples of briefs that were likely in
uential to the extent that they focused
181
Court's attention on particular precedent, places in the record, and on rel-
evant statutes to assist the Court in its decision making. The Lifted cases
that are the focus of this Chapter present puzzles due not only to the extent
of the language shared, but also due to the content of the shared language.
How deeply should we read into the shared language in Lifted cases? If we
dig into other cases and probe the relationships between briefs and opinions,
there are almost assuredly more examples with this type of language shar-
ing, potentially at levels that approximate those in the cases covered in this
section. There are likely even more cases where minor editing changed the
phrasing so that language does not align perfectly between brief and opin-
ion, but where high levels of shared language predominate. Below I brie
y
examine a few of the implications from Lifted relationships.
There are many sources that the Justices or their clerks would have located
without the assistance of briefs and so in some cases it is mere happenstance
that the language is shared. As discussed in this Chapter, judges also of-
ten want briefs to organize and synthesize the information in cases and so
they may actively seek specic language used in briefs (Tate 1978). The
unattributed shared language, however, is not so readily explained by well-
organized and persuasive briefs. Opinions can and do cite to briefs as they
do to other information upon which the Justices base their opinions. In Rus-
sello,
149
for instance, the opinion cites to the petitioner's brief when it says,
\Petitioner himself has not attempted to dene the term \interest" as used
in x1963(a)(1). He insists, however, that the term does not reach money
or prots because, he says: `interest,' by denition, includes of necessity an
interest in something."
150
Similarly in National Bank of Commerce,
151
the
Court cites to the respondent's brief stating, \Common sense dictates that
a right to withdraw qualies as a right to property for purposes of xx6331
149
464 U.S. at 22 (1983)
150
Brief for Petitioner 9
151
472 U.S. At 722 (1985)
182
and 6332. In a levy proceeding, the IRS `steps into the taxpayer's shoes,'
United States v. Rodgers, 461 U. S., at 691, n. 16, quoting 4 Bittker,
x111.5.4, at 111-102; M. Saltzman, IRS Practice and Procedure x14.08, p.
14-32 (1981); Brief for Respondent 8." These examples show that even in the
Type I cases referenced in this Chapter, the Court is willing to cite to the
briefs in certain circumstances where the brief informs the opinion writer's
choice of language.
In the bulk of Lifted instances of language overlap identied in this Chapter,
the Court does not cite to a source, and yet due to the extent of shared
language in these cases and to the proximity of the Justices and the clerks
to the briefs, the likely source of the language is clear. Since clerks are
involved in the process of drafting opinions it is highly likely that some of
these choices of shared language stem from clerks' decisions. To this point,
in an interview Justice Ginsburg related the clerks' role in opinion drafting
when she said, \I would like to do all my own work so I could write my
own opinions, but there is just not enough time to do that" (Ginsburg in
Peppers and Ward (2012, 391). Even if the shared language in some of these
instances stems from a source that both the brief and the opinion share, the
lack of attribution does not change.
There are two additional items worthy of consideration based on this lack
of attribution. The rst has to do with the trustworthiness or credibility of
the source and the second has to do with the normative value we place on
courts to cite to sources when there is a clear indication that the language
did not originate with the writer.
The trustworthiness quandary may be diminished because of the source of
language in these cases. The relationship between the Solicitor General and
the Court is predicated on trust (Salokar 1994; Bailey, Kamoie and Maltz-
man 2005). This is evident from the numerous times the Court has invited
183
the Solicitor General to le amicus curiae briefs in cases where the govern-
ment is not a direct party (Krislov 1963). This trust and the reasons behind
it might allay concerns about the source of the shared language. If there is
a repeat litigator in the Court who is expected to present accurate informa-
tion and who is held to a lofty standard, that litigator is the government's
attorney - the Solicitor General. Nonetheless, the SG is also an advocate
and makes arguments to win cases. Furthermore, the government may have
an agenda in a case that does not coincide with the best interest of Justice
or of the rule of law. If the Solicitor General frames arguments to win cases,
then the language used may be stronger or more argumentative than we
would expect from an agnostic party or from the Court.
Although the Solicitor General, or any another other similarly situated liti-
gator's argument may be accurate, the level of accuracy expected from these
parties does not parallel that which is expected from the Court. Thus, there
may be certain professional and societal expectations of due diligence on the
part of the Justices and the clerks in terms of fact and law-checking that is
not equally expected from others. This is not to say that such due-diligence
does not occur in circumstances with unattributed language sharing; only
that the lack of attribution raises the question of the level of due-diligence
that is performed.
The normative question of when a citation in a Supreme Court opinion is
proper or expected is primarily untapped. There is a small body of schol-
