Elements of Judicial Strategy

By Walter F. Murphy

Now in a readily available digital edition, and adding a substantive 2016 Foreword by Lee Epstein and Jack Knight, this classic of law and political science is presented to a new generation of thoughtful observers of the U.S. Supreme Court and how its justices create judicial decisions. As Epstein and Knight write, this book is "extraordinary. It's the rarest of rare: a breakthrough of the path-marking, even paradigm-shifting, variety...." Its publication offered a "huge conceptual breakthrough. Elements was the first to offer a strategic account" of judging, and its "framework forever changed the study of judicial behavior." It remains influential to current thought, extending even in its "global reach," and is an important part of modern social sciences and law.

First outlining the sources and instruments--and limitations--of judicial power, the author then shows how policy-oriented justices might take advantage of their power positions to maximize their impact on the formation and execution of public policy. In this book Walter F. Murphy attempts to understand how, under the limitations which the American legal and political systems impose, Supreme Court justices can legitimately act to further their policy objectives. Murphy also considers ethical issues raised by the model of judicial decision-making he describes. Throughout, systematic analysis is supported by prodigious research and fascinating real-world examples over the years and in very different judicial administrations.

The new ebook edition of this foundational work features quality digital formatting, active Contents, linked footnotes and endnotes, and fully linked tables of cases and subject-matter index. Also available in new paperback and hardcover editions from Quid Pro Books.

• Language: English
• Publisher: Quid Pro, LLC
• Release date: Jun 2, 2016
• ISBN: 9781610273541

Walter F. Murphy

Novelist and long-time professor of constitutional law at Princeton, Walter F. Murphy authored The Roman Enigma, Upon This Rock, and the New York Times bestselling novel The Vicar of Christ. His much-cited nonfiction works include Elements of Judicial Strategy and Congress and the Court.

Book preview

Elements of

Judicial Strategy

ELEMENTS OF

JUDICIAL STRATEGY

by

WALTER F. MURPHY

Foreword by Lee Epstein and Jack Knight

CLASSICS OF LAW & SOCIETY

qp

QUID PRO BOOKS

New Orleans, Louisiana

Smashwords edition.

Copyright © 1964 by the University of Chicago; reprint edition © 2016 by Doris Maher Murphy. Foreword © 2016 by Lee Epstein and Jack Knight. All rights reserved. No part of this book may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying, copying its digital form, recording, or by any information storage and retrieval system, without written permission from the current publisher (or, in the case of the new foreword, from its authors).

Previously published in 1964 in hardback, and in 1973 in paperback, by the University of Chicago Press, Chicago and London.

Published in 2016 by Quid Pro Books, at Smashwords, as part of the Legal Legends series. Also available in new paperback and hardcover editions.

ISBN 978-1-61027-354-1 (ePUB)

ISBN 978-1-61027-356-5 (pbk.)

ISBN 978-1-61027-353-4 (hbk.)

QUID PRO BOOKS

Quid Pro, LLC

5860 Citrus Blvd., Suite D-101

New Orleans, Louisiana 70123

www.quidprobooks.com

qp

Publisher’s Cataloging-in-Publication

  Murphy, Walter F., 1929-2010.

      Elements of judicial strategy / Walter F. Murphy.

            p. cm. — (Legal legends)

      Reprint edition. Originally published: Chicago and London: University of Chicago Press, 1964. New foreword by Lee Epstein and Jack Knight.

      Includes bibliographical references and index.

1. United States. Supreme Court. 2. Judicial process—United States. 3. Judicial review—United States. 4. Law—political aspects. I. Epstein, Lee. II. Knight, Jack. III. Title. IV. Series.

KF 8748 .M84 2016

For Terry

Contents

FOREWORD • 2016

Lee Epstein & Jack Knight

*

Knowledge in the social sciences develops and cumulates at a snail’s pace—a small hypothesis here, a nugget of an idea there. That’s why Elements of Judicial Strategy is so extraordinary. It’s the rarest of rare: a breakthrough of the path-marking, even paradigm-shifting, variety (though it took some time before anyone quite realized it).

