Über die Autorin bzw. den Autor

Thomas H. Hammond is Professor of Political Science at Michigan State University. Reginald S. Sheehan is Professor of Political Science at Michigan State University. Chris W. Bonneau is Assistant Professor of Political Science at the University of Pittsburgh.

Von der hinteren Coverseite

Aus dem Klappentext

Auszug. © Genehmigter Nachdruck. Alle Rechte vorbehalten.

Strategic Behavior and Policy Choice on the U.S. Supreme Court

Stanford University Press

Contents

List of Figures and Tables...........................................................................................xiPreface..............................................................................................................xvAcknowledgments......................................................................................................xixPART I: THEORIES OF SUPREME COURT DECISION-MAKING1 Introduction.......................................................................................................1Plan of the Book.....................................................................................................62 Seven Distinctions in the Literature...............................................................................8Distinction 1: A Psychological Metaphor vs. a Rational-Choice Metaphor...............................................8Distinction 2: Theories of Attitude Activation vs. Theories of Rational Choice.......................................11Distinction 3: Theories of Choice vs. Theories of Measurement........................................................14Distinction 4: Explaining Final Votes vs. Explaining What Final Opinion Is Adopted...................................15Distinction 5: Explaining Just the Final Vote vs. Explaining All Five Stages of Decision-Making......................17Distinction 6: Theories of "Sincere" Rational Choice vs. Theories of "Strategic" Rational Choice.....................19Distinction 7: A "Status Quo" Policy vs. No "Status Quo" Policy......................................................23Conclusion...........................................................................................................243 Assessing Previous Theories of Supreme Court Decision-Making.......................................................26The Pioneers: Pritchett, Schubert, and Murphy........................................................................26Schubert's Attitude-Activation Model.................................................................................29The Attitudinal Model................................................................................................39Conceptual Problems with the Attitudinal Model.......................................................................41Other Issues Involving the Attitudinal Model.........................................................................52The Literature on Strategically Rational Justices....................................................................55Conclusion...........................................................................................................60PART II: A FORMAL MODEL OF SUPREME COURT DECISION-MAKING4 Why Formal Models?.................................................................................................65The Role of Theories and Models in Empirical Research................................................................66Potential Benefits from Formal Modeling..............................................................................68Potential Costs and Other Criticisms of Formal Modeling..............................................................73How Can We Be Sure that the Potential Benefits Exceed the Potential Costs?...........................................76Conclusion...........................................................................................................785 Definitions and Assumptions........................................................................................79Lines, Points, and Utility Functions.................................................................................80The Status Quo Policy................................................................................................83Preferred-to Sets and Win Sets.......................................................................................85The Number of Justices...............................................................................................89An Informational Assumption..........................................................................................90"Sincere" and "Strategic" Behavior...................................................................................91The Independence of Cases............................................................................................92Joining, Concurring, and Dissenting..................................................................................92Costless Opinion Writing.............................................................................................94Conclusion...........................................................................................................946 Coalition Formation and the Final Vote.............................................................................95When Can the Status Quo Policy Be Upset?.............................................................................97What Are the Constraints on the Set of Policies that Could Be Adopted?...............................................99What Policies Do Different Majority Coalitions Prefer to SQ?.........................................................102How Do Justices Behave When They Dislike the Majority Opinion?.......................................................108The Agenda-Control Version...........................................................................................110The Open-Bidding Version.............................................................................................125The Median-Holdout Version...........................................................................................129Comparison of the Agenda-Control, Open-Bidding, and Median-Holdout Versions..........................................134Is Agenda-Control Behavior Unstable?.................................................................................137Summary of Major Results.............................................................................................1387 Opinion Assignment.................................................................................................139Self-Assignment as an Opinion-Assignment Strategy....................................................................140Alternative Opinion-Assignment Strategies............................................................................142Opinion Assignment by a Justice Outside [W.sub.Jmed](SQ).............................................................143Opinion Assignment by a Justice Inside [W.sub.Jmed](SQ)..............................................................147Opinion Assignment by a Minority-Side Justice........................................................................156Would an Opinion Assigner Prefer Larger Coalitions?..................................................................158How Much Does Opinion Assignment Matter?.............................................................................161Summary of Major Results.............................................................................................1628 The Conference Vote................................................................................................163Different Kinds of Strategic Behavior from Different Kinds of Justices...............................................166Strategic Behavior by the Chief Justice 168Strategic Behavior by an Associate Justice Who Could Become the Opinion Assigner.....................................173Strategic Behavior by a Low-Seniority Justice Who Cannot Become the Opinion Assigner.................................192What If Everyone Behaves Strategically?..............................................................................195When Does the Chief Justice Self-Assign?.............................................................................207Will the Chief Justice Trust What Other Justices Say on the Conference Vote?.........................................209Summary of Major Results.............................................................................................2129 Certiorari.........................................................................................................215Sincere Behavior on Certiorari Decisions.............................................................................216Strategic Behavior on Certiorari Decisions...........................................................................220"Aggressive Granting" and "Defensive Denial" When Justices Are Strategic.............................................226Summary of Major Results.............................................................................................227PART III: FUTURE DIRECTIONS FOR THEORIES OF SUPREME COURT DECISION-MAKING10 Empirical Implications............................................................................................231Understanding the Five Stages of Supreme Court Decision-Making.......................................................233"Nonstrategic" or "Sincere" Behavior on the Supreme Court............................................................241Vote Switching between the Original and Final Votes..................................................................243Problems of Empirical Measurement....................................................................................246Conclusion...........................................................................................................24811 Future Research...................................................................................................249Do the Justices Have Perfect Information about Each Other's Preferences?.............................................250Do the Justices Always Have Clear and Fixed Preferences?.............................................................253Are Supreme Court Cases Independent from Each Other?.................................................................254Regular and Special Concurrences.....................................................................................259How Many Issue Dimensions Are There?.................................................................................262Costly Opinion Writing...............................................................................................268Extensions of the Model..............................................................................................269Exogenous Preferences and the Impact of The Law......................................................................271Broader Applications.................................................................................................272Conclusion...........................................................................................................273Notes................................................................................................................277References...........................................................................................................289Index................................................................................................................295