arship looking specically at judicial plagiarism primarily focusing on the
lower courts and on copyright implications from this practice (Bast and
Samuels 2008; Dursht 1996; Lebovits, Curtin and Solomon 2008; Richmond
2013). This Chapter focuses exclusively on the Supreme Court and unlike
these other articles it focuses on the construction of such opinions rather
than on copyright issues.
184
Sitting at the apex of the judicial hierarchy, the Supreme Court and its opin-
ions are not reviewed by other judicial bodies that could fashion norms and
expectations for this practice. There are reasons why the Justices may want
to provide citations when there are clear sources, including the potential
eect it could have on views of the Court's legitimacy (Baum 2006).
It is possible that the practice of large-scale language borrowing is idiosyn-
cratic to a particular Justice or Justices. Justice Blackmun authored more
than half of the Lifted opinions and this may a part of a larger pattern of his
lack of clerk supervision towards the end of his career on the Supreme Court
(Mauro 2005). It may indicate that he was more willing to directly engage
the language of litigants than other Justices. Although this might be accu-
rate at the level of particular cases, as this Dissertation shows, the average
percentage of overlapping language between briefs and Justice Blackmun's
opinions is not high compared to the other Justices (the mean overlap value
across the 1946 through 2013 terms is 9.54% and Justice Blackmun's mean
overlap value is just above average at 9.90%). Still, many other Justices in
this time period authored Lifted opinions, and if the threshold for Lifted
opinions was loosened, this number would grow considerably.
As Epstein, Posner and Landes (2013, 52) aptly wrote, \There is much to
be learned from studies that compare the judicial opinion with the briefs
and trial transcripts and other materials on which the judge based - or
purported or was expected to base - his opinion." Awareness of the practice
of large-scale language borrowing may help bring clarity to the process of
opinion construction. Still, the rules for drafting opinions when they are
primarily based on contentions found within merits briefs are unscripted.
Supreme Court opinion writing is a unique enterprise and so it may deserve
deference not aorded to other types of writing. By peeling away a layer
of the onion obscuring the opinion construction process, however, we may
begin to develop expectations for the role briefs should play and for the
185
signicance, if any, we place on attribution to briefs when opinions adopt
their substantive language.
Chapter 5
Conclusion
This dissertation is the rst empirical attempt to trace the relationship be-
tween merits brief and Supreme Court majority opinion language across time
and to locate the cases where merits briefs play the largest role in generat-
ing decisional language. Each chapter adds a novel dimension to the study
of how merits briefs impact the language of Supreme Court opinions. The
rst substantive Chapter examines the relative importance of merits briefs,
amicus briefs, and lower court opinions on the Court's opinion construction.
The second substantive Chapter develops a measure for brief quality and
tests the eects of quality along with other theoretically relevant aspects
of briefs on Supreme Court opinion language. The nal substantive Chap-
ter develops a heuristic for locating the instances where merits briefs have
the greatest impact on Supreme Court opinions and examines each of these
instances.
186
187
5.1 Main Findings
On the aggregate, the ndings from this dissertation suggest that merits
briefs matter. The rst Chapter compares the Justices' main tools of opinion
construction: merits briefs, amicus briefs, and lower court opinions, and
presents the instances where they are most relevant. Along with breaking
down the variation in the Justices' individual opinion construction process,
the rst Chapter shows that across Justices, the Court's opinions share more
language with merits briefs than with either amicus briefs or lower court
opinions, and more language with lower court opinions than with amicus
briefs.
The second Chapter presents data on the relationship between merits brief
language and Supreme Court opinions for all Justices between 1946 and
2013. This Chapter tests a range of hypotheses relating to which merits
briefs will have the greatest impact on Supreme Court opinions and under
what circumstances. It also develops an original measure for brief writing
quality. Among the factors that have large eects on the relationship be-
tween merits briefs and the Court's opinions are case importance, attorney
experience, and brief quality.
The third Chapter locates the instances where merits briefs had the great-
est impact on Supreme Court opinions as well as determines commonalities
between such instances. The overwhelming number of instances of this ro-
bust relationship occurred prior to the 1980's and in cases where the S.G.
led a merits brief. Justice Blackmun also wrote the bulk of these opinions
re
ecting the possibility of a proclivity for such highly overlapping opinion
language.
188
5.2 Practical Application
The ndings from this dissertation are applicable to public law and legal
scholars as well as to appellate legal practitioners. By clarifying the instances
when merits briefs should play the largest role in opinion construction at-
torneys may better understand areas where they have the greatest agency
in possibly aecting the Court's opinion language. This may be especially
important to advocates who value the policy repercussions of opinions. It
also provides a better notion of how attorneys can limit damage in cases
that they are not likely to win.