The story of Elements properly begins in the early 1960s when Professor Walter F. Murphy was in the midst of writing Elements. Back then political scientists were engaged in a debate over the study of judicial behavior—though debate is way too mild. It was more like a civil war for the heart and soul of the field. On the one side were the old-guard legalists (Murphy called them traditionalists) who had long dominated the disciplinary study of judging. Even more than many law professors of the day, the legalists favored a brand of Blackstone’s declaratory theory:¹ Not only are judges supposed to maintain and expound the existing law, they in fact do so, deciding cases according to the known customs and law of the land² regardless of their personal preferences. On the other side were the rebel behavioralist-quants. Armed with lessons from the realist movement,³ social-psychological theories, large-n datasets, and fancy statistical methods, the behavioralists contended that we can best explain the decisions of judges merely by identifying their ideological (liberal or conservative) attitudes about the subject of the relevant case. The idea is that judges care less about applying the law than they do about etching their own policy preferences into judicial decisions. They are little more than single-minded seekers of legal policy,⁴ and hardly the depositories or living oracles of law that Blackstone and the legalists posited.⁵

This war-of-words played out at political science conferences and in the pages of our journals—with the battle between the legalist Wallace Mendelson and the behavioralist Harold J. Spaeth typifying the state of affairs. As far as we can tell, it was Spaeth who fired the first shot. In 1964, he published The Judicial Restraint of Mr. Justice Frankfurter—Myth or Reality?⁶ Spaeth’s focus on Frankfurter was no accident. At the time, Frankfurter’s attachment to self-restraint was rather emphatically asserted by the Justice’s defenders, critics, and neutral observers alike.⁷ It turned out, though, that the observers got it wrong, or at least so proclaimed Spaeth. His empirical analysis suggested that Frankfurter was no more a restraintist than Justice James McReynolds—one of the four horsemen responsible for striking down ten federal laws (in whole or in part) between 1934 and 1936. On Spaeth’s account, Frankfurter too had no hesitation invalidating government action that he found ideologically distasteful and upholding action that fit within his value system. The more general takeaway from Spaeth’s paper, of course, was that if Frankfurter—whose name had become synonymous with judicial restraint—was not a restraintist, no one was. It was all one big myth, yet another smokescreen behind which justices mask their ideology.

The legalist Mendelson had had enough. Just three years earlier, Mendelson had written that Frankfurter was the model of restraint, a justice who would resolve all reasonable doubt in favor of the integrity of sister organs of government.⁸ Now this whippersnapper Spaeth was challenging him in print and Mendelson wasn’t going to take it sitting down. In the pages of a leading journal in political science, he took aim at Spaeth and the other behavioralists, tearing into their methods and calling their conclusions downright strange if not over-simple, shallow, mindless, and limited.⁹ Traditional scholarship, Mendelson proclaimed, [is] more perceptive."¹⁰

When Spaeth shot right back¹¹ and Mendelson responded in kind,¹² the journal’s editor finally stepped in:

[Mendelson and Spaeth] are obviously unable to convince each other but their exchanges may serve at least to titillate the profession. Since the controversy seems endless, the editor draws the line on further comments and responses to this provocative article!¹³

The editor was right. Spaeth and Mendelson would never see eye-to-eye, but, ultimately, Spaeth won the day—both with respect to the battle over the myth of Frankfurter’s self-restraint¹⁴ and the war over the general direction the field would take. Political scientists of today use both quantitative methods and softer approaches to analyze law and legal institutions. But few if any would outright reject the idea that politics—in the form of ideology—plays no role in the Court’s decisions.

Where was Professor Murphy during the battles of the 1960s? He wasn’t standing on the sidelines, nor was he caught in the cross-fire; he planted himself firmly in the behavioralist camp.¹⁵ But his worldview was quite different from Spaeth’s and other leading attitudinal-behavioralists of the day, including Glendon A. Schubert¹⁶ and S. Sidney Ulmer.¹⁷ Although Murphy too believed that justices of the U.S. Supreme Court were policy oriented, he departed company with the attitudinalists in how justices go about achieving their policy goals. While the attitudinalists were busy applying social-psychological theories to judicial behavior¹⁸—arguing that liberal justices vote the way they do because they are liberal; and conservatives, because they are conservative¹⁹—Murphy found it implausible that justices reach their decisions, or make any other choices for that matter, in a vacuum. As he later described his thinking, Elements took as a point of departure the individual Supreme Court justice and tried to show how, given his power as one of nine judges and operating within a web of institutional and ideological restraints, he could maximize his influence on public policy.