Chapter One

Despite a half century of research on decision-making by the United States Supreme Court, the answer to one of the most central questions-What explains the justices' choices and decisions?-remains a matter of controversy. The answer provided by many students of public law has long been that legal criteria are the primary guide for the justices' choices and decisions. These criteria include such matters as respect for precedent (stare decisis), for the plain meaning of words in the Constitution and statutes, for the original intentions of the Framers of the Constitution, and for the legislative history of the statutes.

In contrast, the answer provided by many students of judicial politics is that, although the justices' discourse in a case invariably involves reference to legal criteria such as these, the justices' actual decisions stem primarily from their personal attitudes toward the legal issues and objects involved in the case or from their personal preferences over the legal policies under consideration. Although the Legal Realists writing in the 1920s and 1930s helped provide an intellectual foundation for this general argument (see Clayton 1999, 16-22), the argument can be traced most directly to C. Herman Pritchett's 1948 book, The Roosevelt Court: A Study in Judicial Politics and Values, 1937-1947. This book was the first to systematically examine by quantitative methods the extent to which a justice's "liberalism" or "conservatism" could explain his or her choices and decisions.

A large body of literature was stimulated by Pritchett's work, all driven by his arguments about the centrality of the justices' personal views on various legal issues. Indeed, over the next few decades, several rather different kinds of theories of Supreme Court decision-making were developed within this intellectual context.

In our judgment, however, most of the major theories exhibit significant deficiencies. For example, some theories are based on assumptions that are either unclear or that do not appear consistent with each other. This means that it is difficult to determine just what is driving these theories. Of course, when it cannot be determined what is driving a theory, it is difficult to agree on what expectations-about an individual justice's behavior, or about the Court decisions-are implied by the theory. Hence, it is unclear what hypotheses can serve as tests of the theory. And if the assumptions of a theory are left unclear, or if the basic logic is not precisely specified, it will also be unclear as to what the most appropriate statistical tests might be. Thus, it will be difficult to conduct unambiguous tests of the theory.

Note also that decision-making by the Supreme Court involves several stages:

(1) The justices must first decide whether they want to hear some case; that is, the justices must decide whether to grant a writ of certiorari.

(2) If a writ of certiorari is granted, and after hearing oral arguments in the case, the justices meet in conference and make a preliminary decision-the "conference vote," as we will call it-on how they think the case should be decided.

(3) The Chief Justice, or the most senior associate justice in the majority if the Chief Justice is not in the majority on the conference vote, must then assign the writing of the majority opinion to a justice who voted with this majority.

(4) The author of the majority opinion must then try to write an opinion that gains the support of a Court majority.

(5) When the opinion is completed, each justice must then decide whether to join the majority opinion, concur (in various ways), or dissent.