These ndings are important for our understanding of all Supreme Court
merits briefs - both for winning and losing parties. For instance, even when
a party is likely to lose on the merits, the insights about increased brief
quality can benet the party on the margins by leading the Justices and
clerks to insert a greater amount of language from the brief in the opinion.
Conversely, well written briefs may also help win otherwise close cases by
focusing the Court on a particular party's argument in the merits brief.
5.3 Future Research
There are multiple directions in which research can move that builds from
insights provided by this dissertation. On one hand, there are many facets
of opinion construction that deserve closer scrutiny. For instance, what as-
pect(s) of the contents of specic briefs makes them more distinctive than
others? Do more experienced attorneys write dierently than less experi-
enced attorneys? Studies focusing on the content of briefs will help propel
this research forward and may examine whether certain successful attorneys
can attribute at least some of this success to distinctive writing patterns.
189
There are also other ways to leverage the methodological techniques from
this dissertation to better pinpoint important instances of consequential
language overlap. Approaches like the one utilized in the Third Chapter
that help locate the most relevant instances of language overlap could be
used in other situations. One example may be comparing briefs within the
same case. This could help locate the specic briefs that play the largest
roles in particular instances of opinion construction.
As much as this dissertation helps to ll large gaps in our understandings of
opinion construction, it also shows that there is still much room left to study
additional characteristics that aect opinion construction both methodolog-
ically and by testing additional characteristics of court documents.
Bibliography
Aldisert, Ruggero J. 1992. Winning on Appeal: Better Briefs and Oral
Argument. Natl Inst for Trial Advocacy.
Aldisert, Ruggero J. 1996. Winning on Appeal: Better Briefs and Oral
Argument. National Institute for Trial Advocacy.
Alito Jr, Samuel A, Douglas W Kmiec, Carter G Phillips and Kenneth W
Starr. 2012. \The Inaugural William French Smith Memorial Lecture: A
Look at Supreme Court Advocacy with Justice Samuel Alito." Pepperdine
Law Review 35(5):1.
Alito Jr, Samuel A, Michael W McConnell, Kenneth W Starr and Walter E
Dellinger. 2009. \Second Conversation with Justice Samuel A. Alito, Jr.:
Lawyering and the Craft of Judicial Opinion Writing, The." Pepp. L. Rev.
37:33.
Anderson, Helen A. 2010. \Changing Fashions in Advocacy: 100 Years of
Brief-Writing Advice." J. App. Prac. & Process 11:1.
Baccianella, Stefano, Andrea Esuli and Fabrizio Sebastiani. 2010. Senti-
WordNet 3.0: An Enhanced Lexical Resource for Sentiment Analysis and
Opinion Mining. In LREC. Vol. 10 pp. 2200{2204.
Bailey, Michael A, Brian Kamoie and Forrest Maltzman. 2005. \Signals from
the tenth justice: The political role of the solicitor general in Supreme
190
Bibliography 191
Court decision making." American Journal of Political Science 49(1):72{
85.
Baldock, Robert R, Carlos F Lucero and Vicki Mandell-King. 2001. \What
Appellate Advocates Seek from Appellate Judges and What Appellate
Judges Seek from Appellate Advocates." NML Rev. 31:265{651.
Bast, Carol M and Linda B Samuels. 2008. \Plagiarism and legal scholar-
ship in the age of information sharing: the need for intellectual honesty."
Catholic University Law Review 57:777.
Baum, Lawrence. 2006. \Judges and their audiences.".
Baum, Lawrence. 2009a. Judges and their audiences: A perspective on ju-
dicial behavior. Princeton University Press.
Baum, Lawrence. 2009b. The puzzle of judicial behavior. University of
Michigan 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{550.
Benson, Robert W and Joan B Kessler. 1986. \Legalese v. Plain English: An
Empirical Study of Persuasion and Credibility in Appellate Brief Writing."
Loy. LAL Rev. 20:301.
Black, Ryan C and JF Spriggs. 2008. \An Empirical Analysis of the Length
of US Supreme Court Opinions." Houston L. Rev. 45:621{634.
Black, Ryan C and Ryan J Owens. 2012. The Solicitor General and the
United States Supreme Court: Executive Branch In
uence and Judicial
Decisions. Cambridge University Press.
Black, Ryan C and Ryan J Owens. 2013. \A Built-In Advantage The Oce
of the Solicitor General and the US Supreme Court." Political Research
Quarterly 66(2):454{466.
Bibliography 192
Bloomeld, Louis. 2014. \WCopyFind 4.1.3.".
URL: plagiarism.bloomeldmedia.com/z-wordpress/software/
wcopynd/
Box-Steensmeier, Janet M, Dino P Christenson and Matthew P Hitt. 2013.
\Quality over quantity: Amici in
uence and judicial decision making."
American Political Science Review 107(03):446{460.
Cameron, Charles M, Jerey A Segal and Donald Songer. 2000. \Strate-
gic auditing in a political hierarchy: An informational model of the
Supreme Court's certiorari decisions." American Political Science Review
94(01):101{116.
Carter, Lief H and Thomas F Burke. 2007. \Reason in Law.".
Carter, Lief H and Thomas Frederick Burke. 1998. Reason in law. Longman.
Choi, Stephen J and G Mitu Gulati. 2004. \Which judges write their opin-
ions (and should we care)." Fla. St. UL Rev. 32:1077.
Cohen, Jonathan M. 2009. Inside Appellate Courts: The Impact of Court
Organization on Judicial Decision Making in the United States Courts of
Appeals. University of Michigan Press.
Coleman, Brady and Quy Phung. 2010. \Language of Supreme Court Briefs:
A Large-Scale Quantitative Investigation, The." J. App. Prac. & Process
11:75.
Collins, Paul M. 2004. \Friends of the court: Examining the in
uence