And therein lies Professor Murphy’s huge conceptual breakthrough. Elements was the first to offer a strategic account of judicial behavior.²⁰ On this account (1) judges make choices in order to achieve certain goals (with the policy goal paramount in Elements²¹), (2) they act strategically in the sense that their choices depend on their expectations about the choices of their colleagues and other relevant actors, and (3) their choices are structured by the institutional setting in which they are made. Defined in this way, Murphy’s account belongs to a class of non-parametric rational choice explanations as it assumes that goal-directed actors—here, judges—operate in strategic or interdependent decision making contexts. (Elements focuses mainly on the internal context of judging—relations among colleagues; his Congress and the Court,²² another classic, stresses the external constraints placed on justices by the legislature.)

How and why the Elements framework forever changed the study of judicial behavior are interesting questions, and we’ll get to them momentarily. It’s worth pausing, though, and asking how Murphy came to it. We raise this because his fellow realists-behavioralists were all off in social-psych world, while he drew on economics. Why did he depart company on this crucial dimension?

In our discussions with Professor Murphy we asked him just that. He told us that ever since his days in the Marine Corps, he conceptualized the world in strategic terms. Just as no military commander can expect to win a battle without taking into account the position and likely actions of his opponents, he believed that judges can’t expect to establish lasting policy unless they are attentive to the preferences and likely actions of actors who could stand in their way, including colleagues, politicians, and the public.

The parallels between battles in the field and on the bench, though, were not immediately apparent to him. Indeed, in 1960, when Murphy first started working in the papers of several justices—a major source of his data in Elements²³—he did not know what he was doing. From various biographies, especially Alpheus Mason’s Harlan Fiske Stone,²⁴ he could see the importance of strategic behavior in the development of the Court’s doctrine, but his ideas remained unformed. It was not until a year later, in 1961, when he was reading An Economic Theory of Democracy on a train bound for Charleston that the Elements framework popped into his head. He now knew to what use he would put the judicial papers. (An Economic Theory of Democracy [1957], Anthony Downs’ classic work on political parties, along with William Riker’s The Theory of Coalitions [1962], played central roles in bringing the rational choice paradigm to political science.)

Murphy understood he was onto something big or at least distinctive. In the preface to Elements, he wrote: Almost as jarring to some readers as quotations from private papers will be my use of terms which are familiar to economic reasoning and the theory of games but which are alien in the public law literature.²⁵ Perhaps that’s why reactions to early drafts of Elements were not especially favorable.²⁶ Murphy told us that his colleague Mason thought Elements’ emphasis on behind-the-scenes maneuvering on the Court was going to stir up snakes; Schubert was dismayed that Murphy was culling stories and cases from the justices’ papers instead of setting up hypotheses, systematically mining information, and creating large-scale datasets to test them.

And, yet, once Elements appeared in print, a few scholars found aspects of the work attractive. Or at least attractive enough to continue in its path. Particularly noteworthy was Howard’s analysis of fluidity,²⁷ which attempted to provide systematic support for one of Murphy’s key observations: judges work changes in their votes and permit their opinions to be conduits for the ideas of others through internal bargaining.²⁸ Howard’s methodology was similar to Murphy’s in its reliance on a small number of important cases, but Howard cast his argument in general terms: It may come as some surprise to political scientists how commonplace, rather than aberrational, judicial flux actually is.²⁹ He further claimed that hardly any major decision [is] free from significant alteration of vote and language before announcement to the public.³⁰

Howard’s article was not the last of the early post-Elements pieces. Into the next decade, political scientists applied theories grounded in assumptions of rationality (especially game theory) to study opinion coalition formation³¹ and jury selection.³² By the 1970s, there had been enough work invoking game-theoretic analysis, in particular, that Brenner wrote a bibliographic essay devoted exclusively to the subject.³³