However, few of the theories in the literature systematically analyze all five of these stages of the Court's decision-making process. If some theory is not developed for all five stages of the Court's decision-making process, it is difficult to determine how powerful the theory is; a theory that only explains one stage well is obviously less powerful than a theory that explains all five stages well. Because most of the theories have not been developed for all five stages, it remains difficult to determine whether any of the theories is more powerful than any of its competitors.

Finally, we would observe that nothing in the literature presents any kind of explicit and rigorous formal theory of the Court's multistage decisionmaking process. This is important because in multimember bodies with multistage decision-making processes (as with the Supreme Court), decisions made at the later stages may be affected by the decisions made at earlier stages. For such institutions, an informal and purely verbal logic is often an unclear guide-and sometimes even a misleading one-for determining what behavior to expect. Some kind of formalization may thus be necessary to clarify precisely what any theory implies should be expected to happen at each stage.

These kinds of problems with the major theories of Supreme Court decision-making give rise to three critical questions:

(1) By what standards should the adequacy of theories about Supreme Court decision-making be evaluated?

(2) Given these standards, how adequate are the major theories?

(3) Given any deficiencies these theories may have, how can improved theories be constructed?

For Question 1, our answer is that theories in Pritchett's tradition of judicial politics should meet the following kinds of standards:

The fundamental assumptions of the theory-about the nature of individual Supreme Court justices and about the character of the justices' interactions with each other-should be selected so that they are consistent with each other (that is, the set of assumptions should be intellectually coherent).

Each of these fundamental assumptions should be made clear and explicit.

The general logic of the theory-that is, how the fundamental assumptions interact with each other to produce decisions by the Court-should be clearly and explicitly described.

The theory should be operationalized in a model that specifies precisely what is assumed about the nature of the justices and that details exactly how justices interact with each other, at each stage of the Court's multistage decision-making process, to produce a final outcome.

The empirically testable implications that the model may have-that is, what choices and decisions any particular set of justices should be expected to make at any particular stage of the Court's decision-making process-should be logically derived from the basic rules and operations of the model.

These general kinds of standards should be uncontroversial, which is not to say that the standards have always been followed.

For Question 2, we have already outlined our answer: each of the major theories in Pritchett's tradition fails, in some significant manner, to meet the full set of standards. Insufficient effort, it appears, has gone into making these theories clear and specific, and little effort has gone into deriving from them any detailed working models of the Supreme Court's multistage decisionmaking process. Instead, the literature stemming from Pritchett seems to have focused disproportionately on the empirical and methodological issues involved in testing the major theories. Despite the passage of several decades, then, it appears to us that the conceptual and theoretical aspects of Pritchett's legacy have remained surprisingly undeveloped.

For Question 3, involving how to construct improved theories of Supreme Court decision-making, we believe that a necessary first step involves the development of a better understanding of the conceptual deficiencies of the current generation of theories. Indeed, it appears to us that the full extent of these conceptual deficiencies in the major theories of Supreme Court decision-making is insufficiently recognized. Once these problems are identified, then development of an improved generation of theories and models can proceed.

Our answers to this trio of questions thus lead us to the major purposes of this book:

To identify the conceptual problems of the current generation of theories of Supreme Court decision-making.

To formulate a rational-choice theory of Supreme Court decision-making that is built on a clear, consistent, and coherent set of assumptions involving the justices' personal preferences over legal policies.

To construct from this general theory a formal rational-choice model of the Supreme Court's multistage decision-making process.

To derive the logical implications of the formal model in a full range of situations and contexts.

The formal model we develop is a direct descendent of two of the major theories of Supreme Court decision-making: the "attitudinal" theory (see, for example, Segal and Spaeth 1993, 2002) and the "strategic" theory (see Murphy 1964; Epstein and Knight 1998; Maltzman, Spriggs, and Wahlbeck 2000). Both theories presume that the decisions of Supreme Court justices are motivated by the justices' personal views on legal issues and policies. However, the strategic theory further presumes that each justice is strategically rational, in the sense that each justice's decisions throughout a case, from beginning to end, are affected by how the justice expects the other justices to respond.

Because our formal model of multistage decision-making on the Supreme Court assumes that the justices pursue their own personal policy preferences in a strategically rational fashion, our model can thus be seen as formalizing a body of theory that is already central to the judicial politics literature. Indeed, we see ourselves as making relatively few fundamental assumptions that have not already been present in the literature, in one way or another, for quite some time. In fact, we think that our formal model could have been developed-and should have been developed-over 20 years ago: the basic theoretical ideas had already been advanced, and the modeling technology was already available. But for reasons we do not understand, no one took up this challenge: judicial politics scholars apparently considered formal theories to be of little use or relevance to them, and formal theorists were apparently not interested in the Supreme Court's internal decisionmaking practices.