of amicus curiae participation in US Supreme Court litigation." Law &
Society Review 38(4):807{832.
Collins, Paul M. 2007. \Lobbyists before the US Supreme Court Investigat-
ing the In
uence of Amicus Curiae Briefs." Political Research Quarterly
60(1):55{70.
Bibliography 193
Collins, Paul M., Pamela C. Corley and Jesse Hamner. 2015. \The In
uence
of Amicus Curiae Briefs on U.S. Supreme Court Opinion Content." Law
& Society Review 49(4):917{944.
Colton, Charles Caleb. 1837. Lacon: or, Many things in few words: ad-
dressed to those who think. Longman, Orme, Brown, Green, & Longmans.
Cooper, Frank E. 1953. \Stating the Issue in Appellate Briefs: A Matter of
Legal Strategy." American Bar Association Journal pp. 13{16.
Corley, Pamela C. 2008. \The Supreme Court and Opinion Content The
In
uence of Parties' Briefs." Political Research Quarterly 61(3):468{478.
Corley, Pamela C and Justin Wedeking. 2014. \The (Dis) Advantage of
Certainty: The Importance of Certainty in Language." Law & Society
Review 48(1):35{62.
Corley, Pamela C, Paul M Collins and Bryan Calvin. 2011. \Lower Court
In
uence on US Supreme Court Opinion Content." The Journal of Politics
73(01):31{44.
Corley, Pamela, Paul M Collins and Jesse Hamner. 2013. The in
uence of
amicus curiae briefs on US Supreme Court opinion content. In APSA 2013
Annual Meeting Paper.
Cozine, R Kirkland. 1994. \Emergence of Written Appellate Briefs in the
Nineteenth-Century United States, The." Am. J. Legal Hist. 38:482.
Crossley, Scott A, Jerry Greeneld and Danielle S McNamara. 2008. \As-
sessing text readability using cognitively based indices." Tesol Quarterly
42(3):475{493.
Davis, John W. 1940. \Argument of an Appeal, The." The A.B.A. Journal
26:895.
Bibliography 194
De Marnee, Marie-Catherine, Bill MacCartney, Christopher D Manning
et al. 2006. Generating typed dependency parses from phrase structure
parses. In Proceedings of LREC. Vol. 6 pp. 449{454.
De Marnee, Marie-Catherine and Christopher D Manning. 2008. The Stan-
ford typed dependencies representation. In Coling 2008: Proceedings of
the workshop on Cross-Framework and Cross-Domain Parser Evaluation.
Association for Computational Linguistics pp. 1{8.
Diederich, Joachim, J org Kindermann, Edda Leopold and Gerhard Paass.
2003. \Authorship attribution with support vector machines." Applied
intelligence 19(1-2):109{123.
Duncan, Sarah B. 1998. \Pursuing Quality: Writing a Helpful Brief." .
Mary's LJ 30:1093.
Dursht, Jaime S. 1996. \Judicial plagiarism: It may be fair use but is it
ethical." Cardozo L. Rev. 18:1253.
Dworkin, Ronald. 1978. Taking rights seriously. Vol. 136 Harvard University
Press.
Dwyer, Stephen J, Leonard J Feldman and Ryan P McBridge. 2007. \How
to Write, Edit, and Review Persuasive Briefs: Seven Guidelines from One
Judge and Two Lawyers." Seattle UL Rev. 31:417.
Edwards, Harry T. 1992. \The growing disjunction between legal education
and the legal profession." Michigan Law Review pp. 34{78.
Ehrenberg, Suzanne. 2003. \Embracing the Writing-Centered Legal Pro-
cess." Iowa L. Rev. 89:1159.
Ehrenberg, Suzanne. 2004. \Embracing the Writing-Centered Legal Pro-
cess." Iowa Law Review 89:1159.
Bibliography 195
Epstein, Lee, Andrew D Martin, Jerey A Segal and Chad Westerland.
2007. \The Judicial Common Space." Journal of Law, Economics, and
Organization 23(2):303{325.
Epstein, Lee and Jack Knight. 1998. The Choices Justices Make. SAGE.
Epstein, Lee and Jerey A Segal. 2000a. \Measuring issue salience." Amer-
ican Journal of Political Science pp. 66{83.
Epstein, Lee and Jerey A Segal. 2000b. \Measuring Issue Salience." Amer-
ican Journal of Political Science 44(1):66{83.
Epstein, Lee and Joseph Kobylka. 1992. Supreme Court and Legal Change:
Abortion and the Death Penalty: Abortion and the Death Penalty. Univ
of North Carolina Press.
Epstein, Lee, William M Landes and Richard A Posner. 2010. \Inferring
the winning party in the Supreme Court from the pattern of questioning
at oral argument." The Journal of Legal Studies 39(2):433{467.
Epstein, Lee, William M Landes and Richard A Posner. 2012a. \Are Even
Unanimous Decisions in the United States Supreme Court Ideological."
Nw. UL Rev. 106:699.
Epstein, Lee, William M Landes and Richard A Posner. 2012b. \How busi-
ness fares in the supreme court." Minn. L. Rev. 97:1431.
Epstein, Lee, William M Landes and Richard A Posner. 2013. The Behavior
of Federal Judges. Cambridge: Harvard University Press.
Evans, Michael, Wayne McIntosh, Jimmy Lin and Cynthia Cates. 2007. \Re-
counting the Courts? Applying automated content analysis to enhance
empirical legal research." Journal of Empirical Legal Studies 4(4):1007{
1039.
Bibliography 196
Feldman, Adam. 2015a. \Blurred Lines: Merits Briefs as Templates for
Supreme Court Opinions." Available at SSRN 2679625 .
Feldman, Adam. 2015b. \A Brief Assessment of Supreme Court Opinion
Language." Available at SSRN 2574451 .
Flammer, Sean. 2010. \Persuading judges: An empirical analysis of writing
style, persuasion, and the use of plain English." Legal Writing: J. Legal
Writing Inst. 16:183.
Frederick, David C. 2005. \Supreme Court advocacy in the early nineteenth
century." Journal of Supreme Court History 30(1):1{16.
Galanter, Marc. 1974. \Why the" haves" come out ahead: Speculations on
the limits of legal change." Law and society review pp. 95{160.
Gardner, James A. 2007. Legal argument: the structure and language of
eective advocacy. LexisNexis.
Garner, Bryan A. 2010a. \Chief Justice John G. Roberts Jr." The Scribes
Journal of Legal Writing 13:5.
Garner, Bryan A. 2010b. \Interviews with United States Supreme Court
Justices." Scribes J. Legal Writing 13:1{182.
Garner, Bryan A. 2010c. \Justice Anthony Kennedy." Scribes J. Leg. Writ-
ing 13:79{98.
Garner, Bryan A. 2010d. \Justice Antonin Scalia." Scribes J. Leg. Writing
13:51{67.