A glance at the works on Brenner’s list, however, reveals that most were not explicit applications of game theory or were conducted in the late 1960s, just after Murphy published Elements. The collective shrug in the 1970s and 1980s seems to have reflected the view that Elements, while a good read, did not advance the project of illuminating judicial behavior. Far more valuable, political scientists apparently thought, were determinants of the judges’ decisions following from the social-psychological theories (none of which acknowledged a strategic component to decision making).³⁴ The data tell the story. Within a decade of the publication of Elements, papers adopting variants from the social-psychological paradigm accounted for 60% (16 of 27) of the articles published in the American Political Science Review, the discipline’s (then) flagship journal.³⁵ During that same period, only two essays attentive to rational choice appeared—one of which was a critical assessment.³⁶

But why? Why did scholars so fully embrace the social-psychological paradigm and so fully ignore the sort of strategic analysis Murphy conducted in Elements? Two answers come to mind. The first is from Schwartz who pointed to the lack of equilibrium predictions: Murphy, he claims, "only identifies strategies that might be pursued under some circumstances. Often such a pronouncement is immediately followed by a disclaimer that the contrary strategy might be more appropriate in other circumstances. The problem is that he derives no tight predictions about exactly when we should expect to see certain behaviors as opposed to others."³⁷

There’s some merit to this view. In direct contrast to other early advocates of rational choice theory, including Downs, Murphy did not write down any models and derive equilibria that others could go out and test—as did a multitude of scholars with the predictions in An Economic Theory of Democracy. Even more to the point, Murphy’s predictions were a good deal more ambiguous than those offered by early adherents of social-psychological approaches. Compare, for example, a Murphy hypothesis with one offered by Schubert:

Murphy: "When a new Justice comes to the Court, an older colleague might try to charm his junior brother."³⁸

Schubert: "In accordance with modern psychometric theory, which generalizes the basic stimulus-response point relationship, Supreme Court cases are treated as raw psychological data. . . . Each case before the Court for decision is conceptualized as being represented by a stimulus (j) point. . . . The combination of the attitudes of each justice toward these same issues also may be represented by an ideal (i) point. . . . Obviously how the case will be decided will depend upon whether a majority or minority of the i-points dominate the j-point. If a majority of i-points dominate, then the value or values raised in the case will be upheld or supported by the decision ‘of the court’; and if, to the contrary, the j-point dominates a majority of i-points, then the value or values raised will be rejected. . . ."³⁹

And, yet, scholars were able to glean predictions from Murphy’s work and did attempt to test them. This was true of Howard’s work on vote fluidity and, we might add, it is true of the more modern crop of strategic work, much of which explicitly identifies Elements as its starting point—including our own The Choices Justice Make.⁴⁰ In other words, Murphy may not have laid out predictions as boldly as did other rational choice theorists or those advocating variants of the social-psychological model. But within his work were plenty of intuitions that others could use to develop models, tease out empirical implications, and ultimately test against data.

If it wasn’t the lack of precise expectations that led to indifference toward, if not downright rejection of, strategic accounts, what then? We believe the explanation lies in the very nature of those tests and in the results they generated—an explanation, we think, that is a more faithful representation of the tenor of the times. As we already noted, and as Murphy made clear to us, during the 1960s the great battles in the field of judicial politics were not between proponents of the rational choice and social-psychological models but between traditionalists and behavioralists: between those who believed that social scientists should develop realistic and generalizable explanations of social behavior and those who did not; and, increasingly, between those who believed scholars could quantify behavior and those who did not.⁴¹ To be a scientist in the field of judicial behavior by the 1970s was to value data and to believe in the power of statistics. For this reason it isn’t surprising that scholars working in the social-psychological tradition triumphed over their strategically-minded counterparts: As early as 1941⁴² but with particular force in the 1960s,⁴³ they claimed to have boatloads of systematically developed data to support their hypotheses. In other words, unlike Murphy (and Howard), they shunned detailed analysis of particular cases (the modus operandi of the traditionalists) and instead focused on large samples of Court cases—the dispositions of which they claim to predict with a good deal of success.