We use our model to derive a series of propositions about how strategically rational justices should be expected to behave at each stage of the Court's decision-making process. Many of our propositions also serve as hypotheses that could be empirically tested. However, developing our formal model turns out to be a lengthy and complex enterprise. And because the model turns out to have a wide range of testable implications for Supreme Court behavior, a full empirical assessment of the model's implications would require several additional years of work, and the results would require many additional pages to describe. For these reasons, we have chosen to retain a narrow focus for the book: we develop our theory, construct our model, and derive the model's empirical implications, but we do not attempt to conduct an empirical assessment of any hypotheses based on our model's propositions.

Despite the lack of an empirical test here, we do argue that the development of a formal model such as ours is an integral element of the "if ... then" exercises that are critical to scientific progress. That is, if the basic premises of our model are assumed to be true, then certain kinds of behavior should be expected from the Court. If the justices' empirically observed behavior is as expected, this would lend support to our model. If the expected behavior is not empirically observed, this would suggest that something may be wrong with the test (which would mean that it has to be improved) or that something may be wrong with our model (which would mean that it has to be revised).It is when a model is formalized, and its basic assumptions and logic thereby made clear and explicit, that the scholarly community is best able to diagnose what may have gone wrong. In fact, we believe that scientific progress can often be hastened by initiating an ongoing series of cycles of these kinds of "if ... then" exercises, with each cycle involving the development, refinement, or redevelopment of the formal model and then its empirical test. Because our model does highlight several ways in which important pieces of the conventional wisdom about Supreme Court decision-making are at odds with how strategically rational justices should be expected to behave (at least according to our model), this suggests that empirical testing of our model is likely to generate interesting results, no matter what the results turn out to be.

We emphasize that we do not necessarily see ourselves as proponents of the theory, model, and propositions that we develop. In fact, each of us harbors his own doubts and hesitations about the empirical validity of various results we develop. Instead, we prefer that our study be seen as providing a service to the field of judicial politics: by developing the full implications of one important class of theories that has been a significant part of the judicial politics literature for several decades now, we hope to enable scholars in this field to make a more thorough and systematic empirical assessment than has yet been possible.

We also emphasize that our rational-choice model is probably the simplest possible model of rational justices who strategically pursue their policy preferences in the Supreme Court's multistage decision-making process. Although more complex models of judicial decision-making could be developed, only an initial effort like ours, one that uses basic assumptions and technology, will indicate the extent to which further theoretical developments are necessary. Thus, whatever the ultimate outcome of these cycles of theoretical and empirical research that we are trying to initiate here, we are confident our formal analysis will turn out to have been a useful and necessary first step.

Plan of the Book

Our book has three parts. In addition to this chapter, Part I contains two other chapters. Chapter 2 discusses seven pairs of assumptions, prominent in the literature in varying degrees, about the nature of Supreme Court justices and about the characteristics of Supreme Court decision-making. Understanding these seven distinctions helps situate our own model within the larger literature on the Court. Chapter 3 illustrates the importance of these seven distinctions by showing how these various assumptions have structured several major studies of Supreme Court decision-making.

Part II contains six chapters where we present our formal model. Chapter 4 discusses why developing formal models may help solve some of the problems we have identified in the judicial politics literature on Supreme Court decision-making. Although the development of formal models has some costs, we believe that at this time in the field of judicial politics, the benefits of developing the formal models far outweigh the costs. Chapter 5 then provides definitions of our key terms and describes the assumptions on which our model is based.

Chapter 6 focuses on Stages 4 and 5 of the decision-making process: coalition formation and the final vote. We start at the end of the decision-making process because, to model strategic behavior, it is necessary to work backward from the final policy decision (in which the actors are primarily interested) so as to figure out what would be the rational decisions for them to make at the successively earlier stages of the decision-making process. Chapter 7 then models opinion assignment (Stage 3), Chapter 8 models the conference vote (Stage 2), and Chapter 9 models the original decision to grant certiorari (Stage 1).

(Continues...)

Excerpted from Strategic Behavior and Policy Choice on the U.S. Supreme Courtby THOMAS H. HAMMOND CHRIS W. BONNEAU REGINALD S. SHEEHAN Copyright © 2005 by Board of Trustees of the Leland Stanford Junior University. Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.