Garner, Bryan A. 2013. \Interviews with United States Court of Appeals
Judges." Scribes J. Legal Writing 15:1{194.
Garrow, David J. 1998. \Lowest Form of Animal Life?: Supreme Court
Clerks and Supreme Court History.".
Bibliography 197
Gillman, Howard. 1999. \The Court as an idea, not a building (or a
game): Interpretive institutionalism and the analysis of Supreme Court
decision-making." Supreme Court decision-making: New institutionalist
approaches 65:90.
Ginsburg, Ruth Bader. 1985. \Obligation to Reason Why, The." U. Fla. L.
Rev. 37:205.
Ginsburg, Ruth Bader. 1992. \Speaking in a judicial voice." NYUL Rev.
67:1185.
Ginsburg, Ruth Bader. 1998. \Remarks on appellate advocacy." SCL Rev.
50:567.
Grimmer, Justin. 2010. \A Bayesian hierarchical topic model for political
texts: Measuring expressed agendas in Senate press releases." Political
Analysis 18(1):1{35.
Haire, Susan B, Stefanie A Lindquist and Donald R Songer. 2003. \Appellate
court supervision in the federal judiciary: A hierarchical perspective." Law
& Society Review 37(1):143{168.
Hall, Kermit L, James W Ely and Joel B Grossman. 2005. The Oxford
Companion to the Supreme Court of the United States. Oxford University
Press.
Hamilton, Clyde H. 1998. \Eective Appellate Brief Writing." SCL Rev.
50:581.
Hansford, Thomas G and James F Spriggs. 2006. The politics of precedent
on the US Supreme Court. Princeton University Press.
Hart, Herbert Lionel Adolphus. 1961. The concept of law. Oxford University
Press.
Bibliography 198
Hasko, John J. 2002. \Persuasion in the Court: Nonlegal Materials in US
Supreme Court Opinions." Law Libr. J. 94:427.
Hathaway, Oona A. 2001. \Path dependence in the law: The course and
pattern of legal change in a common law system." The Iowa Law Review
86(2).
Hellman, Arthur D. 1982. \Error correction, lawmaking, and the Supreme
Court's exercise of discretionary review." U. Pitt. L. Rev. 44:795.
Hessler, Stephen E. 2000. \Establishing Inevitability Without Active Pur-
suit: Dening the Inevitable Discovery Exception to the Fourth Amend-
ment Exclusionary Rule." Michigan Law Review pp. 238{278.
Hillard, Dustin, Stephen Purpura and John Wilkerson. 2008. \Computer-
assisted topic classication for mixed-methods social science research."
Journal of Information Technology & Politics 4(4):31{46.
Hughey, Rachel Clark. 2010. \Eective Appellate Advocacy before the Fed-
eral Circuit: A Former Law Clerk's Perspective." J. App. Prac. & Process
11:401.
Huisman, Matthew. 2013. \Billing Snapshot." National Law Journal
35(38):11.
Israel, Jerold H. 1963. \Gideon v. Wainwright: The" Art" of Overruling."
The Supreme Court Review 1963:211{272.
Joachims, Thorsten. 1998. Text categorization with support vector machines:
Learning with many relevant features. Springer.
Johnson, Timothy R. 2001. \Information, oral arguments, and Supreme
Court decision making." American Politics Research 29(4):331{351.
Johnson, Timothy R. 2004. Oral arguments and decision making on the
United States Supreme Court. SUNY Press.
Bibliography 199
Johnson, Timothy R, Paul J Wahlbeck and James F Spriggs. 2006. \The
in
uence of oral arguments on the US Supreme Court." American Political
Science Review 100(1):99{113.
Johnson, Timothy R, Ryan C Black, Jerry Goldman and Sarah A Treul.
2009. \Inquiring Minds Want to Know: Do Justices Tip Their Hands
with Questions at Oral Argument in the US Supreme Court." Wash. UJL
& Pol'y 29:241.
Johnson, Timothy R, Ryan C Black and Justin Wedeking. 2009. \Pardon the
Interruption: An Empirical Analysis of Supreme Court Justices' Behavior
during Oral Arguments." Loy. L. Rev. 55:331.
Kainen, James L. 1992. \The Impeachment Exception to the Exclusionary
Rules: Policies, Principles, and Politics." Stanford Law Review pp. 1301{
1372.
Kearney, Joseph D and Thomas W Merrill. 2000. \The in
uence of amicus
curiae briefs on the Supreme Court." University of Pennsylvania Law
Review pp. 743{855.
Kenny, David A. 1996. \Models of non-independence in dyadic research."
Journal of Social and Personal Relationships 13(2):279{294.
Kenny, David A, Deborah A Kashy and William L Cook. 2006. Dyadic data
analysis. Guilford Press.
Klein, Larry A. 2010. \The Evolved Appellate Brief." Litigation pp. 38{41.
Knight, Jack and Lee Epstein. 1996. \The norm of stare decisis." American
Journal of Political Science pp. 1018{1035.
Kravitz, Mark R. 2009. \Written and Oral Persuasion in the United States
Courts: A District Judge's Perspective on Their History, Function, and
Future." J. App. Prac. & Process 10:247.
Bibliography 200
Krislov, Samuel. 1963. \The amicus curiae brief: From friendship to advo-
cacy." Yale Law Journal pp. 694{721.
Kunda, Ziva. 1990. \The case for motivated reasoning." Psychological bul-
letin 108(3):480.
Kundi, Fazal Masud, Shakeel Ahmad, Aurangzeb Khan and Muham-
mad Zubair Asghar. 2014. \Detection and Scoring of Internet Slangs
for Sentiment Analysis Using SentiWordNet." Life Science Journal 11(9).
Lamb, Brian, Susan Swain, Mark Farkas et al. 2011. The Supreme Court: A
C-SPAN Book Featuring the Justices in Their Own Words. PublicAairs.
Lasky, Moses. 1961. \Observing Appellate Opinions from Below the Bench."
California Law Review pp. 831{844.
Lax, Jerey R and Charles M Cameron. 2007. \Bargaining and opinion
assignment on the US Supreme Court." Journal of Law, Economics, and
Organization 23(2):276{302.
Lazarus, Richard James. 2008. \Advocacy Matters Before and Within
the Supreme Court: Transforming the Court by Transforming the Bar."
Georgetown Law Journal 96.
Lebovits, Gerald, Alifya V Curtin and Lisa Solomon. 2008. \Ethical Judicial
Opinion Writing." Geo. J. Legal Ethics 21:237.
Lindquist, Stefanie A and David E Klein. 2006. \The in
uence of jurispru-
dential considerations on Supreme Court decisionmaking: A study of con-