But there’s more: Just as scholars were asserting that hypotheses following from the social-psychological model held up against rigorous data-intensive investigations, they were also arguing Murphy’s strategic account did not withstand similar scrutiny. A critical work here is by Brenner,⁴⁴ which reassessed Howard’s contention that voting fluidity was rampant on the Court. Brenner compared votes cast in conference with those in the published record for major and non-major decisions. Although he found minimal change in case disposition (about 15%), his results for vote shifts were rather dramatic: In 48% of the major cases and in 59% of the non-major cases did at least one justice change his vote. Still, Brenner concluded that Howard (and, by implication, Murphy) was largely incorrect—that, in fact, considerable stability exists in voting. And it was this interpretation of Brenner’s work that became the prevailing wisdom among political scientists. (Though no longer. Subsequent work has shown that Murphy and Howard got it right.⁴⁵)

Given Brenner’s own rendition of his study, given the massive amounts of data analysts gathered to support the social-psychological model, and given the significance political scientists in this field attached to large-scale statistical studies, it is easy to understand why follow-ups to Elements—studies grounded in assumptions of rationality—failed to make any substantial showing in the political science journals of the 1970s and 1980s.

So what happened? How did Elements move from a one-off good read to the path-marking, paradigm-shifting work we now know it to be? Elsewhere we have recounted possible answers,⁴⁶ but to us one stands out: data. If large-n studies explain why Murphy’s strategic account failed to gain traction, they also account for its reclamation. Beginning in the mid-1990s but with particular force in the 2000s, the big data studies, by political scientists, economists, and legal academics, started to appear: at first in a dribble and then in a downpour but all providing systematic empirical support for the plausibility of Elements’ assumptions and implications. Nearly three-quarters (17/23) of the articles published on judicial behavior in the American Journal of Political Science (today’s flagship journal) over the last ten years (2005–2015) adopt a strategic account in full or part. Because a more complete reversal from the 1970s into the early 1990s is hard to imagine(!), we have dubbed the 2000s forward as nothing short of a strategic revolution in the analysis of judicial behavior.⁴⁷

In the early part of the revolution work tended to focus on U.S. Supreme Court justices, as did Elements and as did The Choices Justices Make. There were studies on many of the topics of concern to Murphy: forward thinking,⁴⁸ agenda manipulation,⁴⁹ opinion assignment,⁵⁰ and opinion writing,⁵¹ as well as research on relations between the Court and the elected branches of government.⁵²

This work continues today.⁵³ But as the strategic revolution took hold other U.S. courts moved to the fore. There is now a substantial body of literature on the behavior on the U.S. Courts of Appeals and state supreme courts. Research on the federal circuits tends to focus on so-called panel, collegial, or peer effects, asking whether the case’s outcome (or a judge’s vote) would have been different had a single judge, and not a panel, decided the case⁵⁴—and, if so, why? Murphy would have appreciated one answer: Because appellate judges worry about the Supreme Court (or the circuit en banc) reversing their decisions, a judge on a panel who shares the preferences of the Supreme Court or the circuit (but not the panel’s majority) can constrain her colleagues by threatening to blow the whistle on would-be offenders.⁵⁵ Work on state courts asks whether elected judges must attend to their constituents to retain their jobs. There is plenty of evidence suggesting that they do, whether by holding for in-state plaintiffs to redistribute wealth from out-of-state businesses to the plaintiffs⁵⁶ or by ruling against criminal defendants on the theory that the public doesn’t like judges who appear soft on crime.⁵⁷

Happily, Professor Murphy lived long enough to see the impact of Elements on the study of the behavior of U.S. judges. We only wish he were around today to appreciate the global reach of his work. Scholars in the United States and elsewhere are now adapting his framework to study courts in virtually every region in the world, from Europe⁵⁸ to Latin America⁵⁹ and from Asia⁶⁰ to the Mideast.⁶¹ As a great student of comparative law and legal institutions, Professor Murphy would have appreciated, and most certainly would have contributed to, this latest extension of this work. Equally without doubt he would have reveled in all the social scientists and legal academics so inspired by his sheer intellect, insight, and elegance—including, we hope, the two of us.

Footnote

* Lee Epstein is the Ethan A.H. Shepley Distinguished University Professor at Washington University in St. Louis. Jack Knight is the Frederic Cleaveland Professor of Law and Political Science at Duke University. In past work we have explored Walter F. Murphy’s many contributions to the study of law and legal institutions. See, e.g., Lee Epstein & Jack Knight, Toward a Strategic Revolution in Judicial Politics: A Look Back, A Look Ahead, 53 Political Research Quarterly 625 (2000), and our Walter F. Murphy, in Nancy Maveety, ed., Pioneers of Judicial Behavior (2003). For this Foreword we drew on these works, as well as on interviews we conducted (and many conversations we had) with Professor Murphy. We hope he would approve of our interpretation of the events leading up to Elements, its reception, and its ultimate impact, but we need to be clear: it is our interpretation, not his.

PREFACE • 1964

One of the occupational hazards of writing in a realistic fashion about the Supreme Court of the United States is that one runs the double risk of being accused of muckraking by those who, at a given moment, identify themselves as the friends of the Court and the sometimes even more unpleasant fate of being lauded by those who currently proclaim themselves to be its enemies. When one dares to use the private papers of deceased Justices to bolster one’s analysis, the likelihood of both dangers actually occurring becomes great indeed. Let me make, then, the protestation that, while I bow to none in my respect for the Court, this book is neither an effort to debunk or belittle, nor an effort to praise or defend. It is an attempt to understand—to understand how, under the limitations which the American legal and political systems impose, a Justice can legitimately act in order to further his policy objectives.

Since the aims and approach of this book are set out in detail in the introductory chapter, I shall deal here only with certain peripheral matters. The first concerns the use of private papers of Justices. I have carefully gone over the papers of Charles Evans Hughes, Horace Lurton, James C. McReynolds, Harlan F. Stone, George Sutherland, William Howard Taft, and those parts of Frank Murphy’s papers which were open by August, 1963. In addition, my good friend David F. Hughes of Centre College generously allowed me full use of his copious notes on the Salmon P. Chase Papers at Philadelphia and Washington. My empirical data are largely drawn from these collections. I went through parts of the Franklin D. Roosevelt and Harry S. Truman papers, and the Oral History Files at Columbia University, but I used very little material from them.

There is a feeling in some quarters that the private papers of Justices—papers which the Justices themselves sometimes compiled and arranged for public use—should not be looked into lest the image of the Court be somehow tarnished. Taking an opposite stand, Charles Evans Hughes claimed that if the inner workings of the Court were widely known and understood, popular respect for the judiciary would be vindicated.¹ Certainly Hughes’s faith was well founded.* The private papers of the Justices which have been relied upon in this book do reveal some human frailties. To a much greater extent they reveal also a truly remarkable degree of personal integrity, intellectual vigor, and selfless dedication to duty—qualities which these judges directed toward the attainment of what they earnestly believed to be sound law, wise public policy, and, not least of all, justice.

A more intelligent objection to the use of private papers of Justices is that such use might discourage the free exchange of ideas within the Court. There may be some validity in this objection. Candid discussion among Justices is an important value,† but so is responsibility, and responsibility usually does inhibit free choice. Because they are not restrained directly by the check of the ballot box there is all the more reason why Justices should be checked by the informed judgment of history.

I have not quoted from the private correspondence of any living Justice, sitting or retired, without his permission.* On the other hand, I have used and used heavily the private papers (repeating, I hope, neither gossip nor scandal) of deceased Justices in the belief that judges, like all public officials, have to bear the criticism of good and bad scholars and run the risk of being misunderstood—a risk that every author shares—in the hope that truth will ultimately triumph. With his typical pungent wit, Karl Llewellyn stated my position far better than I ever could:²

It is well to remember that neither secrecy of the court’s deliberation or later secrecy about what went on during that deliberation rests in the nature of things or in any ordinance of God. The roots of each are either practical or accidental, and it is only either ignorance or tradition which makes us feel that we have here something untouchable, a semiholy arcanum. . . . Thus the storied sanctity of the conference room represents to me as pragmatic and nonmystic a phase of appellate judicial work as the handling of the docket. Our modern fetish of secrecy reminds me of the shock German lawyers displayed at the notion of such dangerous things as published dissenting opinions.