ict cases." Law & Society Review 40(1):135{162.
Liptak, Adam. 2015. \Clarence Thomas, a Supreme Court Justice of Few
Words, Some Not His Own." The New York Times p. A8.
Llewellyn, Karl N. 1962. \A Lecture on Appellate Advocacy." The Univer-
sity of Chicago Law Review pp. 627{639.
Bibliography 201
Long, Lance N and William F Christensen. 2011. \Does the Readability of
Your Brief Aect Your Chance of Winning an Appeal." J. App. Prac. &
Process 12:145.
Long, Lance N and William F Christensen. 2012. \When Justices (Subcon-
sciously) Attack: The Theory of Argumentative Threat and the Supreme
Court." Or. L. Rev. 91:933.
Long, Lance N and William F Christensen. 2013. \When Justices (Subcon-
sciously) Attack: The Theory of Argumentative Threat and the Supreme
Court." Or. L. Rev. 91:933{961.
Lupu, Yonatan and James H Fowler. 2013. \Strategic Citations to Precedent
on the US Supreme Court." The Journal of Legal Studies 42(1):151{186.
Maltzman, Forrest, James F Spriggs and Paul J Wahlbeck. 2000. Crafting
law on the Supreme Court: The collegial game. Cambridge University
Press.
Manning, Christopher D and Hinrich Sch utze. 1999. Foundations of statis-
tical natural language processing. MIT press.
Margolis, Ellie. 1999. \Beyond Brandeis: Exploring the Uses of Non-Legal
Materials in Appellate Briefs." USFL Rev. 34:197.
Martin, Andrew D and Kevin M Quinn. 2002. \Dynamic ideal point esti-
mation via Markov chain Monte Carlo for the US Supreme Court, 1953{
1999." Political Analysis 10(2):134{153.
Martineau, Robert J. 1985. Fundamentals of modern appellate advocacy.
Lawyers Co-operative Pub. Co.
Martineau, Robert J. 1986. \Value of Appellate Oral Argument: A Chal-
lenge to the Conventional Wisdom, The." Iowa L. Rev. 72:1.
Bibliography 202
Marvell, Thomas B. 1978. Appellate courts and lawyers: Information gath-
ering in the adversary system. Number 4 Greenwood Press Westport,
CT.
Mauro, Tony. 2005. \Lifting the Veil: Justice Blackmun's Papers and the
Public Perception of the Supreme Court." Mo. L. Rev. 70:1037.
McCarty, L Thorne. 2007. Deep semantic interpretations of legal texts. In
Proceedings of the 11th international conference on Articial intelligence
and law. ACM pp. 217{224.
McGuire, K, Georg Vanberg and A Yanus. 2007. Targeting the median jus-
tice: A content analysis of legal arguments and judicial opinions. In annual
meeting of the Southern Political Science Association, New Orleans.
McGuire, Kevin T. 1993. The Supreme Court bar: Legal elites in the Wash-
ington community. University of Virginia Press.
McGuire, Kevin T. 1995. \Repeat players in the Supreme Court: The
role of experienced lawyers in litigation success." The Journal of Politics
57(01):187{196.
Miner, Roger J. 1992. \Twenty-Five Dos for Appellate Brief Writers."
Scribes J. Leg. Writing 3:19.
Monaghan, Henry P. 1979. \Taking Supreme Court Opinions Seriously."
Md. L. Rev. 39:1.
Morris, Thomas R. 1986. \States before US Supreme Court: State Attorneys
General as Amicus Curiae." Judicature 70:298.
Moss, Scott A. 2013. \Bad Briefs, Bad Law, Bad Markets: Documenting the
Poor Quality of Plaintis' Briefs, Its Impact on the Law, and the Market
Failure It Re
ects." Emory LJ 63:59.
Bibliography 203
O'Brien, David M. 2006. \More Smoke Than Fire: The Rehnquist Court's
Use of Comparative Judicial Opinions and Law in the Construction of
Constitutional Rights." JL & Pol. 22:83.
Oldfather, Chad M, Joseph P Bockhorst and Brian P Dimmer. 2012. \Tri-
angulating Judicial Responsiveness: Automated Content Analysis, Judi-
cial Opinions, and the Methodology of Legal Scholarship." Fla. L. Rev.
64:1189.
Owens, Ryan J and David A Simon. 2011. \Explaining the Supreme Court's
Shrinking Docket." Wm. & Mary L. Rev. 53:1219.
Owens, Ryan J and Justin P Wedeking. 2011. \Justices and legal clarity:
Analyzing the complexity of US supreme court opinions." Law & Society
Review 45(4):1027{1061.
Owens, Ryan J and Justin Wedeking. 2012. \Predicting drift on politically
insulated institutions: A study of ideological drift on the United States
supreme court." The Journal of Politics 74(02):487{500.
Pacelle Jr., Richard L. 2003. Between law and politics: The solicitor gen-
eral and the structuring of race, gender, and reproductive rights litigation.
Texas A&M University Press.
Palmer, Barbara. 2001. \Bermuda Triangle{The Cert Pool and Its In
uence
over the Supreme Court's Agenda, The." Const. Comment. 18:105.
Pang, Xun, Barry Friedman, Andrew D Martin and Kevin M Quinn. 2012.
\Endogenous Jurisprudential Regimes." Political Analysis 20(4):417{436.
Peck, Girvan. 1983. \Strategy of the Brief." Litigation 10:26.
Peppers, Todd C. 2006. Courtiers of the Marble Palace: The Rise and
In
uence of the Supreme Court Law Clerk. Stanford University Press.
Bibliography 204
Peppers, Todd C and Artemus Ward. 2012. In Chambers: Stories of Supreme
Court Law Clerks and Their Justices. University of Virginia Press.
Peppers, Todd C and Christopher Zorn. 2008. \Law Clerk In
uence on
Supreme Court Decision Making: An Empirical Assessment." DePaul L.
Rev. 58:51.
Phillips, Carter G. 1998. \Advocacy Before the United States Supreme
Court." TM Cooley L. Rev. 15:177.
Pitler, Emily and Ani Nenkova. 2008. Revisiting Readability: A Unied
Framework for Predicting Text Quality. In Proceedings of the Conference
on Empirical Methods in Natural Language Processing. Association for
Computational Linguistics pp. 186{195.
Posner, Richard A. 1993. Cardozo: A study in reputation. University of
Chicago Press.
Posner, Richard A. 1998. \Convincing a Federal Court of Appeals." Litiga-
tion 25:3.
Posner, Richard A. 2013a. \Judicial Opinions and Appellate Advocacy in
Federal Courts-One Judge's Views." Duq. L. Rev. 51:3.
Posner, Richard A. 2013b. Re
ections on Judging. Harvard University Press.
Posner, Richard A and Albert H Yoon. 2010. \What Judges Think of the
Quality of Legal Representation." Stan. L. Rev. 63:317.
Pregerson, Harry and Suzianne D Painter-Thorne. 2008. \Seven Virtues of
Appellate Brief Writing: An Update from the Bench, The." Sw. L. Rev.
38:221.
Prettyman, E Barrett. 1953. \Some Observations Concerning Appellate
Advocacy." Virginia Law Review pp. 285{302.
Bibliography 205
Quinn, Kevin M, Burt L Monroe, Michael Colaresi, Michael H Crespin
and Dragomir R Radev. 2010. \How to analyze political attention with
minimal assumptions and costs." American Journal of Political Science
54(1):209{228.
Rehnquist, William H. 1983. \Oral Advocacy: A Disappearing Art." Mercer
L. Rev. 35:1015.
Rehnquist, William H. 1999. \From Webster to Word-Processing: The As-
cendance of the Appellate Brief." J. App. Prac. & Process 1:1.
Richards, Mark J and Herbert M Kritzer. 2002. \Jurisprudential regimes
in Supreme Court decision making." American Political Science Review
96(02):305{320.
Richmond, Douglas R. 2013. \Unoriginal Sin: The Problem of Judicial
Plagiarism." Ariz. St. LJ 45:1077.
Ricks, Sarah E and Jane L Istvan. 2007. \Eective Brief Writing Despite
High Volume Practice: Ten Misconceptions That Result in Bad Briefs."
University of Toledo Law Review 38:1113.
Robbins, Kristen K. 2002. \Inside Scoop: What Federal Judges Really Think
about the Way Lawyers Write, The." Legal Writing: J. Legal Writing Inst.
8:257.
Roberts, John G. 2005. \Oral Advocacy and the Re-emergence of a Supreme
Court Bar." Journal of Supreme Court History 30(1):68{81.
Rohde, David W and Harold J Spaeth. 1976. Supreme Court decision mak-
ing. WH Freeman.
Rose, Daniel E and Richard K Belew. 1989. Legal information retrieval a
hybrid approach. In Proceedings of the 2nd international conference on
Articial intelligence and law. ACM pp. 138{146.
Bibliography 206
Rutledge, Wiley. 1942. Appellate Brief, The. In Dicta. Vol. 19 HeinOnline
p. 109.
Salokar, Rebecca Mae. 1994. The solicitor general: The politics of law.
Temple University Press.
Segal, Jerey A. 1988. \Amicus curiae briefs by the solicitor general during
the Warren and Burger Courts: A research note." The Western Political
Quarterly pp. 135{144.
Segal, Jerey A and Albert D Cover. 1989. \Ideological Values and the
Votes of US Supreme Court Justices." American Political Science Review
83(02):557{565.
Segal, Jerey A and Harold J Spaeth. 1996. \The in
uence of stare decisis
on the votes of United States Supreme Court justices." American Journal
of Political Science pp. 971{1003.
Segal, Jerey A and Harold J Spaeth. 2002. The Supreme Court and the
Attitudinal Model Revisited. Cambridge University Press.
Shapiro, Martin. 1962. \Supreme Court and Constitutional Adjudication:
Of Politics and Neutral Principles." Geo. Wash. L. Rev. 31:587.
Shapiro, Martin. 1968. The Supreme Court and administrative agencies.
Vol. 2 Free Press New York.
Shapiro, Martin. 1981. Courts: a comparative and political analysis. Uni-
versity of Chicago Press.
Sheehan, Reginald S, William Mishler and Donald R Songer. 1992. \Ide-
ology, Status, and The Dierential Success of Direct Parties Before the
Supreme Court." American Political Science Review 86(02):464{471.
Silverstein, Gordon. 2009. Law's allure: how law shapes, constrains, saves,
and kills politics. Cambridge University Press.
Bibliography 207
Simon, Dan. 1998. \Psychological Model of Judicial Decision Making, A."
Rutgers LJ 30:1.
Smith, Christopher E. 1996. \Bent on Original Intent." ABAJ 82:48.
Songer, Donald R, Jerey A Segal and Charles M Cameron. 1994. \The
hierarchy of justice: Testing a principal-agent model of Supreme Court-
circuit court interactions." American Journal of Political Science pp. 673{
696.
Songer, Donald R and Reginald S Sheehan. 1993. \Interest group success
in the courts: Amicus participation in the Supreme Court." Political Re-
search Quarterly 46(2):339{354.
Souter, David, Ruth Bader Ginsburg, David S Tatel and Linda Greenhouse.
2010. \The Supreme Court and Useful Knowledge: Panel Discussion."
Proceedings of the American Philosophical Society pp. 294{306.
Spaeth, Harold J. 2013. The Supreme Court Database. Center for Empirical
Research in the Law at Washington University.
Stark, Steven. 1983. \Why Lawyers Can't Write." Harv. L. Rev 97:1389.
Starr, Kenneth W. 2005. \Supreme Court and Its Shrinking Docket: The
Ghost of William Howard Taft, The." Minn. L. Rev. 90:1363.
Stern, Robert L. 1981. Appellate Practice in the United States. Bureau of
National Aairs.
Sunstein, Cass R. 2001. One case at a time: Judicial minimalism on the
Supreme Court. Harvard University Press.
Swain, Susan. 2013. The Supreme Court: A C-SPAN Book, Featuring the
Justices in their Own Words. PublicAairs.
Sweet, Alec Stone. 1999. \Judicialization and the Construction of Gover-
nance." Comparative Political Studies 32(2):147{184.
Bibliography 208
Tate, Albert. 1978. \The Art of Brief Writing: What a Judge Wants to
Read." Litigation pp. 11{57.
Unah, Isaac. 2010. The supreme court in American politics. Palgrave
Macmillan.
Wedeking, Justin. 2010. \Supreme Court Litigants and Strategic Framing."
American Journal of Political Science 54(3):617{631.
Weiden, David L and Artemus Ward. 2006. Sorcerers' apprentices: 100
years of law clerks at the United States Supreme Court. NYU Press.
White, James Boyd. 1995. \What's an Opinion for?" The University of
Chicago Law Review pp. 1363{1369.
Wiener, Frederick Bernays. 2004. Eective Appellate Advocacy. American
Bar Association.
Woodward, Bob and Scott Armstrong. 2011. The brethren: inside the
Supreme Court. Simon and Schuster.
Wyner, Adam, Raquel Mochales-Palau, Marie-Francine Moens and David
Milward. 2010. Approaches to text mining arguments from legal cases.
Springer.