I find that the use of private papers entails a more serious hazard than that of lifting secret veils. Collections of letters and memoranda are always incomplete records; this is especially true of papers of men who are in daily face-to-face contact with each other or who frequently use the telephone. There is as great a danger in trying to fill the inevitable gaps as in not understanding that a writer’s point of view may destroy the validity of a letter or memorandum as a factual account of contemporary events. These are well-known problems of historical research, and I hope I have avoided at least the more common errors in using personal documents.³

In this context it should be noted that I have drawn most heavily on material from the Taft Papers for illustrations of judicial efforts to exert personal influence in the political processes. Because of his long political career and especially his term of office as President, it may be objected that Taft’s position was unique. It is true that Taft had unique advantages; but he also had unique disadvantages. The advice of a former President, as no one had better reason to know than Taft himself, is often far less welcome in the White House than suggestions from a former senator, governor, or Attorney General. And, over the years, Taft had made enemies as well as friends on Capitol Hill. It should also be kept in mind that most members of the Court have extensive political careers before they come to the bench,⁴ and that not many of them have showed less understanding of the realities of politics than did Taft before 1921. I believe that, insofar as they relate to previous experience, Taft’s net advantages over the typical Justice were of degree rather than of kind, though it is entirely possible he made more of these net advantages in his dealings with other government officials than any other Justice either before or since.*

Almost as jarring to some readers as quotations from private papers will be my use of terms which are familiar to economic reasoning and the theory of games but which are alien to public law literature. Words like bargaining may seem badly out of place—and out of taste—when applied to the judicial process. But if one strips such concepts of their emotive connotation and tries to utilize them as descriptive rather than evaluative expressions, I think any nefarious overtones will be quickly dissipated. We all bargain—unless we are willing and able to employ superior brute force to settle disagreements, or we are in the enviable position of being so obviously and perpetually infallible in our judgment that all affected men agree with us, or we are so supine as never to disagree with our peers. At first reading, other terms may be equally disagreeable as bargaining, but I hope they will become equally neutral if it is understood that they are meant to imply neither diabolic nor angelic qualities.

This book is not intended, however, as an essay in the formal theory of games. Most of what we call game theory is a mathematical development of zero-sum games, or games of pure conflict. The judicial process, I think, is one of the areas in which a zero-sum model is inappropriate. Now a model need not be, if in fact it ever can be, a replica of reality; certainly the model of judicial decision-making I offer in chapter ii is not. But models should be simplifications not distortions of reality, and I hope to demonstrate that judicial decision-making, at least from the point of view of the judges, is not in most of its aspects zero-sum.⁵ Following Thomas Schelling, I believe that games of pure conflict form one limiting point of the real world, with games of pure co-operation at the other extreme.⁶ As I will often state in the text of the book, I have used game-theory concepts and vocabulary, but of a kind of game theory which Schelling calls mixed-motive; that is, where elements of both conflict and co-operation are present. I think these concepts are appropriate and fruitful in the study of judicial behavior.

This book, as with everything else I have ever done, has been built on a foundation of enormous debts. I am obliged to the Social Science Research Council for financing several summers of research and one of writing, as well as to Princeton University for various funds and time off to pursue this work at a pace which for a more industrious person would have been leisurely. The directors and staffs of the Manuscript Division of the Library of Congress, the Alderman Library of the University of Virginia, the Michigan Historical Collections of the University of Michigan, the Franklin D. Roosevelt Library at Hyde Park, New York, and the Harry S. Truman Library at Independence, Missouri, supplied cheerful and valuable assistance to me in my work with document collections. Miss Helen Fairbanks of the Firestone Library of Princeton University helped run down many esoteric bibliographical references, and my research assistant, Stephen Beckwith, of the Princeton class of 1964, aided in checking citations. Mrs. Helen Wright typed the manuscript with her usual skill and acted as a welcome general editor both on matters of style and substance.

A number of people have read all or parts of the manuscript. Stanley Kelley, Jr., of Princeton University read in full an early draft with a care that must have been as tedious to him as it was painful—and valuable—to me. David Danelski of Yale University, Robert K. Faulkner, Alpheus Thomas Mason, Louis Werner of Princeton, Joseph Tanenhaus of New York University, and C. Herman Pritchett of the University of Chicago read the final draft and made many useful suggestions. Others whose comments on particular aspects were especially helpful are Robert H. Birkby of Vanderbilt University, Duane Lockard of Princeton, John Nolan of the