Abstract (if available)

Abstract Merits briefs are litigating parties' main means to convey their construction of their cases in court. In the Supreme Court, they provide the Justices with consolidated arguments and assessments of each case. In doing so they are considered one of the most important inputs for the Justices' opinions and specifically for opinion language. While anecdotal evidence supports this point, minimal empirical scholarship is devoted to verifying this premise. This dissertation is designed to look expansively at the relationship between Supreme Court merits briefs and opinions and in so doing it is designed to fill empirical gaps in our understanding of the inputs into Supreme Court opinions. It also looks specifically at the instances where merits briefs played the largest role in defining the Court's opinion language. It is equally designed to broaden the understanding of the relationship between briefs and opinions with the addition of new indicators, variables, and data, and by clarifying the mechanisms involved in the process linking briefs and opinions. This dissertation is broken into three parts based on interrelated analytic goals: examining the relative impact of merits briefs on opinion language, looking at the factors and mechanisms that are involved in this relationship, and examining cases where merits briefs have the largest impact on Supreme Court opinion language. This dissertation uses an original, large-N dataset of over 9,000 briefs and their respective opinions to test hypotheses regarding factors that make merits briefs more or less impactful on the Court's opinions.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Neither purse nor sword: strategic minority opinions and judicial policy-making

PDF

Supreme Court decision-making in slavery cases

PDF

The Supreme Court and the governing regime in political time

PDF

The separation of powers and the Supreme Court: a new institutional analysis of inter-branch disputes, 1946-2005

PDF

The new dynamics of bureaucratic autonomy: courts, litigation, and agencies in the modern American state

PDF

Death and politics in the United States Courts of Appeals

PDF

Legitimizing self-determination: advancing the sovereignty of separatist movements

PDF

Things fall apart: the unraveling of international institutions through withdrawal

PDF

Creating destruction: the political economy of zombie firms

PDF

Changing political attitudes and behavior in a diverse America: incorporating individual and contextual determinants

PDF

The social making of authoritarian environmentalism: protest-litigation nexus and policy changes in China

PDF

California's adaptation to sea level rise: incorporating environmental justice communities along the California coastline

PDF

Pathways to nuclear weapon reversal: exploring mechanisms and understanding non-proliferation policy

PDF

Public opinion and international affairs: a multi-method approach to foreign policy attitudes

PDF

Selling virtue: how human rights NGOs and their donors work together to create a better world ... for themselves

PDF

The political psychology of democracy promotion, or how democracies democratically promote democracy promotion

PDF

Deciding modalities of global health governance: What facilitates or hinders public-private partnerships?

PDF

Visual reasoning: representations in global politics, economics, and society

PDF

The IO contestation project: a multi-method approach towards understanding the ways states contest international organizations.

PDF

Analysis and prediction of malicious users on online social networks: applications of machine learning and network analysis in social science

Asset Metadata

Creator Feldman, Adam (author)

Core Title The role of merits briefs in the generation of Supreme Court decisional language

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Political Science and International Relations

Publication Date 04/20/2017

Defense Date 03/01/2017

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Justices,Language,litigation,merits briefs,OAI-PMH Harvest,opinions,Supreme Court

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Sandholtz, Wayne (committee chair), Barnes, Jeb (committee member), Klerman, Daniel (committee member), Simon, Dan (committee member)

Creator Email adfeld@gmail.com,asfeldma@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-363726

Unique identifier UC11258218

Identifier etd-FeldmanAda-5247.pdf (filename),usctheses-c40-363726 (legacy record id)

Legacy Identifier etd-FeldmanAda-5247.pdf

Dmrecord 363726

Document Type Dissertation

Rights Feldman, Adam

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA