Beyond the Hero Judge Institutional Reform Litigation as Litigation Margo Schlanger Mich Law Review May 1999
Download original document:
This text is machine-read, and may contain errors. Check the original document to verify accuracy.
BEYOND THE HERO JUDGE: INSTITUTIONAL REFORM LITIGATION AS LITIGATION Margo Schlanger* JUDICIAL POLICY MAKING AND THE MODERN STATE: HOW THE COURTS REFORMED AMERICA’S PRISONS. By Malcolm M. Feeley1 and Edward L. Rubin.2 New York: Cambridge University Press. 1998. Pp. xv, 490. $69.95. In 1955, in its second decision in Brown v. Board of Education, the Supreme Court suggested that federal courts might be called upon to engage in long-term oversight of once-segregated schools.3 Through the 1960s, southern resistance pushed federal district and appellate judges to turn that possibility into a reality.4 The impact of this saga on litigation practice extended beyond school desegregation, and even beyond the struggle for African-American equality; through implementation of Brown, the nation’s litigants, lawyers, and judges grew accustomed both to issuance of permaCopyright by Michigan Law Review and Margo Schlanger. * Assistant Professor of Law, Harvard. B.A. 1989, J.D. 1993, Yale. Many of the citations in this review refer to unpublished sources, mostly case pleadings and unreported court opinions. They are all on file with the author. — Ed. I received helpful comments on a draft of this review from Elizabeth Alexander, Dick Fallon, Jerry Frug, Phil Heymann, Christine Jolls, Duncan Kennedy, Martha Minow, Steve Rosenbaum, Peter Schuck, David Schoenbrod, Bill Stuntz, Mark Tushnet, Lucie White, and, as always, from Sam Bagenstos. I also interviewed a number of current and former prisoners’ lawyers, who were forthcoming and generous with their time. Their names are given in the relevant footnotes. Former Civil Rights Division deputy assistant attorney general Jim Turner was also kind enough to help me reconstruct the near-ancient past of United States Department of Justice policy in the early 1970s. Andrew Ehrlich and Brandon Ponichter provided able research assistance, and the diligence of the Harvard Law Library’s document delivery staff was above any call of duty. Any errors are, of course, my responsibility. 1. Claire Sanders Clements Dean’s Professor of Political Science and Law, University of California, Berkeley. 2. Professor of Law, University of Pennsylvania Law School. 3. Brown v. Board of Educ., 349 U.S. 294, 301 (1955) (Brown II ) (discussing “period of transition” during which district courts should maintain jurisdiction over desegregation cases to “consider the adequacy of any plans the defendants may propose . . . and to effectuate a transition to a racially nondiscriminatory school system”). 4. See, e.g., United States v. Jefferson County Bd. of Educ., 372 F.2d 836, 860 (5th Cir. 1966) (describing challenges posed by 128 school desegregation cases filed in district courts in Alabama, Florida, Georgia, Louisiana, Mississippi, and Texas between 1956 and 1966), adopted in relevant part, 380 F.2d 385 (5th Cir. 1967) (en banc); see also Briefs for the United States, app. Vols. II-III, id. (panel) (No. 23,345 et al.) (setting out the procedural history of each and every one of these cases). 1994 May 1999] Judicial Policymaking 1995 nent injunctions against state and local public institutions, and to extended court oversight of compliance.5 A new kind of case (termed, variously, “public law litigation,” “structural reform litigation,” or “institutional reform litigation”) developed as civil rights plaintiffs and their lawyers began to seek and obtain litigated reform and continuing injunctive relief not only against schools, but also against prisons,6 jails,7 mental health and mental retardation facilities,8 and many other types of institutions.9 Law professors, law students, and political scientists followed a few years behind with descriptions, discussions of origins, efforts at legitimation, critiques, and case studies. Professors Abram Chayes and Owen Fiss set the terms of the scholarly debate;10 both de5. The civil rights injunctions had historical antecedents in earlier antitrust injunctions, court-supervised bankruptcy reorganizations of complex corporate entities such as railroads, and even trust and probate matters. On antitrust injunctions, see, for example, Michael E. DeBow, Judicial Regulation of Industry: An Analysis of Antitrust Consent Decrees, 1987 U. CHI. LEGAL F. 353; OWEN M. FISS, INJUNCTIONS 325-414 (1972) [hereinafter FISS, INJUNCTIONS]; on other historical antecedents, see Theodore Eisenberg & Stephen C. Yeazell, The Ordinary and the Extraordinary in Institutional Litigation, 93 HARV. L. REV. 465, 481-94 (1980). In addition to the Brown decision, and the movement that produced it, another important prerequisite to the “flowering” of the new form of litigation was the 1966 amendments of the Federal Rules of Civil Procedure, which established modern class action practice. STEPHEN C. YEAZELL, FROM MEDIEVAL GROUP LITIGATION TO THE MODERN CLASS ACTION 237 (1987); see generally id. at 238-60; cf. Archibald Cox, The New Dimensions of Constitutional Adjudication, 51 WASH. L. REV. 791, 808-12 (1976); Robert L. Carter, The Federal Rules of Civil Procedure as a Vindicator of Civil Rights, 137 U. PA. L. REV. 2179, 2184-86 (1988). 6. See, e.g., Holt v. Sarver, 300 F. Supp. 825 (E.D. Ark. 1969), 309 F. Supp. 362 (E.D. Ark. 1970), affd., 442 F.2d 304 (8th Cir. 1971); Taylor v. Perini, 413 F. Supp. 189, app. A (N.D. Ohio 1976) (reprinting 1972 order); Battle v. Anderson, 376 F. Supp. 402 (E.D. Okla. 1974). 7. Speaking generally, prisons are state facilities that house long term felony convicts, whereas jails are county and city facilities that house pre-trial detainees, misdemeanant convicts, and shorter term felony offenders. For examples of jail cases, see cases cited infra note 146. 8. See, e.g., Wyatt v. Stickney, 334 F. Supp. 1341 (M.D. Ala. 1971), 344 F. Supp. 373 (M.D. Ala. 1972), affd. in part and revd. in part sub nom. Wyatt v. Aderholt, 503 F.2d 1305 (5th Cir. 1974) (mental health and mental retardation); New York State Assoc. for Retarded Children v. Rockefeller, 357 F. Supp. 752 (E.D.N.Y. 1973), New York State Assoc. for Retarded Children v. Carey, 393 F. Supp. 715 (E.D.N.Y. 1975) (mental retardation). 9. Other examples include housing authorities, see Gautreaux v. Chicago Hous. Auth., 296 F. Supp. 907 (N.D. Ill. 1969); the federal Department of Health, Education and Welfare’s Office for Civil Rights, see Adams v. Richardson, 356 F. Supp. 92 (D.D.C. 1973), affd., 480 F.2d 1159 (D.C. Cir. 1973); and a wide variety of public employers, see, e.g., Kirkland v. New York State Dept. of Correctional Servs., 374 F. Supp. 1361 (S.D.N.Y. 1974), affd., 520 F.2d 420 (2d Cir. 1975). 10. See Abram Chayes, The Role of the Judge in Public Law Litigation, 89 HARV. L. REV. 1281 (1976); FISS, INJUNCTIONS, supra note 5, at 415-81; OWEN M. FISS, THE CIVIL RIGHTS INJUNCTION (1978); Owen M. Fiss, Foreword: The Forms of Justice, 93 HARV. L. REV. 1 (1979) [hereinafter Fiss, The Forms of Justice]. Chayes’s piece was, at last survey, the sixth or seventh most frequently cited of all law journal articles. See James E. Krier & Stewart J. Schwab, The Cathedral at Twenty-five: Citations And Impressions, 106 YALE L.J. 2121, 2136, 2139 (1997); Fred R. Shapiro, The Most-Cited Law Review Articles Revisited, 71 CHI.-KENT L. REV. 751, 767 (1996); see also Richard L. Marcus, Public Law Litigation and Legal Scholarship, 21 U. MICH. J.L. REFORM 647 (1988). 1996 Michigan Law Review [Vol. 97:1994 scribed and defended civil rights injunctive cases in opposition to Lon Fuller’s vision of private dispute resolution by adversarial litigation,11 and both took as their central concern the role of the judge. The many siblings of Chayes’s and Fiss’s work,12 and its numerous progeny,13 have, with some exceptions,14 shared these two features. Malcolm Feeley and Edward Rubin’s history and analysis of prison reform litigation, Judicial Policy Making and the Modern State: How the Courts Reformed America’s Prisons, is a work in this tradition. Like other scholars since the 1970s concerned with structural reform cases, Feeley and Rubin aim to rebut Fuller by “rethink[ing] the forms and limits of adjudication” (p. 3). And like other scholars since the 1970s, Feeley and Rubin pay most attention to judges, although they narrow the focus even further, to judges’ creation of legal doctrine (albeit broadly defined).15 11. See Lon L. Fuller, The Forms and Limits of Adjudication, 92 HARV. L. REV. 353 (1978) (originally written and circulated in 1957); see also Melvin Aron Eisenberg, Participation, Responsiveness, and the Consultative Process: An Essay for Lon Fuller, 92 HARV. L. REV. 410, 431 (1978) (“The development of public law litigation challenges in an important way Fuller’s view of the limits of litigation. Chayes’s article marks the beginning of an effort to rationalize this development . . . .”). 12. See, e.g., DONALD L. HOROWITZ, THE COURTS AND SOCIAL POLICY (1977); Ralph Cavanagh & Austin Sarat, Thinking About Courts: Toward and Beyond a Jurisprudence of Judicial Competence, 14 L. & SOCY. REV. 371 (1980) [hereinafter Cavanagh & Sarat, Beyond Judicial Competence]; Colin S. Diver, The Judge as Political Powerbroker: Superintending Structural Change in Public Institutions, 65 VA. L. REV. 43 (1979) [hereinafter Diver, Judge as Powerbroker]; William A. Fletcher, The Discretionary Constitution: Institutional Remedies and Judicial Legitimacy, 91 YALE L.J. 635 (1982) [hereinafter Fletcher, Discretionary Constitution]; Gerald E. Frug, The Judicial Power of the Purse, 126 U. PA. L. REV. 715 (1978); Donald L. Horowitz, Decreeing Organizational Change: Judicial Supervision of Public Institutions, 1983 DUKE L.J. 1265; Robert F. Nagel, Controlling the Structural Injunction, 7 HARV. J.L. & PUB. POLY. 395 (1984); Judith Resnik, Managerial Judges, 96 HARV. L. REV. 374 (1982) [hereinafter Resnik, Managerial Judges]. 13. See, e.g., PHILLIP J. COOPER, HARD JUDICIAL CHOICES: FEDERAL DISTRICT COURT JUDGES AND STATE AND LOCAL OFFICIALS (1988); GERALD N. ROSENBERG, THE HOLLOW HOPE: CAN COURTS BRING ABOUT SOCIAL CHANGE (1991); Barry Friedman, When Rights Encounter Reality: Enforcing Federal Remedies, 65 S. CAL. L. REV. 735 (1992). Like Colin Diver, fifteen years before her, see Diver, Judge as Powerbroker, supra note 12, Susan Sturm has added a full cast of characters to the judge-centered action, but retains the judge and appropriate judicial role as her primary focus. See Susan P. Sturm, Lawyers at the Prison Gates: Organizational Structure and Corrections Advocacy, 27 U. MICH. J.L. REFORM 1 (1993) [hereinafter Sturm, Lawyers at the Prison Gates]; Susan P. Sturm, The Legacy and Future of Corrections Litigation, 142 U. PA. L. REV. 639 (1993) [hereinafter Sturm, Legacy and Future]; Susan P. Sturm, The Promise of Participation, 78 IOWA L. REV. 981 (1993); Susan Sturm, A Normative Theory of Public Law Remedies, 79 GEO. L.J. 1355 (1991) [hereinafter Sturm, Normative Theory]; Susan Sturm, Resolving the Remedial Dilemma: Strategies of Judicial Intervention in Prisons, 138 U. PA. L. REV. 805 (1990); Susan P. Sturm, Note, “Mastering” Intervention in Prisons, 88 YALE L.J. 1062 (1979). 14. See, e.g., JENNIFER L. HOCHSCHILD, THE NEW AMERICAN DILEMMA: LIBERAL DEMOCRACY AND SCHOOL DESEGREGATION (1984); JAMES B. JACOBS, The Prisoners’ Rights Movement and Its Impacts, in NEW PERSPECTIVES ON PRISONS AND IMPRISONMENT 33 (1983); PETER H. SCHUCK, SUING GOVERNMENT: CITIZEN REMEDIES FOR OFFICIAL WRONGS (1983); Sturm, Lawyers at the Prison Gates, supra note 13. 15. Other scholars deal more with judges’ roles in case management and negotiation, and frequently separate remediation from other kinds of decisionmaking. See, e.g., Diver, Judge May 1999] Judicial Policymaking 1997 Feeley and Rubin use the first fifteen years of systemic prison reform litigation as both context and source for a theoretical description and legitimation of a judicial activity they contrast to interpretation and label policymaking — “the process by which [judges] exercise power on the basis of their judgment that their actions will produce socially desirable results” (p. 5). The purpose of the authors’ “microanalysis” of prison cases is to understand judges and how they are motivated and constrained and to construct a “theory of judicial policy making from the different, complex features that the [prison litigation] example offers.”16 Feeley and Rubin’s most basic point is that “judicial policy making [is] a separate judicial function with its own rules, its own methods, and its own criteria for measuring success or failure” (p. 3), and their book proceeds to describe these elements (pp. 380-81). It is Feeley and Rubin’s grand design to transform our vision of judging and of law by expanding it to include policymaking as well as interpretation. Their subsidiary goals are to describe the history of prison reform litigation in this country; to present a sociological description of the “institutional phenomenology of judicial decision making” (p. 212); to persuade readers that federalism and separation of powers have no normative (and, as a result, little positive) force; and to recast the concept of the rule of law. Succeeding in any of these projects is worth a book; shedding as much light on all of them as the authors do is a major achievement. The strengths of this work are formidable. It is well written, interesting, nuanced, and erudite. The authors’ account of pre-1960 prison cases (pp. 30-34) is itself a brief but important historical contribution. Their normative analysis of federalism (pp. 171-203) is creative and provocative and has garnered sustained scholarly attention elsewhere.17 Particularly insightful is the discussion, in the final chapter, of the abiding paradox of litigated prison reform: even if litigation has eliminated the worst abuses — the Tucker teleas Powerbroker, supra note 12, at 50-53, 62-64, 80; Fletcher, Discretionary Constitution, supra note 12; Resnik, Managerial Judges, supra note 12; Sturm, Normative Theory, supra note 13. 16. P. 211. I agree entirely with the authors on the appeal of such “microanalyses.” See also Edward L. Rubin, The New Legal Process, the Synthesis of Discourse, and the Microanalysis of Institutions, 109 HARV. L. REV. 1393, 1424, 1425, 1425-26 (1996) (calling for a “new synthesis for legal scholarship” with a “new unified methodology” — “the microanalysis of institutions”). 17. The authors published this section of the book five years ago, with a somewhat different framework, as an article that has provoked much discussion. See Edward L. Rubin & Malcolm Feeley, Federalism: Some Notes on a National Neurosis, 41 UCLA L. REV. 903 (1994). For responses to the article, see, for example, DAVID L. SHAPIRO, FEDERALISM: A DIALOGUE 7 nn.25-26, 107 n.1 (1995); Barry Friedman, Valuing Federalism, 82 MINN. L. REV. 317 (1997); Vicki C. Jackson, Federalism and the Uses and Limits of Law: Printz and Principle?, 111 HARV. L. REV. 2180, 2216-23 (1998); Daniel B. Rodriguez, State Constitutional Theory And Its Prospects, 28 N.M. L. REV. 271, 288 (1998); John C. Yoo, Sounds of Sovereignty: Defining Federalism in the 1990s, 32 IND. L. REV. 27, 43 (1998). 1998 Michigan Law Review [Vol. 97:1994 phones,18 the bread and water diets, the complete failure to provide medical care, and the authorized violence of convict “trusties” assigned to guard and punish other inmates — its outcome may nonetheless not be one that inmates or their advocates would have chosen. As the authors explain, The modern constitutional prison is a mixed blessing . . . . Conditions and practices are much improved and the constitutionalization of the process assures that these improvements are likely to be permanent. But the mission of prisons and jails remains safety and security by means of a tight system of control. Judicial reform has, on balance, enhanced the ability of officials to pursue this mission: they are now more, not less, effective and efficient. As such, the courts may have contributed to an increased willingness to rely on prisons and even to the increasing oppressiveness that results from the development of supermaximum institutions.19 For all its virtues, however, Feeley and Rubin’s exploration is flawed by two mismatches between the litigation underlying their theory and the theory itself. I take issue, first and probably less significantly, with their strong insistence that the prison cases amount to “policymaking” untethered to the Constitution’s text. The authors consider the Eighth Amendment’s prescription against “cruel and unusual punishments” a basically contentless cipher that acts only to give courts “jurisdiction” over policy disputes (pp. 14, 146); they present in support of this contention evidence that judges presiding over prison cases thought morality and national practice relevant considerations in determining the scope of the Eighth Amendment. They are persuasive on the factual point that morality and national practice played a role in convincing judges to hold some prisons constitutionally liable to inmates, but not on the theo18. As Feeley and Rubin describe, the Tucker telephone was a torture device used to punish inmates in Arkansas; it was attached by electrodes to a prisoner’s extremities (including his genitals), and guards would use its hand crank to generate electricity. See p. 56 n.*. 19. P. 375. Based on my experience as a prison and jail litigator (at the U.S. Department of Justice Civil Rights Division), I agree with Feeley and Rubin that, at least, well-conceived and well-executed prison litigation can be instrumental in turning around troubled facilities. It is not this review’s purpose (as it is not Feeley and Rubin’s effort) to evaluate if prison cases have, overall, led to “better” prisons, whether that means facilities that are more humane, safer, more orderly, or more successful at some project such as rehabilitation or deterrence, or whether the costs of any improvements were justified. Such an evaluation would be extraordinarily difficult, both practically and theoretically. See pp. 362-66. I would, however, second Feeley and Rubin’s worry that by promoting the comforting idea of the “lawful prison,” the litigation movement may have smoothed the way for ever-harsher sentences and criminal policies and contributed to the current situation, in which our prisons and jails confine over 1.8 million people at last count — .66% of the nation’s total population. See Darrell K. Gilliard, Prison and Jail Inmates at Midyear 1998, in U.S. DEPT. OF JUSTICE, BUREAU OF JUSTICE STATISTICS BULLETIN (March 1999, NCJ 173434) (on June 30, 1998, the nation’s prisons and jails incarcerated approximately 1,802,496 persons); U.S. Bureau of the Census, Monthly Estimates of the United States Population: April 1, 1980 to May 1, 1999 (internet release June 25, 1999) <http://www.census.gov/population/estimates/nation/intfile1-1.txt> (estimated population on July 1, 1998 was 270,299,000). May 1999] Judicial Policymaking 1999 retical claim that these liability assessments amounted to something quite different from constitutional interpretation. I think it is quite within the bounds of interpretation, traditionally defined, for judges to read the constitutional words “cruel and unusual” to forbid purposeful (or deliberately indifferent) infliction of pain on prisoners — by torture, starvation, denial of medical care, failure to protect from known dangers of violence from other inmates, or excessive force. For judges to reach this conclusion, in the early 1970s and today, by evaluating what they learn from litigants about conditions in defendants’ facilities in light of a conception of national morality and prison practices, seems similarly reasonable. “Cruel,” after all, is a word with moral content,20 and “unusual” is best read in a national charter of rights to direct a national comparison. In any event, this dispute is not the focus of this review, because it is indisputable that where judges continued after assessing liability to craft remedial orders, those orders were not similarly “interpretive” of the Constitution. As in most areas of injunctive practice, design of prison remedies requires the kinds of instrumental judgments that we typically label “policy,” as decree-drafters decide how to bring about institutional changes that will ensure that the rights at stake are respected in the future.21 Even further lessening the impact of this disagreement is that much of what Feeley and Rubin describe as the “rights” announced in the prison cases (p. 320 n.*), I would say were part of the admittedly instrumental “remedies.” For example, where Feeley and Rubin argue that federal judges announced that nonbureaucratic prisons violated the Constitution (pp. 271-90), I would say, rather, that judges more typically first found that certain prisons violated the Constitution, and then imposed bureaucratizing solutions to the problems. So I agree with Feeley and Rubin that the prison cases are sensibly thought of as “policymaking” in part if not in whole. But Feeley and Rubin’s analysis has a deeper flaw, which is the major subject of this review (following brief summaries of the early history of the litigation and of the authors’ theoretical framework). The history of litigated prison reform reveals it to be an intricate set of interactions framed by the rules of litigation and involving many groups, with varying roles, interests, and constraints. Feeley and Rubin’s theory, however, almost exclusively concerns the sole institution of the judiciary, and even more narrowly, the judicial activity of doctrine creation. The authors fail to assess the significance of the larger context (or, as they might prefer, the “institution”) of the 20. See, e.g., Richard H. Fallon, Jr., A Constructivist Coherence Theory of Constitutional Interpretation, 100 HARV. L. REV. 1189, 1205 (1987). 21. See, e.g., Fletcher, Discretionary Constitution, supra note 12. For a description of the rules limiting remedial orders, in both settled and contested situations, see infra text accompanying notes 56-65. 2000 Michigan Law Review [Vol. 97:1994 litigation, and their theory fails to reckon with litigation realities such as the burden of proof; the resources, goals, and strategies of counsel; or the difference between settled and litigated outcomes. Yet these are unmistakably important to the cases. It may be that there are other areas of criminal or civil litigation which have been more exclusively judge-driven (the Fourth Amendment law of policing, perhaps, or abortion law). But Feeley and Rubin ask a question of the prison cases — How do judges make policy? — that these cases, at least, cannot answer. The authors do not engage what is for me not only the more interesting but more appropriate question: How do courts function as an arena of policy disputation? I. PRISON LITIGATION A. AND ITS LESSONS History Feeley and Rubin begin their account with an excellent history of early prison litigation, which I briefly recap here (with supplementation). Until the 1960s, federal judges almost invariably refused to intervene in civil cases about prison conditions or the institutional rules to which federal and state inmates were subjected.22 In taking this “hands-off” approach, judges explained that the judicial role simply did not encompass prison reform.23 The 22. See pp. 30-37. An isolated early exception was In re Birdsong, in which Emory Speer, a federal judge in Georgia, held that a federal prisoner’s Eighth Amendment rights had been violated by a county jailer who chained him by the neck to a grating in his cell at night “so that he could not put his heels to the ground.” 39 F. 599, 602 (S.D. Ga. 1889). Speer ordered the prisoner released from this torture. For biographical information on Judge Speer, a Confederate veteran and a stalwart critic of the slave-like peonage and chain-gang systems under which laborers were forcibly conscripted throughout the South, see Benno C. Schmidt, Jr., Juries, Jurisdiction, and Race Discrimination: The Lost Promise of Strauder v. West Virginia, 61 TEXAS L. REV. 1401, 1484 (1983); Benno C. Schmidt, Jr., Principle and Prejudice: The Supreme Court and Race in the Progressive Era. Part 2: The Peonage Cases, 82 COLUM. L. REV. 646, 669-71 (1982) [hereinafter Schmidt, Peonage Cases]. A more significant but nonetheless isolated precursor to the cases of the 1960s was Coffin v. Reichard, 143 F.2d 443, 444 (6th Cir. 1944), in which an inmate alleged that he was being “subjected to assaults, cruelties and indignities from guards and his co-inmates” and the Sixth Circuit Court of Appeals held that the writ of habeas corpus could be granted to remedy unlawful conditions of confinement. The court commented in a much repeated formulation that “[a] prisoner retains all the rights of an ordinary citizen except those expressly, or by necessary implication, taken from him by law.” Id. at 445. 23. See pp. 36-37; Note, Beyond the Ken of the Courts: A Critique of Judicial Refusal to Review the Complaints of Convicts, 72 YALE L.J. 506 (1963) (popularizing the phrase “hands off,” and listing and quoting cases). The language of the hands-off cases emphasized noninterference rather than, as courts had earlier done, that prisoners had no rights to be respected by prison officials. See Ruffin v. Commonwealth, 62 Va. 790, 796 (1871) (naming prisoners “slave[s] of the State”); Ex parte Sherwood, 29 Tex. App. 334, 15 S.W. 812 (Tex. Ct. App. 1890) (same). Thus, in a sense, the hands-off principle was one of “underenforcement” of constitutional norms, rather than rejection of the concept of inmates’ constitutional rights. See Lawrence Gene Sager, Fair Measure: The Legal Status of Underenforced Constitutional Norms, 91 HARV. L. REV. 1212 (1978) (arguing that courts “underenforce” certain constitutional rights because of institutional concerns like judicial capacity, but that government offi- May 1999] Judicial Policymaking 2001 first serious hole in the solid barrier of the hands-off policy came in 1941, in Ex parte Hull, in which the Supreme Court prohibited prison officials from screening inmates’ habeas corpus petitions prior to forwarding them to a court.24 But while Hull allowed pleas for relief from abusive conditions to arrive, in habeas petitions and in other kinds of filed complaints, judges declined to answer those pleas for nearly twenty years.25 Many of the obstacles to judicial oversight of prisons were doctrinal. Before courts could plausibly undertake to reform prisons, numerous questions had to be resolved: whether and which guarantees of the Bill of Rights govern state as well as federal officials;26 whether an action for damages or injunctive relief (other than release from prison) could be brought against state or local officials under 42 U.S.C. § 1983;27 whether inmates would be required to exhaust state remedies prior to bringing such an action.28 Gradually, these questions did get resolved, each in favor of judicial power, and judges began to intervene, rather than expressing regret that they could not (pp. 34-39). cials are nonetheless legally obligated to respect such rights fully); see also Lewis v. Casey, 518 U.S. 343, 349 (1996) (“[I]t is not the role of courts, but that of the political branches, to shape the institutions of government in such fashion as to comply with the laws and the Constitution.”). 24. 312 U.S. 546 (1941); see also Cochran v. Kansas, 316 U.S. 255 (1941) (state must allow inmates to file appellate papers from prison); Burns v. Ohio, 360 U.S. 252, 257 (1959) (state must allow indigent inmate to file appeal without payment of fees); Smith v. Bennett, 365 U.S. 708 (1961) (same, for habeas petition); Johnson v. Avery, 393 U.S. 483 (1969) (striking down prison regulation prohibiting prisoners from assisting each other with habeas corpus applications and other legal matters); Wolff v. McDonnell, 418 U.S. 539, 577-80 (1974) (extending protection of jailhouse lawyers to cover assistance in civil rights actions); Bounds v. Smith, 430 U.S. 817, 828 (1977) (“[T]he . . . right of access to the courts requires prison authorities to assist inmates in the preparation and filing of meaningful legal papers by providing prisoners with adequate law libraries or adequate assistance from persons trained in the law.”). 25. Federal courts did allow the Justice Department to bring criminal civil rights prosecutions of prison guards for their brutal treatment of inmates; the cases sometimes described the charge as “summary punishment.” See United States v. Best, Crim. No. 13,157 (D. Colo. Apr. 2, 1952); United States v. Irby, Crim. No. 19072, (E.D.S.C. Feb. 23, 1952)); United States v. Jones, 207 F.2d 785 (5th Cir. 1953); United States v. Walker, 216 F.2d 683 (5th Cir. 1954); United States v. Jackson, 235 F.2d 925 (8th Cir. 1956). For descriptions of the Justice Department’s 1950s initiative to prosecute prison brutality cases, see JOHN T. ELLIFF, THE UNITED STATES DEPARTMENT OF JUSTICE AND INDIVIDUAL RIGHTS, 1937-1962, at 304-09 (1987) (originally written in 1967 as Ph.D. dissertation, Harvard University); Arthur B. Caldwell & Sydney Brodie, Enforcement of the Criminal Civil Rights Statute, 18 U.S.C. Section 242, in Prison Brutality Cases, 52 GEO. L.J. 706 (1964) (describing the above-cited cases). 26. P. 37. See Palko v. Connecticut, 302 U.S. 319, 325 (1937) (holding that only those guarantees “implicit in the concept of ordered liberty” are applicable against the states). The Eighth Amendment was not held applicable against the states until 1962, in Robinson v. California, 370 U.S. 660 (1962). For a description of the varying ways the Supreme Court has approached this issue, see Duncan v. Louisiana, 391 U.S. 145 (1968). 27. P. 37. See Monroe v. Pape, 365 U.S. 167 (1961). 28. See Pierce v. La Vallee, 293 F.2d 233 (2d Cir. 1961) (no exhaustion required); Houghton v. Shafer, 392 U.S. 639 (1968) (same). 2002 Michigan Law Review [Vol. 97:1994 Two groups of claims in the early 1960s, especially, enlisted court involvement.29 First, Black Muslim inmates attacked widespread prison policies that denied them access to religious literature, clergy, and services, but granted similar requests by Christians and adherents of other religions.30 Supported by the Black Muslim organization on the outside, inmates around the country brought lawsuits; they were almost immediately successful in the Supreme Court, which in 1964 handed down a one paragraph per curiam opinion in Cooper v. Pate,31 tersely reversing the Seventh Circuit’s dismissal of one such claim (pp. 37-38). The second group of cases (unmentioned by Feeley and Rubin) was tied directly to the project of desegregating public facilities. In the early to mid-1960s, African Americans, especially in the South, sought to realize in facilities other than schools the desegregative promise of Brown v. Board of Education.32 Correctional facilities were not exempt from that effort; civil rights plaintiffs and lawyers targeted them both in omnibus suits seeking to desegregate a range of public facilities,33 and in 29. See, e.g., JIM THOMAS, PRISONER LITIGATION: THE PARADOX OF THE JAILHOUSE LAWYER 81-92 (1988) [hereinafter THOMAS, PRISONER LITIGATION] (summarizing trends that influenced upsurge in prisoner litigation in the 1960s and 1970s). In addition, a third type of case involved only federal prisoners in federal prisons. See United States v. Muniz, 374 U.S. 150 (1963) (holding that federal inmates can sue under the Federal Tort Claims Act for injuries suffered while in federal custody); Eugene N. Barkin, The Emergence of Correctional Law and the Awareness of the Rights of the Convicted, 45 NEB. L. REV. 669, 686-89 (1966) (article by legal counsel for the federal Bureau of Prisons describing early impact of Muniz). 30. See pp. 37-38. For a summary and analysis of the earliest Black Muslim cases, see Comment, Black Muslims in Prison: Of Muslim Rites and Constitutional Rights, 62 COLUM. L. REV. 1488 (1962). 31. 378 U.S. 546 (1964). 32. See, e.g., Holmes v. City of Atlanta, 223 F.2d 93 (5th Cir.) (forbidding segregation of public golf courses), affd. per curiam, 350 U.S. 879 (1955); Mayor of Baltimore v. Dawson, 220 F.2d 386 (4th Cir.) (public beaches), affd. per curiam, 350 U.S. 877 (1955); Gayle v. Browder, 142 F. Supp. 707 (M.D. Ala.) (buses), affd. per curiam, 352 U.S. 903 (1956); New Orleans City Park Improvement Assn. v. Detiege, 252 F.2d 122 (5th Cir.) (parks), affd. per curiam, 358 U.S. 54 (1958). 33. See Wood v. Vaughan, 209 F. Supp. 106 (W.D. Va. 1962) (omnibus suit seeking to desegregate all public facilities in Lynchburg, Virginia, including the city jails); Coleman v. Aycock, 304 F. Supp. 132 (N.D. Miss. 1969) (granting injunction requiring desegregation of county jail in Belzoni, Mississippi, in lawsuit “encompass[ing] practically all public facilities operated by the county and city and many of the services rendered by the municipality”); Palmer v. Thompson, 391 F.2d 324 (5th Cir. 1967) (denying standing as to the city jail in Jackson, Mississippi, to noninmate African-American citizens bringing omnibus public facilities desegregation suit), adhered to without comment on rehg. en banc, 419 F.2d 1222 (5th Cir. 1969), affd. on other grounds, 403 U.S. 217 (1971). The lawyers in these cases were generally very involved in civil rights matters, but not (so far as I can tell), affiliated with any kind of national organization with an overall litigating strategy. Indeed, the general counsel of the NAACP Legal Defense and Education Fund, the principal institutional national desegregation litigator, did not think that omnibus suits were likely to be effective. See JACK GREENBERG, CRUSADERS IN THE COURTS: HOW A DEDICATED BAND OF LAWYERS FOUGHT FOR THE CIVIL RIGHTS REVOLUTION 352-53 (1994) [hereinafter GREENBERG, CRUSADERS]. May 1999] Judicial Policymaking 2003 suits focusing on jails and prison more particularly.34 But it was not until later in the 1960s that federal judges began to move beyond claims about in-prison violation of generally applicable constitutional guarantees (such as the Equal Protection Clause or the First Amendment) and to entertain seriously the claim that the Eighth Amendment’s prescription against “cruel and unusual punishments” might provide a judicially enforceable right to at least minimally adequate prison conditions. The first such cases involved prison discipline — corporal punishment and conditions in disciplinary isolation — presumably because these were easiest to conceptualize as “punishment” separable from the sentence of incarceration.35 But soon, faced with sometimes uncontested proof of brutal and unhealthful jail and prison environments not just in isolation cells but throughout facilities, judges began to find that such conditions also violated the constitutional rights of inmates36 and to issue injunctive orders requiring remediation of the unconstitutional practices. The first case to require wholesale reform of a prison occurred in Arkansas, where by 1970 a federal district judge undertook to reform not just one institution, but the 34. See, e.g., Bolden v. Pegelow, 329 F.2d 95 (4th Cir. 1964) (requiring integration of the District of Columbia’s Lorton prison’s barber shops); Washington v. Lee, 263 F. Supp. 327 (M.D. Ala. 1966) (desegregating all penal and detention facilities in Alabama), affd. per curiam, 390 U.S. 333 (1968); Toles v. Katzenbach, 385 F.2d 107 (9th Cir. 1967) (suit seeking the cessation of Federal Bureau of Prisons segregation in Washington prison; relief denied after the Bureau responded to the lawsuit by issuing a policy statement forbidding racial distinctions in housing, work, or program assignments); Wilson v. Kelley, 294 F. Supp. 1005 (N.D. Ga. 1968) (requiring desegregation of all Georgia jails and prisons); Mason v. Peyton, Civ. No. 5611-R (E.D. Va. order entered Oct. 16, 1969) (desegregating Virginia’s prisons); Pounds v. Theard, 230 So. 2d 861 (La. App. 4th Cir. 1970) (desegregating New Orleans city jail); McClelland v. Sigler, 327 F. Supp. 829 (D. Neb. 1971) (desegregating Nebraska state prison), affd., 456 F.2d 1266 (8th Cir. 1972); Guthrie v. Evans, Civ. A. No. CV 3068 (S.D. Ga. Apr. 1974) (desegregating Georgia State Prison), contempt motion settled, 93 F.R.D. 390, 39193 (S.D. Ga. 1981) (describing procedural history); cases cited infra notes 119 & 124; see also Nichols v. McGee, 169 F. Supp. 721 (N.D. Cal. 1959) (denying relief to inmate who sought to desegregate California’s Folsom prison). 35. See, e.g., Jordan v. Fitzharris, 257 F. Supp. 674 (N.D. Cal. 1966) (finding that conditions in isolation constituted cruel and unusual punishment); Wright v. McMann, 257 F. Supp. 739 (N.D.N.Y. 1966) (denying relief on similar claim), revd., 387 F.2d 519, 521-26 (2d Cir. 1967), on remand, 321 F. Supp. 127 (N.D.N.Y. 1970), affd. in part and revd. in part, 460 F.2d 126 (2d Cir. 1972); Jackson v. Bishop, 404 F.2d 571, 579-80 (8th Cir. 1968) (holding whipping of prisoners unconstitutional); see also Fulwood v. Clemmer, 206 F. Supp. 370, 37879 (D.D.C. 1962) (holding confinement in “strip cell” for prison rule violations unconstitutional because disproportional to offense); cf. Johnson v. Glick, 481 F.2d 1028 (2d Cir. 1973) (Friendly, J.) (discussing scope of Cruel and Unusual Punishments Clause’s application to prison conditions, and analyzing previous cases). 36. Convicts’ Eighth Amendment protections correspond to similar protections for pretrial detainees provided by the Fourteenth Amendment. See Ingraham v. Wright, 430 U.S. 651, 671 n.40 (1977) (“Eighth Amendment scrutiny is appropriate only after the State has complied with the constitutional guarantees traditionally associated with criminal prosecutions. . . . Where the State seeks to impose punishment without such an adjudication, the pertinent constitutional guarantee is the Due Process Clause of the Fourteenth Amendment.”). 2004 Michigan Law Review [Vol. 97:1994 entire penal system (p. 39). Arkansas had a fairly small system: just two facilities, one for blacks and one for whites, housing a total of about 1600 inmates.37 But the case, Holt v. Sarver,38 augured a nationwide flood of class-action lawsuits leading to major court orders requiring reform in such areas as housing conditions, security, medical care, mental health care, sanitation, nutrition, and exercise.39 By 1984 (the first year for which data are accessible), 24% of the nation’s 903 state prisons (including at least one in each of forty-three states and the District of Columbia) reported to the federal Bureau of Justice Statistics that they were operating under a court order. In 1983 (the first year these data exist for jails), 15% of the nation’s 3338 jails (including at least one in all but two of the forty-five states that had jails, and the District of Columbia) reported court orders.40 Litigants had been particularly active — or particularly successful — in large facilities: the prisons under court order housed 42% of the nation’s state prisoners, and the jails under court order housed 44% of the nation’s jail inmates, and for both jails and state prisons, about half of the nation’s largest facilities were under court order.41 The series of Arkansas cases that led up to Holt, and Holt itself, are the subject of the first of five “case studies” Feeley and Rubin 37. See NATIONAL CRIMINAL JUSTICE INFORMATION & STATISTICS SERVICE, U.S. DEPT. JUSTICE, NO. SD-NPS-PSF-1, PRISONERS IN STATE AND FEDERAL INSTITUTIONS FOR ADULT FELONS ON DECEMBER 31, 1971, 1972, AND 1973, at 12 tbl.1 (1975). The total number of prisoners in custody in the nation was 198,000; Arkansas ranked 32nd in number of prisoners, and 22nd in incarceration rate per 1000 population. Id. at 18 tbl.4. 38. Many reported opinions chronicle the course of the litigation. Holt v. Sarver, 309 F. Supp. 362 (E.D. Ark. 1970), affd., 442 F.2d 304 (8th Cir. 1971), on remand sub nom. Holt v. Hutto, 363 F. Supp. 194 (E.D. Ark. 1973), affd. in part and revd. in part sub nom. Finney v. Arkansas Bd. of Correction, 505 F.2d 194 (8th Cir. 1974), on remand sub nom. Finney v. Hutto, 410 F. Supp. 251 (E.D. Ark. 1976), affd., 548 F.2d 740 (8th Cir. 1977), affd., 437 U.S. 678 (1978), Finney v. Marbry, 455 F. Supp. 756 (E.D. Ark. 1978), 458 F. Supp. 720 (E.D. Ark. 1978) (approving consent decree), 528 F. Supp. 567 (E.D. Ark. 1981), 534 F. Supp. 1026 (E.D. Ark. 1982), 546 F. Supp. 626 (E.D. Ark. 1982), 546 F. Supp. 628 (E.D. Ark. 1982) (finding of compliance and dismissal of action). 39. Concurrently, courts frequently began to find that prison disciplinary procedures (or their absence) violated the Due Process Clause. The landmark prison due process case in the lower courts was Landman v. Royster, 333 F. Supp. 621 (E.D. Va. 1971); see also Wolff v. McDonnell, 418 U.S. 539 (1974). 40. Data are derived from BUREAU OF JUSTICE STATISTICS, U.S. DEPT. OF JUSTICE, CENSUS OF STATE ADULT CORRECTIONAL FACILITIES, 1984 (1988) (ICPSR 8444) [hereinafter BUREAU OF JUSTICE STATISTICS, 1984 PRISON CENSUS] and BUREAU OF JUSTICE STATISTICS, U.S. DEPT. OF JUSTICE, NATIONAL JAIL CENSUS, 1983 (1987) (ICPSR 8203) [hereinafter BUREAU OF JUSTICE STATISTICS, 1983 JAIL CENSUS] (raw data for both available from the National Archive of Criminal Justice Data (visited June 1, 1999) <http://www.icpsr.umich.edu/ NACJD/archive.html>). 41. In 1983, 175 jails (5% of all jails), each with average daily population over 257, housed 50% of the nation’s total jail population; of these jails 92 (53%) reported court orders. In 1984, 105 prisons (12% of all state prisons), each with average daily population over 991, housed 50% of the nation’s total state prison population; of these prisons 52 (50%) reported a court order. Data are derived from BUREAU OF JUSTICE STATISTICS, 1983 JAIL CENSUS, supra note 40; BUREAU OF JUSTICE STATISTICS, 1984 PRISON CENSUS, supra note 40. OF May 1999] Judicial Policymaking 2005 include in their early chapters to deepen their description of the prison reform cases (pp. 51-79). The second case study is of the Ruiz litigation,42 which concerns the mammoth Texas prison system (pp. 80-95). The authors also provide descriptions of Ramos v. Lamm,43 which shut down Colorado’s “Old Max” maximum security prison (pp. 96-111), Branson v. Winter,44 a California state court litigation about conditions in the Santa Clara County jails (pp. 11128), and a series of cases about conditions and policies at the United States Penitentiary at Marion, Illinois (pp. 128-43). As with all case studies, it is difficult to know how representative these are in any given respect.45 Nonetheless, the case studies are useful and concrete descriptions of a varied set of reform litigations. B. Policymaking When Feeley and Rubin talk about “prison reform cases,” they mean the kinds of cases that are discussed in their five case studies — injunctive actions brought pursuant to the Eighth Amendment.46 They summarize the course of the litigation movement: 42. When Ruiz began, it was captioned Ruiz v. Estelle. It remains an ongoing litigation, and (at the time this review went to press) most recently appeared in the reporters as Ruiz v. Johnson, 37 F. Supp.2d 855 (S.D. Tex. 1999); see id. at 862-69 for a description of the procedural history. 43. Ramos v. Lamm, 485 F. Supp. 122 (D. Colo. 1979), affd. in part and revd. in part, 639 F.2d 559 (10th Cir. 1980), on remand, 520 F. Supp. 1059 (D. Colo. 1981). 44. Branson v. Winter (Super. Ct. County of Santa Clara, No. 78807). 45. Indeed, we know that the Arkansas litigation, as the first wholesale reform case, and the Texas litigation, as the largest and one “more troublesome” than most, Malcolm M. Feeley & Roger A. Hanson, The Impact of Judicial Intervention on Prisons and Jails: A Framework for Analysis and a Review of the Literature, in COURTS, CORRECTIONS, AND THE CONSTITUTION: THE IMPACT OF JUDICIAL INTERVENTION ON PRISONS AND JAILS 12, 21 (John J. DiIulio, Jr. ed., 1990), are not representative. This, of course, is not lost on Feeley and Rubin. See pp. 28-29. Also, of the five case studies, all but the Santa Clara case included judicial resolution of the question of liability, a litigation record that is highly unusual. See infra text accompanying notes 66-72; see also Feeley & Hanson, supra, at 42 (calling for “more careful research” that does not “rest[ ] on a limited number of cases, not necessarily representative of the world at large”). 46. Damage actions by and on behalf of prisoners are themselves a complex and interesting subject, worthy of more scholarly attention than they have received. At last published count, inmate civil rights petitions amounted to 11% of the federal civil docket. ADMINISTRATIVE OFFICE OF THE UNITED STATES COURTS, JUDICIAL BUSINESS OF THE UNITED STATES COURTS: 1997 REPORT OF THE DIRECTOR 128-29 tbl.C-2. These cases, brought almost entirely by inmates without counsel, see Roger A. Hanson & Henry W.K. Daley, Challenging the Conditions of Prisons and Jails: A Report on Section 1983 Litigation, at 22 tbl.6 (Jan. 1995) (Bureau of Justice Statistics discussion paper, NCJ-151652), frequently seek damages, William Bennett Turner, When Prisoners Sue: A Study of Prisoner Section 1983 Suits in the Federal Courts, 92 HARV. L. REV. 610, 623 & n.80 (1979) [hereinafter Turner, When Prisoners Sue], and have been important both doctrinally, see, e.g., Sandin v. Connor, 515 U.S. 472 (1995); Farmer v. Brennan, 511 U.S. 825 (1994), and politically. A number of the provisions of the Prison Litigation Reform Act, Pub. L. No. 104-134, 110 Stat. 1321 (1996), were a response to the “alarming expansion in the number of frivolous lawsuits filed by State and Federal prisoners.” 141 Cong. Rec. S14413-14 (daily ed. Sept. 27, 1995) (Statement of Sen. Dole). See 42 U.S.C. § 1997a-1997h, 28 U.S.C. §§ 1915, 1915A. But (although I know of no 2006 Michigan Law Review [Vol. 97:1994 This massive intervention into state corrections was an act of judicial policy making. Over the course of a single decade, the federal courts fashioned a comprehensive set of judicially enforceable rules for the governance of American prisons. They derived these rules from existing correctional literature, sociology, and their own perceptions of political morality. Such a new code of legal rules, inspired by general moral and empirical considerations and derived from a model that had been hovering near but had not yet appeared upon any accepted agenda, is a typical product of the policy-making process, not very different from a statute or an administrative regulation. [pp. 13-14] Although Feeley and Rubin make some extravagant assertions about the uniqueness of the prison reform cases, naming them “the most striking example of judicial policy making in modern America” (p. 13), and “the high-water mark of judicial policy making” (p. 336), they nonetheless use the cases as exemplars of what they argue is a less unusual process. They posit that “the prison reform cases represent a standard mode of judicial action, that is, policy making of the same kind that the legislature or the executive pursues” (p. 146). The authors then attempt both a general description and general justification of such policymaking.47 Feeley and Rubin conceptualize the judicial practice of policymaking as having four parts: problem, goal, solution, and implementation: Problem and goal: The authors argue that policymaking activity by the federal judiciary is triggered by a perceived problem — specifically, a conflict between judges’ role, or their understanding of what the current law requires or allows, and their own moral beliefs (pp. 161-62). But such a conflict is sufficient to prompt judges to engage in policymaking, the authors add, only when the judges feel their moral beliefs to be widely, and nationally, shared (p. 352). In the prison cases, Feeley and Rubin argue, the major trigger to the earliest judicial action was the discrepancy between southern prison systems and those elsewhere (pp. 150-58). By 1965, the “plantation” prisons of the South were simply no longer morally acceptable data on this point), it seems to be extremely rare for class-action inmate litigation to include a damages component, perhaps because the kinds of harms that might justify damages are so individual as to preclude class certification. See FED. R. CIV. P. 23. But cf. Amy Beth Graves, Inmates Share $1.65 Million; Settlement Made in Suit Against Prison Operator, CINCINNATI ENQUIRER, Mar. 3, 1999, at B2 (describing cash settlement of class-action inmate lawsuit alleging unsafe conditions in private prison operated by Corrections Corporation of America). 47. Although they intended their book to legitimate litigated prison reform, Feeley and Rubin’s opus has already provided ammunition to reform’s opponents, who agree with the authors’ description of the cases as judicial policymaking without constitutional excuse, but not with their phenomenological justification of such activity by federal judges. See David Schoenbrod & Ross Sandler, By What Right Do Judges Run Prisons?, WALL ST. J., Aug. 31, 1998, at A19. For my position on the constitutional point, see supra text accompanying notes 20-21. May 1999] Judicial Policymaking 2007 to federal judges, who had grown accustomed to their part as foot soldiers in the “wide-ranging, nationally initiated attack on southern institutions that took place in the decades following World War II” (p. 159). The goal of the judges’ policymaking followed naturally from the perceived problem, Feeley and Rubin argue: the judiciary sought to impose national norms on state prisons, to stamp out the problematic variation.48 Solution: When judges make policy, the authors say, they do so by creating new legal doctrine (pp. 204-96). There are important constraints on this process, which come from the institutional structure of the judiciary and from judges’ felt institutional roles. First, judicial policymaking must be allowed by an affirmative grant of “jurisdiction” from a constitutional or statutory source; in the prison cases, the Eighth Amendment served this function (p. 206). In addition, doctrine creation is constrained by the need of each judge to operate in a way generally consistent with the broader legal discourse and persuasive and attractive to other judges.49 What is required, the authors argue, is a “coordinating idea” — one that “can be communicated to, and followed by, a large number of dispersed individuals within the judiciary” (p. 242). Only ideas that are simple, clear, and incrementally connected or related to existing legal doctrine have sufficient appeal for members of the role-bound group that forms the federal judiciary to become successful “coordinating ideas.”50 In the prison reform cases, Feeley and Rubin argue, the coordinating idea developed by judges was the concept of the “moral, legally justifiable prison” (p. 239). More particularly, the authors contend, judges adopted a two-prong solution to the policy problem created by southern prisons (and later applied this 48. See pp. 162-71. The authors contend that this kind of judicial imposition of national norms would seem to run headlong into the obstacle of federalist constraints on federal power. See pp. 171-203. But, they argue, “[i]f hundreds of federal judges were willing to reject federalism, they must have perceived or sensed that this principle is not as hallowed as our public rhetoric suggests.” P. 177. The authors conclude, after sustained attention, that, “[i]n fact, what is wrong with federalism is that it has become obsolete.” Id. For scholarly reactions to the authors’ antifederalism arguments, see sources cited supra note 17. 49. Pp. 226-52. The authors concede that their description of doctrinal creation, which does not include any determinative guidelines for any judge who is making doctrine, runs headlong into some concepts of “the rule of law.” Specifically, they note that it conflicts with a conceptualization of the rule of law that requires that “‘government in all its actions is bound by rules fixed and announced before — rules which make it possible to foresee with fair certainty how authority will use its coercive powers in given circumstances.’” P. 347 (quoting F.A. HAYEK, THE ROAD TO SERFDOM 80 (15th ed. 1994)). But Feeley and Rubin argue that this concept of the rule of law, is, like any concept of federalism as a constraint on judicial power, outdated. Instead, “[a] modern version of the rule of law . . . incorporates the concept of constraint, but jettisons the idea that the constraint must necessarily consist of fixed, preestablished rules. To locate the sources of constraint, we must examine contemporary attitudes and governmental arrangements.” P. 350. 50. This section of the book was published, in somewhat different form, as Edward Rubin & Malcolm Feeley, Creating Legal Doctrine, 69 S. CAL. L. REV. 1989 (1996). 2008 Michigan Law Review [Vol. 97:1994 solution to nonsouthern prisons): requiring, first, that prisons be designed to further rehabilitation of the prisoner, and second, that they be bureaucratically organized, capable of declaring, implementing, and observing the implementation of policy changes. Implementation: Once judges have made policy by creating doctrine, say Feeley and Rubin, they must implement that policy (pp. 297-335). Indeed, the authors name “administrative implementation” “an integral aspect of the policy-making process” (p. 299). Moreover, it is “probably the single most controversial” part (p. 299). In the prison cases, Feeley and Rubin argue, it was because the very “notion of prisoners’ rights was simply incomprehensible” to many wardens who thought of inmates as “slaves of the state” (p. 301) that judicial implementation was so “decisive[ ] and comprehensive[ ]” (p. 299). Judges got drawn into undertaking a “type of organizational therapy whose purpose was to transform the institution’s collective understanding of itself” (p. 302). The two principal strategies they used were familiar administrative methods: imposition of standards (pp. 303-07) and appointment of “special masters,” as a sort of administrative staff, used in appropriately diverse ways: “as informants, always, as consultants and advisors to the tractable, and as supervisors or punishers of the intransigent.”51 Having surveyed the components of judicial policymaking both in the prison cases and more generally, Feeley and Rubin next place such policymaking in the broader context of modern attitudes about government. They argue that the courts’ actions in the prison cases exemplify what they see as the modern conception of the role of government, in which “[t]he state is held responsible for social problems and is expected to combat them by developing new governmental programs” (p. 23). The method we as a polity have chosen for government to carry out its new duties is administrative — “the conscious, coordinated effort of a central authority that represents our entire political community” (p. 343). To tell judges that they may not make policy, or to require them to be constrained by structural principles of federalism and separation of powers, “particularly when other branches have abandoned them, [would] simply exclude[ ] the courts from the modern governmental process” (p. 344). Feeley and Rubin conclude it would be descriptively silly, 51. P. 310. Feeley and Rubin argue that such implementation appears to contradict another “constitutional fixture: the separation of powers.” P. 311. But, seeking to avoid accusing hundreds of judges of simply ignoring this contradiction (an explanation they characterized as premised on mass judicial “brain fever,” p. 217), the authors undertake an extensive review of claims that separation of powers in fact poses an obstacle to active implementation efforts. They conclude that it does not, for a variety of reasons, including that the subject entities were “states, not coordinate federal branches,” and “that the checks-andbalances doctrine pointed in the opposite direction, that social circumstances had so clearly changed, and that broad-ranging implementation powers were traditional judicial functions.” Pp. 329-30. Judicial Policymaking May 1999] 2009 and normatively wrong, to believe courts are appropriately excluded in this way. Instead, we should acknowledge that “[w]hile certain disadvantages attend their [policymaking] efforts . . . the courts have certain strengths as well, and are fully able to function in this dominant mode of modern governmental action” (p. 388). II. BEYOND JUDICIARY: THE LITIGATEDNESS LITIGATED REFORM THE OF As is evident from the above description, Judicial Policy Making and the Modern State “really combines two different books,”52 the first about litigated prison reform, and the second about judicial policymaking. In a recent review, Donald Dripps comments that the second book is not completely justified by the first — that is, that the facts of court-supervised prison reform as Feeley and Rubin present them don’t fully support their claims about the desuetude of federalism and separation of powers, and their reconceptualization of the constraint imposed by the rule of law.53 My criticism goes deeper: it is that the authors’ theory obscures rather than illuminates the facts. Feeley and Rubin’s theoretical vision is so tightly focused on judges and doctrinal creation that they seem nearly blind to most of the other relevant players and the rules and contours of other types of court action. This defect mars their account both of litigated prison reform in particular and of litigation as a realm of policy disputation and resolution in general. Obviously, it is beyond the scope of this review to complete the picture. But I can offer some, limited, description of how the first generation of prison cases were shaped not only by judicial doctrine creation, but also by the identity, goals, resources, and strategies of some of the nonjudicial participants,54 and by the forms and rules of liti52. Donald A. Dripps, Prisons, Public Policy, and Legal Theory (reviewing Judicial Policy Making and the Modern State), 2 JURIST BOOKS-ON-LAW No. 5 (May 1999) <http:// www.jurist.law.pitt.edu/lawbooks/reviews.htm>. 53. Id. 54. My approach builds on the work of Colin Diver. In a 1979 article, Diver writes of structural reform litigation as a “component of the continuous political bargaining process that determines the shape and content of public policy,” Diver, Judge as Powerbroker, supra note 12, at 45. He begins by discussing the rules of litigation, and the non-judicial players in the cases, id. at 64-76. But this description simply sets the scene for Diver’s analysis of the ways in which “institutional reform litigation . . . presents the power-conscious trial judge with numerous opportunities to influence directly the distribution of effective power within the institutional defendant,” id. at 88, and of the factors contributing to and limiting the legitimacy of such a role. Thus the judge stays at the center of the inquiry. For a discussion of individual inmate litigation that seeks to broaden the prevalent focus on judges and to cover other “players and processes,” see THOMAS, PRISONER LITIGATION, supra note 29, at 155–90, especially id. at 155 (“The general image promoted of processing prisoner litigation by media accounts and implied by critics entails a two-stage process: the prisoner writes the story, and then sends it to the judge who decides the case. This image promotes only a two-stop tour, the first offering a cursory glimpse of the prisoner’s story- 2010 Michigan Law Review [Vol. 97:1994 gation.55 I look especially at two topics — settlement and plaintiffs’ counsel. Feeley and Rubin mention both in some of their case studies, but not in their proffered interpretation and theoretical account of the prison cases and of judicial policymaking more generally. Settlement. The rules governing entry of various types of contested federal court orders set up substantial limits on the discretion of trial judges. Before awarding damages in a case, for example, a trial judge must articulate the applicable legal rules of liability and damages (subject to de novo appellate review), and she must base her decision on what she believes are the most likely facts in light of record evidence (generally subject to more deferential appellate review).56 Before issuing an injunction against a governmental entity, a trial judge must, likewise, articulate and apply a rule of liability. In addition, moreover, the injunction is subject to far more substantial limitations than the award of damages. It must simultaneously “so far as possible eliminate the [unconstitutional] effects of the past as well as bar like [unconstitutionality] in the future,”57 and yet be “no broader than necessary” to accomplish this end.58 Appellate policing of the trial judges’ chosen balance between cure and intrusion is deferential,59 but the trial judges’ obligation to attempt the contradictory task exists nonetheless. Until 1996, when Congress rewrote the rules for cases about the conditions of confinement in prisons, jails, and juvenile facilities,60 document, and the second paying brief homage before the shrine of the judicial decision. The organizational processing, however, is far more complex.”). See also Turner, When Prisoners Sue, supra note 46 (providing an empirical study of inmate litigation, including causes, processing, and results). 55. The connection between procedural rules and litigation practice and outcomes has been the topic of much interesting scholarly work. See, e.g., Stephen C. Yeazell, The Misunderstood Consequences of Modern Civil Process, 1994 WIS. L. REV. 631, 647 (arguing that the impact of the Federal Rules has been that “control of litigation has moved further down the legal food chain — from appellate to trial courts, and from trial courts to lawyers”); Carter, supra note 5 (describing how the Federal Rules impact civil rights). These rules have not been neutrally derived, but are rather themselves the result of political contests occurring in many fora, from the Congress to the trial courts to the federal Rules committees to the Supreme Court. See Roy B. Flemming, Contested Terains and Regime Politics: Thinking About America’s Trial Courts and Institutional Change, 23 L. & SOC. INQUIRY 941, 945 (1998) (describing courts as “contested terrain,” in which “courts and their processes are often . . . objects of broader political conflicts that occur outside courthouses”). 56. See, e.g., Anderson v. City of Bessemer City, 470 U.S. 564 (1985); Pullman Standard v. Swint, 456 U.S. 273 (1982). 57. Louisiana v. United States, 380 U.S. 145, 154 (1965). 58. Newman v. Alabama, 683 F.2d 1312, 1319 (11th Cir. 1982); see also Ruiz v. Estelle, 679 F.2d 1115, 1145 (5th Cir. 1982) (“We should, therefore, fashion ‘the least intrusive remedy that will still be effective.’ ”), amended on rehg., in part, on other grounds, 688 F.2d 266 (5th Cir. 1982) . 59. See, e.g., Milliken v. Bradley, 433 U.S. 267, 280 (1976) (“Once invoked, ‘the scope of a district court’s equitable powers to remedy past wrongs is broad’ . . . .”). 60. Prison Litigation Reform Act, Pub. L. No. 104-134, 110 Stat. 1321, §§ 801-810 (Apr. 26, 1996) (codified in pertinent part at 18 U.S.C. § 3626 (1997 Supp.) (“PLRA”)). For discussion of the PLRA’s provisions, see infra notes 170-177 and accompanying text. May 1999] Judicial Policymaking 2011 all this changed when a court handed down its order as part of a settled outcome. By settling, the parties preempted the need for the judge to make liability findings, issue remedial orders, or both. Settlements do not require formal proof or legal argument. When parties settle remedial issues in a civil rights injunctive suit they usually submit their agreement to the court in the form of a proposed court order; on adoption by the court the settlement becomes enforceable by and against the parties to the same extent as contested orders based on pleadings, briefs, and record evidence. Such consent decrees have frequently incorporated terms that a judge could not lawfully include in a contested order.61 In addition, consent decrees are more permanent than contested orders.62 And while judges were not supposed to (and did not quite) rubber stamp consent decrees,63 the pre-1996 rules governing acceptance of a prison or jail conditions consent decree by a trial judge were far from strict. A decree needed only “spring from and serve to resolve a dispute within the court’s subject-matter jurisdiction[,] . . . ‘com[e] within the general scope of the case made by the pleadings,’ and . . . further the objectives of the law upon which the complaint was based.”64 Finally, a judge must examine any settlement of a class action litigation to guard against the possibility that plaintiffs’ counsel have sold out some or all of their clients’ interests too cheap.65 The ordinary litigation incentives favoring settlement operate strongly for parties and judges in structural reform cases. Settle61. See Local Number 93, Intl. Assn. of Firefighters v. City of Cleveland, 478 U.S. 501, 525 (1986) (“[A] federal court is not necessarily barred from entering a consent decree merely because the decree provides broader relief than the court could have awarded after a trial.”). 62. Courts must, on a motion by a party, alter a contested injunction to respond to subsequent changes in law. See Pennsylvania v. Wheeling & Belmont Bridge Co., 59 U.S. (18 How.) 421, 431-32 (1855). By contrast, consent decrees generally are amended only where there has been a significant and unanticipated change in fact or law; even when such a change occurs, a consent decree should not be rewritten to “conform[ ] to the constitutional floor.” Rufo v. Inmates of the Suffolk County Jail, 502 U.S. 367, 389-91 (1992). Note, however, that in institutional reform cases, the standard for termination of contested and consent judgments is generally the same — compliance. See Board of Educ. of Oklahoma City v. Dowell, 498 U.S. 237, 246-49 (1991). But see Prison Litigation Reform Act, 18 U.S.C. § 3626(b)(1), (b)(3) (changing this rule in prison, jail, and juvenile corrections cases by requiring court termination of any litigated or consent judgment, on motion by a party, if the relief is no longer necessary to remedy a current and ongoing constitutional violation). 63. See United States v. Miami, 664 F.2d 435, 440-42 (5th Cir. 1981). 64. Firefighters, 478 U.S. at 525 (citation omitted). 65. That is, the judge must find that the action satisfies the requirements for class disposition, see FED. R. CIV. P. 23, including that there are no intra-class conflicts, see Amchem Prods. v. Windsor, 521 U.S. 591, 621-28 (1997), and must declare the settlement “fair, adequate, and reasonable.” Officers for Justice v. Civil Service Commn., 688 F.2d 615, 625 (9th Cir. 1982) (describing this formulation as the “universally applied standard”). This second standard is a judicial gloss on the requirement set out in FED. R. CIV. P. 23(e) that “[a] class action shall not be dismissed or compromised without the approval of the court.” 2012 Michigan Law Review [Vol. 97:1994 ment saves the enormous expense and uncertainty of trial and appeal, and it gives the parties augmented control over the specifics of a remedy. More speculatively, defendants who agree to a decree may transform themselves in the eyes of the public, and even in their own eyes, from “lawbreakers to law implementers.”66 And there are also more situation-specific incentives. Plaintiffs or their counsel, and judges, may push especially hard for settlement if they believe that necessary institutional change requires the cooperation of the defendants, which is more easily obtained by consent than by judicial fiat.67 Another frequently remarked dynamic favoring settlement in institutional reform cases, duly noted by Feeley and Rubin, is the high level of cooperation by defendants (pp. 59-62, 116, 373-74, 378). The explanation seems clear: defendants, who are government officials operating under fiscal and political constraints, frequently win by losing. The result of a consent decree can be more resources and freedom from entrenched restrictions on changes in policy and practice. “The court is making me do it” trumps many ordinary political considerations.68 In the particular context of prison litigation, defendants were often themselves interested in the professionalization, and concurrent bureaucratization, of the prisons under their supervision.69 Finally, with a consent decree, defendant officials can even gain a power, unavailable through the ordinary political process, to bind their successors.70 For all these reasons, settlements of various kinds do indeed seem to be the primary source of judgments in prison and jail cases;71 the litiga66. Ross Sandler & David Schoenbrod, Government by Decree: The High Cost of Letting Judges Make Policy, CITY J., Summer 1994, at 54, 58. 67. See, e.g., Maimon Schwarzschild, Public Law by Private Bargain: Title VII Consent Decrees and the Fairness of Negotiated Institutional Reform, 1984 DUKE L.J. 887, 898-901 (discussing incentives for settlement in institutional reform cases). Schwarzschild summarizes with reference to employment discrimination decrees: “[A] judge quite properly encourages a settlement when it fosters a conciliatory atmosphere in which the employer is more likely to comply with the letter and the spirit of the decree. Less creditably, perhaps, some judges may welcome . . . consent decrees as an opportunity to avoid grappling with the policy dilemmas and moral ambiguities lurking about [the issue being litigated].” Id. at 901 (footnote omitted). 68. See, e.g., Mark Kellar, Responsible Jail Programming, AM. JAILS, Jan.-Feb. 1999, at 78, 79 (“To be sure, we used ‘court orders’ and ‘consent decrees’ for leverage. We ranted and raved for decades about getting federal judges ‘out of our business’; but we secretly smiled as we requested greater and greater budgets to build facilities, hire staff, and upgrade equipment. We ‘cussed’ the federal courts all the way to the bank.”). 69. See Elizabeth Alexander, The New Prison Administrators and the Court: New Directions in Prison Law, 56 TEXAS L. REV. 963, 967-71 (1978); see also p. 378 (“[A]s we have seen, the more progressive administrators either viewed the courts as allies or were eager to conclude a truce with them that traded their discretion for increased resources and a tighter organizational structure.”). 70. See Michael W. McConnell, Why Hold Elections? Using Consent Decrees to Insulate Policies from Political Change, 1987 U. CHI. LEGAL F. 295 (criticizing this result). 71. Observers and participants all seem to agree that settlements are very prevalent. See, e.g., Lloyd C. Anderson, Implementation of Consent Decrees in Structural Reform Litigation, May 1999] Judicial Policymaking 2013 tion has frequently been, to use Marc Galanter’s coinage, “litigotiation” — “the strategic pursuit of a settlement through mobilizing the court process.”72 Of course, that the parties are able to reach agreement on all or part of the questions of liability or initial remedy does not mean that the case is over, or that contested litigation ceases. Enforcement is where the action is, in corrections as in other complex mandatory injunction cases,73 and a case can settle easily but be hard-fought post-judgment.74 Even a judge who was not involved during settlement negotiations may become a central actor during enforcement. Moreover, judges frequently do play a substantive role in encouraging and crafting complex settlements of all kinds, including consent decrees, both actively and indirectly through the parties’ surmises or knowledge about a judge’s substantive inclinations.75 But even so, no consent judgment is the pure result of judicial decisionmaking. Decrees develop out of the complex interplay of the judges’ promotion of settlement76 and the parties’ expecta1986 U. ILL. L. REV. 725, 725; Chayes, supra note 10, at 1298-1302; William C. Collins, Medical Authorities Identify Consent Decree Syndrome, May Be Endemic Among Correctional Administrators, CORRECTIONAL L. RPTR. Apr. 1992, at 83; see also Glen R. Jeffes, The Thirteen Commandments of Negotiating and Living with Consent Decrees, AM. JAILS, May-June 1990, at 38. Of course, this does not make institutional reform cases any different from other types of litigation: depending on how you count, anywhere from 67% to 95% of civil cases settle, see Marc Galanter & Mia Cahill, “Most Cases Settle”: Judicial Promotion and Regulation of Settlements, 46 STAN. L. REV. 1339, 1339-40 (1994) [hereinafter Galanter & Cahill, “Most Cases Settle”]. And guilty pleas constitute 92% of all criminal convictions in federal courts (78% of all criminal dispositions, including dismissals and acquittals), and 92% of all criminal convictions in state courts (information on state dismissals and acquittals is not easily available, but the best available statistic is that a 1987 sampling of 23 jurisdictions revealed a mean rate of guilty pleas as 68% of all criminal dispositions), see BUREAU OF JUSTICE STATISTICS, U.S. DEPT. OF JUSTICE, SOURCEBOOK OF CRIMINAL JUSTICE STATISTICS 1997, at 392 tbl.5.16 (federal), 422 tbl.5.47 (state); BUREAU OF JUSTICE STATISTICS, U.S. DEPT. OF JUSTICE, THE PROSECUTION OF FELONY ARRESTS, 1987, at 6 tbl.3 (1990). 72. Marc Galanter, Worlds of Deals: Using Negotiation to Teach About Legal Process, 34 J. LEGAL EDUC. 268, 268 (1984); see also Cavanagh & Sarat, Beyond Judicial Competence, supra note 12, at 405 (“Extended impact litigation does not displace negotiation and compromise but is frequently an essential precondition to it.”). 73. See, e.g., Sturm, Legacy and Future, supra note 13, at 724-28; Karla Grossenbacher, Note, Implementing Structural Injunctions: Getting a Remedy When Local Officials Resist, 80 GEO. L.J. 2227, 2235 (1992); Anderson, supra note 71, at 737; Jonathan M. Smith, Enforcing Corrections-Related Court Orders in the District of Columbia, 2 D.C. L. REV. 237 (1994). 74. See, e.g., United States v. Michigan, 62 F.3d 1418 (6th Cir. 1995) (unpublished table op., available in 1995 WL 469430) (describing post-settlement litigation). 75. See, e.g., Galanter & Cahill, “Most Cases Settle,” supra note 71, at 1340 (1994) (“In the two-thirds of cases that do settle without a definitive judicial ruling, judges are by no means absent. Rather, they are a ghostly but influential presence, through their rulings in adjudicated cases and their anticipated response to the case at hand.”); Peter H. Schuck, The Role of Judges in Settling Complex Cases: The Agent Orange Example, 53 U. CHI. L. REV. 337 (1986). 76. See Cavanagh & Sarat, Beyond Judicial Competence, supra note 12, at 373, 385-86 (describing “providing a framework within which parties negotiate and bargain” as “perhaps [courts’] most important function,” and emphasizing “the possibly unique ability of courts to promote informal social ordering”). 2014 Michigan Law Review [Vol. 97:1994 tions as to the outcome of litigation77 and varying stakes and information.78 A “microanalysis” of cases that looks chiefly at judicial doctrine creation fails to notice the significant explanatory value of these crucial factors. Feeley and Rubin do comment on the pre-judgment settlement of one of their five case studies, noting that the sheriff of the Alameda County jail “helped write specifications into the consent decree that went well beyond anything the court would have ordered” (p. 363). But none of their other featured cases settled without a judicial determination of liability, which probably makes them quite unusual. And, correspondingly, the authors’ theoretical account simply does not incorporate settlement as a concern. It seems unlikely that this omission is mere oversight — settlement is too obvious and remarked a feature of institutional reform litigation. Perhaps it was their very desire to legitimate judges as policymakers that led Feeley and Rubin to slight aspects of the prison cases in which judges were so far from being the lone policy mouthpieces. If the “policy” embodied in a court decree is the result of a complex interaction between the parties and other political players, the judge, and the rules of litigation, it does not look like the kind of independent, principled, reasoned elaboration that the strongest defenders of judicial action highlight.79 (On the other hand, perhaps its legitimacy as policy is enhanced, because it was not imposed on the polity but came out of the negotiations and consent of democratically accountable officials.) Whatever the source of the theoretical lacuna, the prison cases simply do not support Feeley and Rubin’s single-minded consideration of how judges act when they decide cases and originate remedies; understanding the cases calls for analysis of the other ways court judgments and outcomes are derived, along with an assessment of the differing contours and relative importance of contested judgments and settlements. Plaintiffs’ Counsel. Similarly, Feeley and Rubin do describe at least some of the parties and their lawyers, in the case studies and in 77. Bargaining occurs, that is, “in the shadow of the law.” Robert H. Mnookin & Lewis Kornhauser, Bargaining in the Shadow of the Law: The Case of Divorce, 88 YALE L.J. 950 (1979). Procedural as well as substantive law can cast such a shadow. See Marc S. Galanter, Federal Rules and the Quality of Settlements: A Comment on Rosenberg’s The Federal Rules of Civil Procedure in Action, 137 U. PA. L. REV. 2231, 2234 (“Substantive and procedural rules, and the practices of courts and lawyers, confer bargaining endowments upon the parties in settlement negotiations.”); Martha Minow, Politics and Procedure, in THE POLITICS OF LAW 79, 93 (David Kairys ed., 3d ed. 1998) (“The relative proportion of fully litigated trials, court-sponsored mediations, and out-of-court settlements reflect rules about attorneys’ fees, filing dates, discovery, . . . jurisdiction, and choice of law.”). 78. See George L. Priest & Benjamin Klein, The Selection of Disputes for Litigation, 13 J. LEGAL STUD. 1 (1984); Steven Shavell, Any Frequency of Plaintiff Victory at Trial is Possible, 25 J. LEGAL STUD. 493 (1996) (focusing on asymmetric information). 79. See, e.g., Fiss, The Forms of Justice, supra note 10, at 13-14; see also Owen M. Fiss, Against Settlement, 93 YALE L.J. 1073, 1077 (1984). May 1999] Judicial Policymaking 2015 their final chapter, but they omit these nonjudicial actors from their judge-centered theoretical framework. This omission obviously compounds the authors’ failure to take account of the prevalence of settlement, because the litigants necessarily have an overwhelming effect on the shape of settled outcomes. But non-judicial parties are important to understanding fully litigated cases, as well. I focus here on plaintiffs’ lawyers, a particularly crucial set of players because our system of procedure makes the plaintiff, by counsel, “master of the suit.”80 In particular, the rules of litigation largely confine judicial response to the record developed and the arguments presented by the parties; for a plaintiff’s judgment, there must be a connection between the order a court issues, and the claims, evidence, and requested relief plaintiffs’ counsel submits. Thus, unlike efforts to urge new executive or legislative policy, litigation gives those seeking change a formal and unique ability to shape the contest. In addition, class-action litigation creates a particularly distant relationship between the real parties in interest (here, the prisoners) and their champions (class counsel).81 Accordingly, the identity, priorities, litigating strategies, and resources of plaintiffs’ counsel have been of great importance to the shape and success of litigated prison reform.82 The district judges who eventually oversaw the litigated reform of prisons could decide the cases, if those cases did not settle. But judges could not easily (or appropriately) put together the evidentiary records needed to survive an appeal, and they did not, in fact, themselves invent the legal theories underlying their decisions. Rather, they generally acted by 80. FLEMING JAMES, JR., GEOFFREY C. HAZARD, JR., & JOHN LEUBSDORF, CIVIL PROCE542 (4th ed. 1992). DURE 81. Nonlawyers may represent themselves in individual lawsuits, but statutory bans on the practice of law by nonlawyers mean that only lawyers may represent other people, so only lawyers may serve as class counsel in a class action. Once appointed as class counsel, a lawyer is ethically obligated to serve the interests of the entire class, not any individual member in it; given the difficulty of ascertaining what a class of people as diverse as inmates actually wants, this means that even a lawyer who seeks direction from the class (as many lawyers do not) is somewhat on his or her own. See Deborah L. Rhode, Class Conflicts in Class Actions, 34 STAN. L. REV. 1183 (1982). 82. See Alvin J. Bronstein, Prisoners and Their Endangered Rights, PRISON J., Mar. 1985, at 3, 11 (“[I]t is likely that cases will succeed or fail not on the basis of how unconstitutional the conditions are, but on the basis of how resourceful the lawyers and experts are.”); JOEL F. HANDLER, SOCIAL MOVEMENTS AND THE LEGAL SYSTEM: A THEORY OF LAW REFORM AND SOCIAL CHANGE 35 tbl.1.6 (1978) [hereinafter HANDLER, SOCIAL MOVEMENTS] (identifying numerous characteristics of law reform groups that bear on the probability of success in their reform efforts, including their size, funding, institutional affiliation, technical expertise, and political resources); WAYNE N. WELSH, COUNTIES IN COURT: JAIL OVERCROWDING AND COURT-ORDERED REFORM 40-41, 49-53, 62-63, 79 (1995) (analyzing the difference in the profile of jail reform litigation conducted by different types of plaintiffs’ counsel); see also Sturm, Lawyers at the Prison Gates, supra note 13 (examining characteristics, strategies, and interests of various groups of plaintiffs’ counsel in correctional cases, including the ACLU National Prison Project, the Youth Law Center, legal services organizations, law firms, and law school clinics). 2016 Michigan Law Review [Vol. 97:1994 following a path proposed by plaintiffs’ counsel and by building on the foundation laid at trial. It is true, as Feeley and Rubin set out (pp. 59, 61, 81, 100), that reform-minded judges sometimes initiated the early prison cases, usually by turning the individual petitions of one or more inmates into the basis of a class action. But even in such circumstances, the judges began by appointing appropriate lawyers (frequently handpicking them for their expertise), and then quickly resumed the traditional stance of arbiter rather than originator. Federal District Judge William Wayne Justice started the Ruiz litigation in Texas by consolidating several inmate petitions, including at least one by a well known inmate “writ writer.” He explained fifteen years later that the purpose of his next step — choosing a skilled and aggressive plaintiffs’ lawyer for the inmates — was to put the job of developing the inmates’ case back in counsel’s hands, to “accord with the goals and aspirations of our adversarial system of justice.”83 To assess the contribution of plaintiffs’ lawyers in the early prison cases, one must first understand their background as participants in the civil rights movement. Feeley and Rubin state elliptically that “the basic relationship between the civil rights movement and prison reform is causal” (p. 159), but they do not, in fact, discuss the civil rights movement. Rather, they spend a page or two describing a series of judicial actions — doctrinal developments that “were part of the effort to secure decent treatment for black citizens in general” (p. 159), and Brown v. Board of Education, to which they ascribe a metaphysical impact: [T]he case that was most important for placing prison reform on the judicial agenda was Brown itself, whose moral message was the fountainhead of our postwar constitutional jurisprudence. . . . [Brown’s] real meaning was that America would finally fulfill the broken promise of its founding, that the full panoply of rights would be extended to everyone, including the people it had formerly enslaved. [p. 160] The authors do not find Brown’s specific antisegregation holding terribly important in the prison context (p. 160), but focus instead on its transformative moral message, which “produce[d] a sort of legal epiphany for federal judges” (p. 160). What Feeley and Rubin leave out are the complex and much less judge-focused ways in which the civil rights movement contributed to the commencement and early history of litigated prison reform. The civil rights movement as a whole both depended on and spurred the project of litigation as an engine of social change,84 and prison litigation was a small piece of this larger project. 83. William Wayne Justice, The Origins of Ruiz v. Estelle, 43 STAN. L. REV. 1, 3 (1990). 84. See, e.g., MARK V. TUSHNET, MAKING CIVIL RIGHTS LAW: THURGOOD MARSHALL AND THE SUPREME COURT, 1936-1961 (1994); MARK V. TUSHNET, THE NAACP’S LEGAL STRATEGY AGAINST SEGREGATED EDUCATION, 1925-1950 (1987). May 1999] Judicial Policymaking 2017 Civil rights lawyers got involved in the prison cases in a variety of ways. A number of the lawyers who represented plaintiffs in the core cases of the movement — school desegregation, voting rights, criminal defense of civil rights protestors, and the like — started doing prison litigation after, in a sense, “follow[ing] their clients into jail.”85 The NAACP Legal Defense and Education Fund (the organization, often called “LDF” or the “Inc. Fund,” that represented African Americans in many of the well known litigation efforts of the civil rights era)86 was the first national group to become heavily involved in attempts to reform prisons through litigation. In one early foray into prison litigation, LDF lawyers brought damage actions over the treatment of civil rights protesters in Mississippi’s notorious Parchman farm prison. The same lawyers subsequently brought Gates v. Collier,87 a broad-gauged reform litigation challenging segregation and conditions at Parchman.88 More generally, LDF’s initiation of prison litigation was part of its major effort in the mid-1960s to expand the organization’s docket beyond explicitly racial claims to cases relating to poverty, crime, and related issues.89 Starting in the late 1960s, LDF frequently asked its cooperating attorneys to handle prison cases90 and quickly became a significant force, coordinating the litigation of new legal theories and the development of evidence to support them. By 1975, it had a docket of more than fifty jail and prison cases. Around 1977, 85. David J. Rothman, Decarcerating Prisoners and Patients, 1 CIV. LIBERTIES REV. 8, 14 (1973). My interpretation here accords more generally with Rothman’s; he writes that “[t]he chronicle of many prisoners’ rights lawyers appears in their movement from civil rights litigation to contesting prison segregation to arguing the constitutionality of prison practices.” Id.; see also Francis B. Stevens & John L. Maxey, II, Representing the Unrepresented: A Decennial Report on Public-Interest Litigation in Mississippi, 44 MISS. L.J. 333, 333-34 (1973) (attributing the onset of litigated reform of Mississippi institutions to the new availability of public interest representation). 86. See GREENBERG, CRUSADERS, supra note 33. 87. 501 F.2d 1291 (5th Cir. 1974). 88. See DAVID M. OSHINSKY, WORSE THAN SLAVERY: PARCHMAN FARM ORDEAL OF JIM CROW JUSTICE 238-41 (1996). AND THE 89. See GREENBERG, CRUSADERS, supra note 33, at 430-60 (discussing broad set of LDF initiatives funded by a Ford Foundation grant). 90. “Cooperation” was an arrangement where lawyers would agree to handle civil rights cases with some financial support, direction, and advice from LDF. See Robert L. Rabin, Lawyers for Social Change: Perspectives on Public Interest Law, 28 STAN. L. REV. 207, 21618 (1976). For example, Sanford Bishop (now a member of Congress) was a young lawyer from Georgia who had received an LDF-initiated fellowship for African Americans at historically white law schools in the South. After working for LDF for a year in New York, he went back to Georgia, where he worked on LDF’s cases. It was at LDF’s request that he took on the Georgia prison litigation, Guthrie v. Evans, Civ. A. No. 3068 (S.D. Ga. filed Sept. 1972), contempt motion settled, 93 F.R.D. 390, 391-93 (S.D. Ga. 1981) (describing procedural history). See BRADLEY STEWART CHILTON, PRISONS UNDER THE GAVEL: THE FEDERAL COURT TAKEOVER OF GEORGIA PRISONS 22-23 (1991); GREENBERG, CRUSADERS, supra note 33, at 458, 601. 2018 Michigan Law Review [Vol. 97:1994 however, when the staff lawyers responsible for the cases left LDF, it essentially ended its prison and jail litigation efforts.91 Joining LDF in the early days of litigated prison reform was the American Civil Liberties Union.92 The ACLU was plaintiffs’ counsel for at least two of the major prison desegregation lawsuits in the late 1960s,93 but its lawyers did not start doing broader prison conditions suits for several years. By 1971, the ACLU supported two small prison litigation projects.94 After the Attica riot, that year,95 the public — and, crucially, philanthropic foundations — showed increased interest in prison conditions.96 In 1972, the foundations funding the ACLU’s two small projects proposed merging them; the ACLU’s National Prison Project was the result.97 The Prison Project immediately became a force in the national litigated reform movement, serving as counsel in a state-wide prison case in Alabama, and then in cases in Rhode Island, Tennessee, and New Mexico. It has remained the leading national inmates’ litigator.98 91. Telephone Interviews with former LDF attorneys Stanley A. Bass (Jan. 12, 1999) [hereinafter Bass interview] and William Bennett Turner (Jan. 18, 1999) [hereinafter Turner interview]. 92. On the importance of the ACLU to the origins of contemporary public interest law, see Rabin, supra note 90, at 209-14. 93. See Washington v. Lee, 263 F. Supp. 327 (M.D. Ala. 1966), affd. per curiam, 390 U.S. 333 (1968); Wilson v. Kelley, 294 F. Supp. 1005 (N.D. Ga. 1968) (both listing as plaintiffs’ counsel, inter alia, ACLU Legal Director Melvin Wulf). 94. One, in Buffalo, New York, was founded by law professor Herman Schwartz, and funded by the Playboy Foundation; it dealt primarily with issues at the Attica state prison both before and after the riot there in September 1971. The other was founded by civil rights lawyer Philip Hirschkop, in Virginia, and funded by the Stern Family Fund and the Field Foundation. It focused on the Virginia penal system. Telephone Interview with former ACLU National Prison Project Executive Director Alvin J. Bronstein (Dec. 21, 1998) [hereinafter Bronstein interview]. 95. In September 1971, after an uprising by inmates at New York’s Attica prison, state officials regained control in an armed attack that killed 29 inmates and 10 of the (previously unharmed) officer hostages. See generally NEW YORK STATE SPECIAL COMMISSION ON ATTICA, ATTICA: THE OFFICIAL REPORT OF THE NEW YORK STATE SPECIAL COMMISSION ON ATTICA (1972); TOM WICKER, A TIME TO DIE: THE ATTICA PRISON REVOLT (1975). Attica captured the imagination of the public (Wicker’s book was on the New York Times Best Seller List for 40 weeks), catalyzed a season of prison disturbances and years of foundation support for prison reform work, and inspired a generation of prison lawyers. It remains the most famous of American prison disturbances. See, e.g., id.; Alvin J. Bronstein, Reform Without Change: The Future of Prisoners’ Rights, CIV. LIBERTIES REV., Sept.-Oct. 1977, at 27, 32. 96. The National Prison Project was funded at its start by the Field Foundation, the Stern Family Fund, and the Playboy Foundation; it soon gained more ample support from the Edna McConnell Clark Foundation, and was, in fact, Clark’s largest grantee for a time. Bronstein interview, supra note 94. On the importance of foundations in building a support structure for law reform litigation, see, for example, CHARLES R. EPP, THE RIGHTS REVOLUTION: LAWYERS, ACTIVISTS, AND SUPREME COURTS IN COMPARATIVE PERSPECTIVE 58-59 (1998) [hereinafter EPP, THE RIGHTS REVOLUTION]. 97. Bronstein interview, supra note 94. 98. In 1995, the National Prison Project was involved in at least one case in 30 of the 43 states or territories whose prisons were subject to court orders to improve conditions. See ACLU National Prison Project, Status Report: State Prisons and the Courts (1995), in 3 PRISONERS AND THE LAW app. B at 109-42 (Ira P. Robbins ed., 1998). May 1999] Judicial Policymaking 2019 Finally, other “public interest” lawyers99 handled prison cases as well—especially lawyers from the large array of new legal services offices that received federal funding starting in 1965.100 Although the Reagan budget cuts of 1981 reduced their involvement,101 legal services offices were important players in the first years of the litigated reform movement; they were regular, if not leading, litigators of prison cases, and the primary litigators of jail cases around the country.102 In 1996, Congress basically ended the role of legal services programs in prison reform by prohibiting use of federal legal services funding for class action litigation and the representation of prisoners.103 99. On the profile of the “public interest law industry” in the 1970s, see Joel F. Handler, Betsy Ginsberg, & Arthur Snow, The Public Interest Law Industry, in PUBLIC INTEREST LAW: AN ECONOMIC AND INSTITUTIONAL ANALYSIS 42 (1978). See also id. at 58 tbl.4.8 (of 72 public interest law firms sampled, 14 did prison cases, devoting an average of 30% of their time to representing inmates). 100. See EARL JOHNSON, JR., JUSTICE AND REFORM: THE FORMATIVE YEARS OF THE OEO LEGAL SERVICES PROGRAM 39-70 (1974) (describing birth process of War on Poverty’s Office of Economic Opportunity (OEO) legal program, with appointment of first director on September 24, 1965). Statutory authority came a little later. See Economic Opportunity Amendments of 1967, Pub. L. No. 90-222, 81 Stat. 672; Economic Opportunity Amendments of 1966, Pub. L. No. 89-794, 80 Stat. 1415. Prior to the OEO program, just a few hundred full-time lawyers performed legal aid work. By 1972, OEO had added over 2600 lawyers, who worked in 850 offices in over 200 communities. See JOHNSON, supra, at 188. OEO was replaced in 1974 by the Legal Services Corporation, which continues to exist today. See Legal Services Corporation Act, Pub. L. 93–355, 88 Stat. 378 (1974) (codified as amended at 42 U.S.C. § 2996). For descriptions and histories of legal assistance for poor people in America, see, for example, JOHN A. DOOLEY & ALAN W. HOUSEMAN, LEGAL SERVICES HISTORY (1984); JACK KATZ, POOR PEOPLE’S LAWYERS IN TRANSITION (1982); Gary Bellow, Legal Aid in the United States, 14 CLEARINGHOUSE REV. 337 (1980); Warren E. George, Development of the Legal Services Corporation, 61 CORNELL L. REV. 681 (1976); Carrie Menkel-Meadow, Legal Aid in the United States: The Professionalization and Politicization of Legal Services in the 1980’s, 22 OSGOODE HALL L.J. 29 (1984); . 101. See Sturm, Lawyers at the Prison Gates, supra note 13, at 4 n.5; Bronstein, supra note 82, at 11. 102. See Philip B. Taft, Jr., Jail Litigation: Winning in Court is Only Half the Battle, CORMAG. June 1983, at 23, 23 (“Because jails are locally controlled, most of the battles have been waged piecemeal by local legal service attorneys.”). For early examples of legal services jail and prison litigation, see Sinclair v. Henderson, 331 F. Supp. 1123 (E.D. La. 1971) (first successful reported prison litigation in Louisiana, challenging conditions on death row, litigated by legal aid society lawyers); Pounds v. Theard, 230 So. 2d 861 (La. Ct. App. 4th Cir. 1970) (desegregation of New Orleans City jail; listed attorneys worked for New Orleans Legal Assistance); LEO CARROLL, LAWFUL ORDER: A CASE STUDY OF CORRECTIONAL CRISIS AND REFORM 50-51, 66 (1998) (describing Rhode Island Legal Services representation of inmates in early systemic litigation). Between 1970 and 1990, the National Clearinghouse for Legal Services’ Clearinghouse Review reported on 327 jail and prison conditions cases, almost all conducted by legal services organizations. See also Inmates of Suffolk County Jail v. Eisenstadt, 360 F. Supp. 676, 684 (D. Mass. 1973) (“During the past few years, due largely to the courage of young poverty-program lawyers, the soul-chilling inhumanity of conditions in American prisons has been thrust upon the judicial conscience.”). For a more recent, but pre-1996, assessment of legal services involvement in corrections litigation, see Sturm, Lawyers at the Prison Gates, supra note 13, at 53-69. RECTIONS 103. See Omnibus Consolidated Rescissions & Appropriations Act of 1996, Pub. L. No. 104-134 § 504(15); 110 Stat. 1321, 1321-55; see also 45 C.F.R. §§ 1632.1-.5. 2020 Michigan Law Review [Vol. 97:1994 Whatever their organizational home, the repeat plaintiffs’-side prison litigators shared information and strategy, both informally and formally. For example, in 1972, the federal Office of Economic Opportunity and the Ford Foundation gave a grant to the American Bar Association to start up the National Resource Center on Correctional Law and Legal Services, a “backup center,” which could provide legal services lawyers with advice and model pleadings.104 Between 1970 and 1990, the National Clearinghouse for Legal Services also published relevant articles and descriptions of jail and prison cases in its Clearinghouse Review and made pleadings available to legal services lawyers. For several years in the early 1970s, the American Bar Association’s Young Lawyers Section and its Commission on Correctional Facilities and Services published the Prison Law Reporter, which reported on judicial decisions, reprinted plaintiff’s briefs and other pleadings, and published topical bibliographies and news of various organizations’ activities. And from 1978 to 1981, the National Prison Project, among other groups, supported a similar publication called Prison Law Monitor. Notwithstanding all this communication, the varying resources, goals, and strategies of these groups also shaped prison litigation’s history, affecting what claims the groups made, what violations were found, and the eventual remedies chosen.105 For example, LDF pioneered the argument that conditions in a given facility or part of a facility violated the Eighth Amendment.106 The ACLU, by contrast, maybe because of its civil liberties background, focused on due process issues in its early prison cases.107 Once the ACLU broadened its approach, it began to emphasize overcrowding, per104. See Richard G. Singer, National Resource Center on Correctional Law & Legal Services: Resource Center for Correctional Law Begins Operation, 6 CLEARINGHOUSE REV. 253 (1972). 105. See Stephen L. Wasby, Civil Rights Litigation by Organizations: Constraints and Choices, 68 JUDICATURE 337 (1985) (describing factors that affect the planning and execution of planned litigation compaigns by groups). 106. An LDF attorney, William Bennett Turner, conducted what seems to be only the second trial in the country that asserted that prison conditions amounted to cruel and unusual punishment. The case, which concerned conditions in the “isolation unit” of New York’s Dannemora prison, was Wright v. McMann, 257 F. Supp. 739 (N.D.N.Y. 1966), revd., 387 F.2d 519, 521-26 (2d Cir. 1967), on remand, 321 F. Supp. 127 (N.D.N.Y. 1970) (trial held Oct. 15, 1968), affd. in part and revd. in part, 460 F.2d 126 (2d Cir. 1972). The first such trial had been conducted the year before, in Jordan v. Fitzharris, 257 F. Supp. 674 (N.D. Cal. 1966); inmate Robert Charles Jordan, Jr’s lawyer, Charles Cohler, had graduated the year before from Harvard Law School and was appointed by the court. Cohler explains that the theory on which he tried and won the case had been pleaded by his client, a prolific and accomplished writ-writer. Telephone Interview with Charles B. Cohler (March 16, 1999). 107. See Alvin J. Bronstein, Representing The Powerless: Lawyers Can Make a Difference, 49 ME. L. REV. 1, 13 (1997) (“In our early years, we were doing a lot of narrow cases — you know, the lawyer’s hangup — procedural due process, which basically doesn’t change anything, at least in the prison context. It gets you the fair procedures and then the prison officials make the same old unfair decisions. Their decisions then were insulated from court review because you had all these due process procedures.”). Landman v. Royster, 333 F. May 1999] Judicial Policymaking 2021 haps because of an interest in encouraging decarceration techniques such as alternative sentencing and early release,108 and perhaps because the threat of overcrowding remedies unpalatable to prison officials, such as population caps, induced defendants to make settlement concessions in other areas.109 Cases brought by legal services offices tended to be more limited, perhaps because prison litigation was not their primary purpose.110 And because they relied on government rather than foundation funding, legal services offices were more subject to political pressure.111 To illustrate how discussion of the prisoners’ lawyers might have deepened Feeley and Rubin’s account, consider the Arkansas litigation, to which the authors devote a good deal of attention. As they note, federal District Judge J. Smith Henley was a particularly active presence in several of the Arkansas cases (pp. 56–73), and the defendants were especially willing to admit, on the record, to their own problems (pp. 60–61). Nonetheless, the series of cases about conditions in Arkansas’ prisons stayed very limited for their first five years, as attorney after attorney was appointed to represent the inmates. It was not until the fifth case that Judge Henley appointed a civil rights lawyer as plaintiffs’ counsel, when he chose Philip Kaplan, an LDF cooperating attorney who had litigated numerous school desegregation cases in Arkansas.112 As Feeley and Rubin describe, Kaplan and his appointed co-counsel Jack Holt113 broadSupp. 621 (E.D. Va. 1971), the landmark prison discipline due process case, was litigated by Philip J. Hirschkop of the ACLU. 108. See pp. 375-76 (describing one prisoners’ advocate’s “pincer strategy” of driving up the costs of prisons while promoting nonincarceration alternatives). 109. See Jeff Bleich, The Politics of Prison Crowding, 77 CAL. L. REV. 1125, 1170-73 (1989) (discussing complicated politics of the ACLU National Prison Project’s overcrowding claims). 110. See, e.g., Sturm, Lawyers at the Prison Gates, supra note 13, at 53–69; Telephone Interview with former U.S. Civil Rights Division attorney Stephen A. Whinston (approx. Jan. 20, 1999) [hereinafter Whinston interview]. 111. See Mark Kessler, Legal Mobilization for Social Reform: Power and the Politics of Agenda Setting, 24 L. & SOC. REV. 121, 132-36 (1990) (describing mechanisms by which legal services lawyers are discouraged from doing reform litigation, including jail and prison cases); supra notes 101-103 and accompanying text. 112. Long-time LDF general counsel Jack Greenberg recounts that Kaplan received $200 and costs from LDF for any civil rights case he handled. See GREENBERG, CRUSADERS, supra note 33, at 457. For some of Kaplan’s other civil rights litigation, see Chase v. Twist, 323 F. Supp. 749 (E.D. Ark. 1970); Arkansas Educ. Assn. v. Board of Educ., 446 F.2d 763 (8th Cir. 1971). 113. Holt was a “local lawyer with extensive experience as a criminal prosecutor and inpeccable conservative credentials.” P. 61. When he was elected (not appointed, as Feeley and Rubin state, p. 62) Chief Justice of the Arkansas Supreme Court, in 1984, he was the third member of his extended family to serve on that court. His father had been Arkansas Attorney General for three terms. See Holt, Top Judge in State Since ‘84, Retiring Sept. 1, ARK. DEM.-GAZETTE, May 13, 1995, at 1A. But while Kaplan was the first plaintiffs’ counsel in the Arkansas prison cases who was a civil rights lawyer, Holt was not the first laywer appointed to represent the inmates who had a gold-plated membership in the legal establishment. In the second of the Arkansas cases, Jackson v. Bishop, 268 F. Supp. 804 (E.D. Ark. 2022 Michigan Law Review [Vol. 97:1994 ened the Arkansas litigation to include a Thirteenth Amendment attack on the prison systems’ use of forced labor, an Eighth Amendment attack on its general conditions and practices, and an Equal Protection Clause attack on its system of race segregation (p. 62). Where the earlier stages of the litigation had been short-lived and piecemeal, this fifth stage was the occasion for more intrusive relief of far longer duration. What Feeley and Rubin fail to note is that the inspiration and support for this comprehensive approach came from LDF staff: according to a history of LDF, when Kaplan was appointed to the Arkansas litigation, he called William Bennett Turner, one of LDF’s prison litigators (who later represented the plaintiffs in the Texas Ruiz litigation) and the two drafted the amended complaint together.114 Thus Holt did not become the landmark case Feeley and Rubin describe until it was taken on by a civil rights lawyer with ties to the evolving national prisoners’ rights bar. Feeley and Rubin’s failure to incorporate such circumstances undermines the usefulness of their theory in understanding these cases. Another important plaintiffs’-side player in prison litigation was the Department of Justice’s Civil Rights Division.115 If not quite a trailblazer in prison litigation,116 the Department was an early force. Title III of the Civil Rights Act of 1964 gave the Department statutory litigating authority to sue for the integration of public facilities, including jails and prisons.117 After filing its first correc1967), relief augmented, 404 F.2d 571 (8th Cir. 1968), one of the appointed lawyers was Edward L. Wright, who concurrently headed up the American Bar Association committee that wrote the 1970 Code of Professional Responsibility. Wright was elected president of the ABA shortly thereafter. See Edward L. Wright, The Code of Professional Responsibility: Its History and Objectives, 24 ARK. L. REV. 1. 1-2 (1970); Geoffrey C. Hazard, Jr., The Future of Legal Ethics, 100 YALE L.J. 1239, 1252 (1991) (describing Wright). And Steele Hays, the lawyer Judge Henley appointed in the fourth case, Holt v. Sarver, 300 F. Supp. 825 (E.D. Ark. 1969) (“Holt I”), was the son of a long-time Arkansas congressman; he was appointed to the state judiciary the same year he represented the inmates (and was eventually elected to the Arkansas Supreme Court, where he served for years alongside Chief Justice Holt). See Brief Sketches of Judges on State Supreme Court, ARK. GAZETTE, July 23, 1990, at 6A; Editorial, For Justice Hays Again, ARK. GAZETTE, Oct. 14, 1990, at 2C. 114. GREENBERG, CRUSADERS, supra note 33, at 457. 115. In the interests of full disclosure, I should note that I served as a trial attorney in the Special Litigation Section of the Civil Rights Division from 1995 to 1998 and that much of my work there was on jail and prison litigation. 116. The Department did bring a number of criminal prosecutions of prison guards, for violation of inmates’ civil rights, as early as the 1950s. See sources cited supra note 25. 117. See Title III of the Civil Rights Act of 1964, 42 U.S.C. § 2000b (1994); HANDLER, SOCIAL MOVEMENTS, supra note 82, at 114-15 (arguing that the Civil Rights Act of 1964 reflected Congress’s recognition that “the courts and social-reform groups could not fight desegregation battles alone”; “[t]he federal government . . . would, it was hoped, . . . relieve some of the burden on the civil-rights organizations”) (citation omitted). In addition, the Attorney General had enforcement authority under Title VI of the Act, which provided that “[n]o person . . . shall, on the ground of race, color, or national origin, be excluded from participation in, be denied the benefits of, or be subjected to discrimination under any program or activity receiving Federal financial assistance.” Title VI of the Civil Rights Act of May 1999] Judicial Policymaking 2023 tional desegregation lawsuit in 1969,118 the Justice Department soon got into the prison and jail desegregation business in a fairly serious way, initiating its own desegregation investigations and lawsuits119 and intervening in a number of privately initiated desegregation lawsuits.120 Justice Department authority to participate in more general conditions cases was somewhat shaky.121 Nonetheless, the Civil Rights Division typically used the occasion of desegregation lawsuits to reach more general conditions as well, whatever the technicalities of litigating authority.122 For example, 1964, § 601, 42 U.S.C. § 2000d (1994); see also State and Local Fiscal Assistance Act of 1972, § 122(a), 31 U.S.C. § 1242 (1982) (current version at 31 U.S.C. § 6716 (1994)) (same rule, for recipients of revenue sharing funds). 118. See United States v. Ashley, Civ. A. No. 69-739 (D.S.C., Anderson Div. consent judgment entered Mar. 2, 1970) (desegregating the Anderson County jail). 119. Other desegregation lawsuits included: United States v. Wyandotte County, Civ. A. No. K.C-3163 (D. Kan. filed June 5, 1970), relief denied, 343 F. Supp. 1189 (D. Kan. 1972), revd., 480 F.2d 969 (10th Cir. 1973), relief granted, Jan. 29, 1974; United States v. McCall, Civ. No. 71-2 (M.D. Fla. filed Jan. 18, 1971) (county jail); United States v. Rowan County Jail, C.A. No. C63-S-73 (M.D.N.C. filed Feb. 19, 1973; summary judgment granted May 24, 1974); United States v. Polk County, No. 73-252-Civ. (M.D. Fla. filed May 23, 1973; settled Mar. 8, 1974); United States v. Sumter County, C.A. No. 73-9-Civ-OC (M.D. Fla. filed May 23, 1973; settled Mar. 4, 1974); United States v. Pompano Beach City Jail, C.A. No. 73-820-Civ-JE (S.D. Fla. filed May 14, 1973; subsequently settled); United States v. Elrod, No. 76-C-4768 (N.D. Ill. filed 1976; consent decree entered Jan. 4, 1980) (Cook County Department of Corrections facilities); United States v. Reid, No. CI-77-0010 (W.D. La. filed Jan 6, 1977; consent decree entered May 16, 1977) (Calcasieu Parish jail); United States v. Illinois, No. S-Civ-76158 (S.D. Ill. filed Dec. 29, 1976; consent decree entered July 26, 1978) (Illinois Correctional Center system). See also Memorandum from Stephen A. Whinston, Attorney, Special Litigation Section, to Paul S. Lawrence, Acting Chief, Special Litigation Section (Mar. 3, 1978) (discussing proposed litigation concerning the Vernon Parish Jail); Memorandum from Civil Rights Division to the Attorney General (Aug. 8, 1978) (setting out details of desegregation investigations and litigations against numerous Louisiana parish jails). 120. See, e.g., intervention cases listed infra note 124; see also Memorandum from J. Stanley Pottinger, Assistant Attorney General, Civil Rights Division, to Director, Federal Bureau of Investigation (Aug. 11, 1975) (requesting FBI assistance with ongoing Division enforcement in Georgia jails of the desegregation judgment obtained by private parties in Wilson v. Kelley); Wilson v. Thompson, No. 75-36-ALR (M.D. Ga. consent decree entered Aug. 22, 1975) (Early County jail); Wilson v. Bloodworth, No. 2922-Mac (M.D. Ga. opened as separate case Oct. 12, 1973) (Bibb County jail); Stewart v. Rhodes, 473 F. Supp. 1185 (S.D. Ohio 1979) (U.S. was amicus in a case about segregation at Columbus Correctional Facility, the reception facility for the Ohio state penal system); Lamar v. Coffield, 951 F. Supp. 629 (S.D. Tex. 1996) (case filed in 1972; U.S. intervened; consent decree approved 1977; denying post-judgment motions for modification). For a description from the ground level of the Justice Department’s commitment to correctional desegregation, see Wilbert Rideau & Billy Sinclair, Prisoner Litigation: How It Began in Louisiana, 45 LA. L. REV. 1061, 1072 (1985) (account by two inmates of 1973 desegregation of Louisiana State Prison at Angola). 121. See United States v. Solomon, 419 F. Supp. 358 (D. Md. 1976), affd., 563 F.2d 1121 (4th Cir. 1977) (holding that in the absence of statutory authority the Attorney General may not sue to remedy unconstitutional conditions of confinement for institutionalized people with mental retardation); United States v. Mattson, 600 F.2d 1294 (9th Cir. 1979) (same); United States v. Philadelphia, 644 F.2d 187 (3d Cir. 1980) (same, for allegedly unconstitutional actions by police department). 122. See Memorandum from James P. Turner, Deputy Assistant Attorney General, to Jerris Leonard, Assistant Attorney General, re. Wyandotte County jail lawsuit (Oct. 22, 2024 Michigan Law Review [Vol. 97:1994 the Department relied on its Title III desegregation authority to intervene in 1971 in the second of the statewide prison conditions cases, Gates v. Collier, which dealt with the Mississippi State Penitentiary at Parchman.123 Once a part of the case, the Department challenged the conditions at Parchman on grounds moving far beyond the Equal Protection Clause.124 In addition, through the 1970s, the Department was asked or ordered by numerous judges to appear in various kinds of non-desegregation institutional reform cases, including jail and prison conditions cases, and it appeared in others on its own initiative.125 In total, prior to 1980, the Department of Justice was either plaintiff, plaintiff-intervenor, or amicus (almost always “litigating amicus,” participating in discovery, negotiation, and presentation of evidence) in more than ten of the largest and most comprehensive prison cases (four of which had desegregation components)126 and in a number of jail cases.127 1971) (“It is my hope to use the case not only to desegregate but to restructure the disciplinary and correctional system.”); Whinston interview, supra note 110. 123. Gates v. Collier, 349 F. Supp. 881 (N.D. Miss. 1972), affd., 501 F.2d. 1291 (5th Cir. 1974); see OSHINSKY, supra note 88, at 245. 124. See Gates, 501 F.2d 1291. Other examples of Justice Department race-related interventions or complaints coupled with broader issues, either in the complaint or in the eventual remedies include: Stockton v. Alabama Industrial School for Negro Children, No. 2834-N (M.D. Ala. filed Jan. 23, 1969; judgment entered July 23, 1971); Wyandotte, 480 F.2d 969 (filed June 5, 1970; relief granted Jan. 29, 1974); Williams v. Edwards, 547 F.2d 1206, 1207, 1208 (5th Cir. 1977); Battle v. Anderson, 376 F. Supp. 402, 407 (E.D. Okla. 1974); United States v. Illinois, No. S-Civ.-76-158 (S.D. Ill. filed Dec. 29, 1976; consent decree entered July 26, 1978); Vest v. Lubbock County, No. CA-5-76-53 (N.D. Tex. filed May 14, 1976; judgment entered June 7, 1977); Soileau v. Phelps, No. 76-1302 (W.D. La. filed Dec. 9, 1976; consent decree entered Sept. 18, 1978); Breaux v. Phelps, No. 77-0121 “C” (W.D. La. filed Feb. 3, 1977; consent decree entered Aug. 21, 1980). 125. For description and discussion of one order requesting Justice Department participation, see Estelle v. Justice, 426 U.S. 925 (1976) (Rehnquist, J.) (dissenting from denial of certiorari). See also Civil Rights for Institutionalized Persons: Hearings on H.R. 2439 and H.R. 5791 before the Subcommittee on Courts, Civil Liberties, and the Administration of Justice of the House Committee on the Judiciary, 95th Cong. 297-306 (1977) (written submission by Civil Rights Division, listing other jail and prison cases in which judges requested or ordered the United States to appear, either as a party or as amicus). 126. See Gates, 349 F. Supp. at 885, 887, 893, 900-01, affd., 501 F.2d at 1299-1301 (judgment entered in 1972; had a desegregation component); Newman v. Alabama, 349 F. Supp. 278 (M.D. Ala. 1972), affd. in part, 503 F.2d 1320 (5th Cir. 1974); Battle, 376 F. Supp. at 402 (had a desegregation component); Harris v. Cardwell, No. CIV-75-185-PHX-CAM (D. Ariz. consent order entered Oct. 14, 1980) (Arizona State Penitentiary); Costello v. Wainwright, 397 F. Supp. 20 (M.D. Fla. 1975); Williams, 547 F.2d at 1206 (had a desegregation component); Pugh v. Locke, 406 F. Supp. 318 (M.D. Ala. 1976); Kendrick v. Bland, 541 F. Supp. 21 (W.D. Ky. 1982); Guthrie v. Evans, Civ. A. No. CV 3068 (S.D. Ga. filed Sept. 1972), contempt motion settled, 93 F.R.D. 390, 391-93 (S.D. Ga. 1981) (describing procedural history) (had a desegregation component); Ruiz v. Estelle, 503 F. Supp. 1265 (S.D. Tex. 1980); Gavin v. Ray, No. 4-78-CV-70062 (S.D. Iowa settlement entered, June 13, 1984) (Iowa State Penitentiary); Parrott v. Ray, C.A. No. 78-174-2 (S.D. Iowa) (Iowa State Penitentiary); Hoptowit v. Ray, 682 F.2d 1237 (9th Cir. 1982) (affirming in part and reversing in part 1980 district court (E.D. Wash.) order); French v. Owens, 538 F. Supp. 910 (S.D. Ind. 1982); Balderas v. Matheson, 75C-220 (D. Utah consent decree entered Aug. 31, 1979) (Utah State Prison). 127. In addition to the jail desegregation and conditions cases cited supra notes 118-120 & 124, see also: Tyler v. Percich, No. 74-40 C(2) (E.D. Mo. judgment entered Oct. 2, 1974) May 1999] Judicial Policymaking 2025 Assessment of the Department of Justice’s resources, goals and strategies aids a great deal in understanding the shape of litigated prison reform.128 When the Civil Rights Division was involved, general conditions cases could more easily be statewide, and more comprehensive, because the Division called upon the FBI to perform statewide investigations and paid the (very high) expenses of such comprehensive litigation, including expert fees.129 Even when the Division did not seek, or was not successful in seeking, to widen the issues past desegregation, the consent decrees the Division negotiated typically required the defendant jail or prison to devise a standardized scheme for assigning inmates to housing and custody and security levels.130 Such “objective classification” is a substantial safeguard against segregation, but it is also much more. Penologists consider “objective classification” a significant reform measure contributing to prison safety as well as regularity (and lack of favoritism) because it forces prison officials to gather and use individual information about each inmate and his or her background and adaptation to prison life.131 This kind of decree provision contributed to the “bureaucratization” that Feeley and Rubin, like other scholars, identify as an important outcome of litigated prison reform.132 Most significantly, though, the Civil Rights Division’s participation was subject to political changes to an even greater extent than the activities of legal services offices. The inauguration of the Reagan administration halted Justice Department initiation of new lawsuits, (St. Louis city jail); Adams v. Mathis, 458 F. Supp. 502 (M.D. Ala. 1978) (Houston County jail); Brown v. Vann, C.A. No. 77-P-0554-S (N.D. Ala. judgment entered Apr. 18, 1978) (Birmingham city jail); Devonish v. Garza, 510 F. Supp. 658 (W.D. Tex. 1981) (Bexar County jail). 128. Cf. HANDLER, SOCIAL MOVEMENTS, supra note 82, at 117 (“During the period of the [school] desegregation campaigns, social change through law-reform litigation simply required too many individual lawsuits in too many places. The social-reform groups required the active intervention of the federal government. When this happened the pace of desegregation quickened. When the federal government backed off, the pace slackened.”). 129. The litigation of Ruiz v. Estelle reportedly cost the Justice Department more than $1 million. See Elizabeth Alexander, The Overall Context of Prison Litigation, 449 PLR/Lit. 401, 412 (1992). 130. See, e.g., Compliance Report filed Dec. 15, 1982, app. H, United States v. Elrod, No. 76-C-4768 (N.D. Ill.) (classification plan for Cook County jail); Wilson v. Thompson, No. 7536-ALR (M.D. Ga. consent decree entered Aug. 22, 1975) (desegregation of Early County jail). 131. On classification, see, for example, PREDICTION AND CLASSIFICATION: CRIMINAL JUSTICE DECISION MAKING (Don M. Gottfredson & Michael Tonry eds., 1987). On classification litigation, see, for example, Barbara A. Belbot & James W. Marquart, The Political Community Model and Prisoner Litigation: Can We Afford Not to Try a Better Way?, 78 PRISON J. 299 (1998). 132. Pp. 271-90. See, e.g., JACOBS, supra note 14, at 54; JAMES B. JACOBS, STATEVILLE: THE PENITENTIARY IN MASS SOCIETY 105-23 (1977); Sturm, Legacy and Future, supra note 13, at 665-68. 2026 Michigan Law Review [Vol. 97:1994 at least for several years,133 and the Department even switched sides in a number of its ongoing litigations.134 Consideration of the nonjudicial participants in litigated prison reform not only augments Feeley and Rubin’s factual account of the movement, it also enables more nuanced, and I think more accurate, interpretation of those facts. For example, Feeley and Rubin identify the difference between prisons in the South and prisons elsewhere as the “problem” that prompted litigated prison reform. They argue that it was this variance in penal philosophy that cued judicial embrace of prison reform, and that the judicial response to the southern variant of prison was homogenization, forcing the outlier southern prisons to conform to the national model. I think they greatly overstate the point. It is true that a number of prison systems of the former Confederate states — especially systems in Mississippi,135 Arkansas (pp. 52-55), Louisiana,136 and the 133. During the 12 years of the Reagan and Bush administrations, the Department brought, or participated as a party in, fewer than ten new prison cases: Canterino v. Wilson, 546 F. Supp. 174 (W.D. Ky. 1982) (women’s facilities); Davis v. Armistead, 575 F. Supp. 695 (M.D. La. 1983) (forensic facility); United States v. Michigan, 680 F. Supp. 928 (W.D. Mich. 1984-1987) (compilation of many orders, including 1984 consent decree); United States v. Virgin Islands, No. 86-265 (D.V.I. filed Nov. 21, 1986; consent decree entered Dec. 1, 1986); United States v. Hawaii, 564 F. Supp. 189 (D. Haw. 1983) (dismissing the action); United States v. Alabama, No 77-V-14-N (M.D. Ala. dismissed Sept. 14, 1987); United States v. California, No. CV-89-1233-EJG-JFM (N.D. Ca. filed Sept. 12, 1989; consent decree entered concurrently) (California Medical Facility, Vacaville); United States v. Guam, No. CIV9100020 (D. Guam filed Feb. 13, 1991; initial consent decree entered May 16, 1991). Under the Clinton administration, the Department has continued to litigate a number of previously filed cases, but has filed just five new lawsuits involving prison conditions. See United States v. Montana, No. 94-90-H-CCL (D. Mont. settled by conditional dismissal, Sept. 12, 1996 and Jan. 20, 1997); Williams v. Lynn, No. 92-0001-B (E.D. La.) (medical care in Louisiana State Penitentiary); United States v. Michigan, No. 97-40053 (E.D. Mich. filed Mar. 10, 1997) (Michigan women’s prisons; currently pre-trial); United States v. Arizona, No. 97-476-PHXROS (D. Ariz. settled Mar. 11, 1999) (Arizona women’s prisons); United States v. Commonwealth of the Northern Mariana Islands, No. CV 99-0017 (D. N. Mar. I. consent decree entered Feb. 25, 1999). The Clinton Justice Department has, however, focused more attention on jails and, especially, juvenile corrections facilities. See “Activity under Civil Rights of Institutionalized Persons Act by Type of Facility” (Aug. 10, 1998) (Civil Rights Division memorandum). 134. See Sturm, Legacy and Future, supra note 13, at 738; Attorney General William P. Barr, Remarks at the Attorney General’s Summit on Corrections 13, 14 (April 27, 1992) (unpublished speech) (setting out new Justice Department policy regarding its affirmative prison litigation, including decision to support the defendants in Texas and Michigan litigation in which United States was plaintiff). See generally Robert D. Dinerstein, The Absence of Justice, 63 NEB. L. REV. 680 (1984) (discussing and criticizing the Reagan Justice Department’s implementation of CRIPA). 135. See OSHINSKY, supra note 88; WILLIAM BANKS TAYLOR, BROKERED JUSTICE: RACE, POLITICS, AND MISSISSIPPI PRISONS, 1798-1992, at xii (1993); David M. Lipman, Mississippi’s Prison Experience, 45 MISS. L.J. 685, 694-95 (1974). 136. See MARK T. CARLETON, POLITICS STATE PENAL SYSTEM 6–7 (1971). SIANA AND PUNISHMENT: THE HISTORY OF THE LOUI- May 1999] Judicial Policymaking 2027 major facilities in Texas137 — were run for many years on a “plantation” model.138 These self-financing forced-labor farm prisons were direct heirs to the slave plantation and the near-slavery systems of labor peonage139 and convict-leasing140 that succeeded the end of Reconstruction in the South.141 (Not all the southern prisons were run on a plantation model — something Feeley and Rubin would 137. See BEN M. CROUCH & JAMES W. MARQUART, AN APPEAL TO JUSTICE: LITIGATED REFORM OF TEXAS PRISONS 15-16 (1989); STEVEN J. MARTIN & SHELDON EKLAND-OLSON, TEXAS PRISONS: THE WALLS CAME TUMBLING DOWN 5 (1987). 138. Mark Carleton wrote in 1971 that “[b]oth the facilities and the philosophy of prisons in the South, especially in the Deep South, were tailor made for black convicts as viewed by their white former masters in the post-Civil War period. Today, despite gradual alterations and nominal progress, these institutions remain much as they were at the turn of the century . . . . Systems which have managed most successfully to diversify their operations away from farming, thereby serving more effectively the needs of modern rehabilitation, are located where there are fewer blacks and, hence, fewer black convicts — the border states and Texas.” CARLETON, supra note 136, at 197; see id. at 197-98 (identifying Arkansas, Mississippi, and Louisiana as the worst of the southern systems); see also BLAKE MCKELVEY, AMERICAN PRISONS: A HISTORY OF GOOD INTENTIONS 197-216 (1977). 139. See Schmidt, Peonage Cases, supra note 22, at 650-55. As Schmidt describes, peonage was a system under which private employers forcibly conscripted laborers who they claimed had signed and broken labor contracts; in his description, it encompasses convict leasing as well. 140. See, e.g., Jamison v. Wimbish, 130 F. 351, 355-57 (S.D. Ga. 1904) (Speer, J.) (describing chain gang); EDWARD L. AYERS, VENGEANCE AND JUSTICE: CRIME AND PUNISHMENT IN THE 19TH-CENTURY AMERICAN SOUTH 185-222 (1984); Lipman, supra note 135, at 688-92. For a comprehensive look at the convict leasing system, state by state, that describes the “penal plantation[s]” of Texas, Arkansas, Mississippi, and Louisiana, and the farming operations of Florida, Georgia, and North Carolina, as part of the effort to “solve the labor problem” caused by the reformist abolition of convict leasing, see Hilda Jane Zimmerman, Penal Systems and Penal Reforms in the South Since the Civil War 444-73 (1947) (unpublished Ph.D. dissertation, University of North Carolina). 141. See SOUTHERN REGIONAL COUNCIL, THE DELTA PRISONS: PUNISHMENT FOR PROFIT 1 (March 1968) (“In general, the abolition of [convict] leasing produced two different systems: a work camp system in the Southeastern states, and a state farm system in the South central states. . . . The county camp system . . . allows local governments to use prison labor to maintain roads and other public works.”); JOHN BARTLOW MARTIN, BREAK DOWN THE WALLS: AMERICAN PRISONS: PRESENT, PAST, AND FUTURE 205-09 (1954) (describing road crew work as leading farm work in southern prisons, and observing that “[t]he main object of southern penology is to get some work out of wrongdoers”). Virginia and Georgia, for example, were mixed systems. See Landman v. Royster, 333 F. Supp. 621, 626 (E.D. Va. 1971) (describing mixed system with non-farm penitentiary holding 1100 inmates, various farm units holding 2400, and road camps holding 2200); PAUL W. KEVE, THE HISTORY OF CORRECTIONS IN VIRGINIA 5 (1986) (describing penal farm and road camps); id. at 119-25 (describing road camps and lime-grinding plants); GEORGIA ADVISORY COMM. TO THE U.S. COMM. ON CIVIL RIGHTS, GEORGIA PRISONS 14-36 (1976) (describing a mixed system, with industries and farming at Reidsville, the largest Georgia facility, and other kinds of operations elsewhere; one quarter of the states’ inmates were housed in county facilities to do road work). Moreover, many states that had once run either road or farm operations had ended them by the time of the onset of litigated reform. Alabama, for example, had largely ended its plantation system a few years prior to the onset of prison reform litigation. See Pugh v. Locke, 406 F. Supp. 318, 326 (M.D. Ala. 1976) (describing pervasive idleness in Alabama system); RAY A. MARCH, ALABAMA BOUND: FORTY-FIVE YEARS INSIDE A PRISON SYSTEM 67 (1978) (oral history of Warden Oscar Dees, describing reforms of 1955 or 1960); id. at 7074 (describing unused farm land). See also MCKELVEY, supra note 138, at 328-29 (describing reforms in Maryland, South Carolina, and Florida); SOUTHERN REGIONAL COUNCIL, supra, at 3 (describing “many of the state systems, especially the work camp-oriented systems of the 2028 Michigan Law Review [Vol. 97:1994 have done better to acknowledge. Road camp systems, in which inmates were spread around the state to work in chain gangs along county roads, were equally, if not more, common throughout the South.) But although the plantation model, and, more generally, southern prisons, were the subject of litigated attack and resulting reform, the historical evidence belies the claim that the litigated reform movement had its origins in the southern plantation prisons’ deviance from a widely accepted national norm of what prisons should look like. The evidence against this claim of origins is simple: the southern cases happened concurrent with, not earlier than, prison and jail cases all over the nation in which courts ordered remedies for unconstitutional conditions. In 1966, for example, a California district court issued an injunction regulating solitary confinement cells, which were an important instrument of control in that state’s prison system.142 The Second Circuit made a similar ruling relating to New York’s system in 1967.143 In 1969, Judge Raymond Pettine entered the first of many orders regulating conditions at Rhode Island’s one-prison correctional system.144 In 1971, a lawsuit on behalf of inmates at Attica over prison officials’ post-riot abuse and interrogation won quick preliminary injunctive relief.145 And it was only September 1972 when Ohio saw its first major prison decree in Taylor v. Perini, which governed many aspects of conditions at the Marion Correctional Institution, a facility that housed over 1300 inmates.146 Moreover, though comprehensive data is not available, it seems that the majority of the earliest jail litigation Southeast,” as “hav[ing] undergone some degree of reform in recent years,” and naming as sites of reform Alabama, Florida, North Carolina, South Carolina, and Tennessee). 142. Jordan v. Fitzharris, 257 F. Supp. 674 (N.D. Cal. 1966). See discussion supra notes 35 & 106 and accompanying text. 143. Wright v. McMann, 387 F.2d 519 (2d Cir. 1967). See discussion supra notes 35 & 106 and accompanying text. 144. Morris v. Travisono, 310 F. Supp. 857, 857 (D.R.I. 1970) (describing earlier temporary order). The initial Morris order was a limited one; it covered only “minimal maintenance of personal hygiene[,] . . . outdoor exercise and access to religious services.” Id. at 858. Soon, however, Judge Pettine handed down a far more sweeping set of procedural rules governing discipline. See id. Two years later, an even wider decree established regulations for discipline, classification, and mail. Prison Reform: The Judicial Process: A BNA Special Report on Judicial Involvement in Prison Reform, 23 CRIMINAL L. REP., Aug. 2, 1978, supp. at 4. And Palmigiano v. Garrahy, 443 F. Supp. 956 (D.R.I. 1977) was a comprehensive litigated reform effort. See CARROLL, supra note 102, at 128-307. 145. Inmates of Attica Correctional Facility v. Rockefeller, 453 F.2d 12 (2d Cir. 1971). 146. See Taylor v. Perini, 413 F. Supp. 189, app. A (N.D. Ohio 1976) (reprinting order of September 12, 1972, adopting consent decree). May 1999] Judicial Policymaking 2029 was nonsouthern;147 this regional pattern has continued into the present.148 Besides, since Feeley and Rubin set up the issue not as one of fact, but of felt experience, it is even more significant that when contemporary observers wrote about early prison cases, they did not focus on the distinctive southern flavor of the plantation prison. They did not, that is, trumpet Holt v. Sarver as the paradigm case; rather, they joined description of Holt with discussion of Jordan v. Fitzharris (from California’s Soledad prison), Wright v. McMann (from New York’s Dannemora prison), and Rhem v. Malcolm (from the Manhattan House of Detention for Men, known as “the Tombs”).149 Although scholarly case studies written more recently almost all concern the South,150 this should not hide the very real impact of the nonsouthern cases in shaping the litigation as a national movement. At the same time, Feeley and Rubin are indisputably correct that prison (if not jail) litigated reform was more prevalent in the South than elsewhere.151 But several explanations with more per147. The “‘granddaddy’ of jail suits” was a case that “forced comprehensive changes” at the Lucas County Jail in Toledo, Ohio. Taft, supra note 102, at 23, 24 (discussing Jones v. Wittenberg, 323 F. Supp. 93 (N.D. Ohio 1971), affd. sub nom. Jones v. Metzger, 456 F.2d 854 (6th Cir. 1972)). There were also, for example, major early jail orders in New York and California in 1971, Rhem v. McGrath, 326 F. Supp. 681 (S.D.N.Y. 1971); Brenneman v. Madigan, 343 F. Supp. 128, 133 (N.D. Cal. 1972) (discussing preliminary order of 1971); in Boston in 1973, Inmates of Suffolk County Jail v. Eisenstadt, 360 F. Supp. 676 (D. Mass. 1973), affd., 494 F.2d 1196 (1st Cir. 1974); and in Puerto Rico in 1976, Martinez Rodriguez v. Jimenez, 409 F. Supp. 582 (D.P.R. 1976). 148. Indeed, southern jails have consistently been underrepresented among those jails subject to court order. For example, at the time of the most recent national jail census, in 1993, jails in the South housed about 40% of the nation’s jail inmates, but southern jails housed just 33% of those inmates in jails subject to court order. Data are derived from BUREAU OF JUSTICE STATISTICS, U.S. DEPT. OF JUSTICE, NATIONAL JAIL CENSUS, 1993 (1996) (ICPSR 6648) [hereinafter BUREAU OF JUSTICE STATISTICS, 1993 JAIL CENSUS] (raw data available from the National Archive of Criminal Justice Data (visited June 1, 1999) <http://www.icpsr.umich.edu/NACJD/archive.html>). I am including as “southern” the states of the Confederacy — Florida, Georgia, Louisiana, Arkansas, Alabama, Mississippi, North Carolina, South Carolina, Tennessee, Texas, and Virginia — and the District of Columbia. 149. See, e.g., Joseph C. Kearfott, Note, Decency and Fairness: An Emerging Judicial Role in Prison Reform, 57 VA. L. REV. 841 (1971); Note, Prisoners’ Rights Under Section 1983, 57 GEO. L.J. 1270 (1969). 150. An important exception is last year’s addition to the case study literature, CARROLL, supra note 102, which examines the Rhode Island prison litigation. A great deal of the case study literature is catalogued by Sturm, Legacy and Future, supra note 13, at 648-52. 151. The six earliest successful large southern cases (counting Oklahoma as southern) were also the six first successful statewide prison reform litigations. The cases were in: Arkansas (Holt v. Sarver, 309 F. Supp. 262 (E.D. Ark. 1970), affd., 442 F.2d 304 (8th Cir. 1971)); Mississippi (Gates v. Collier, 349 F. Supp. 881 (N.D. Miss 1972), affd., 501 F.2d 1291 (5th Cir. 1974)); Oklahoma (Battle v. Anderson, 376 F. Supp. 402 (E.D. Okla 1974)); Florida (Costello v. Wainwright, 397 F. Supp. 20 (M.D. Fla. 1975)); Louisiana (Williams v. Edwards, 547 F.2d 1206 (5th Cir. 1977) (order entered 1975)); Alabama (Pugh v. Locke, 406 F. Supp 318 (M.D. Ala. 1976)). Even today, southern prisons house about 35% of the nation’s inmates, but about 53% of the nation’s inmates currently in facilities subject to court order. Data are derived from BUREAU OF JUSTICE STATISTICS, U.S. DEPT. OF JUSTICE, CENSUS OF STATE 2030 Michigan Law Review [Vol. 97:1994 suasive power than Feeley and Rubin’s incubator theory suggest themselves once the theoretical lens is widened so that sustained prison litigation is not conceptualized as primarily caused by judges’ perception of a “problem.” Looking, instead, at the interaction between sympathetic judges and a set of advocates who saw a potential for urging change by lawsuit and had both resources to bring case after case and expertise to work effectively within the legal frameworks governing both contested and settled orders,152 I can propose several alternative explanations. Each is consistent, as Feeley and Rubin’s southern incubator theory is not, with the nonsouthern flavor of jail orders, which were largely litigated by federally funded legal services offices spread around the country.153 Perhaps the South was the site of the biggest and most sweeping judicial interventions into prison administration because the South was where LDF had cooperating attorneys interested in prison reform. Perhaps it was because the South was where the segregated prisons were most concentrated (in both plantation and nonplantation systems), giving the litigants and judges a clear doctrinal hook for a federal case.154 Or perhaps it was that the Civil Rights Division focused its efforts on the South, where its desegregation authority was most helpful, and where judges sympathetic to civil rights litigation had by 1970 grown accustomed, in school desegregation cases, to looking to the Division for assistance in civil rights injunctive suits.155 Or, to leave the realm of institutional answers for a moment, perhaps the South simply had worse prisons, so that a national trend toward litigated reform had its greatest impact there. Whatever the reason, it seems to me unlikely that its answer can be derived through a description so exclusively focused on judges. For these cases, Feeley and Rubin’s theory obscures rather than furthers analysis. FEDERAL ADULT CORRECTIONAL FACILITIES, 1995 (1998) (ICPSR 6953) [hereinafter BUREAU OF JUSTICE STATISTICS, 1995 PRISON CENSUS] (raw data available from the National Archive of Criminal Justice Data (visited June 1, 1999) <http://www.icspr.umich.edu/NACJD/ archive.html>). As with jails, I am including as “southern” the states of the Confederacy (see supra note 148 for list) and the District of Columbia (which houses its prisoners in Virginia). AND 152. National prison litigation shops not only had resources and expertise; they also had the geographic scope to bring lawsuits in districts whose federal judges were likely to hand down “some useful precedent.” Bass interview, supra note 91. 153. See DOOLEY & HOUSEMAN, supra note 100, at 26-27. 154. See, e.g. Chilton, supra note 90, at 19-22 (describing the start of Guthrie v. Evans, C.A. No. 73-3068 (S.D. Ga.), contempt motion settled, 93 F.R.D. 390 (S.D. Ga. 1981), by inmates seeking desegregation of the Georgia State Prison). 155. For cases in which the United States intervened as plaintiff, see supra notes 120 & 124. Judicial Policymaking May 1999] III. BEYOND THE 2031 JUDGE As I stated at this review’s outset, Feeley and Rubin have a great deal of company when they place the judge at the center of the action in litigated institutional reform.156 At the same time, my advocacy of a more populated analysis is hardly new. As its footnotes reveal, this review has followed the lead of, particularly, Susan Sturm, who has written a rich set of articles about prison litigation.157 More theoretically, it was more than fifteen years ago that Robert Cover described members of social movements as “jurisgenerative” — law-creating — and of judges as only, and derivatively, “jurispathic” — law-killing, by their choice of one or another legal visions presented to them.158 And many political scientists and sociologists have long advocated “bottom-up” analysis of litigation as reform tool, focusing less on judges and more on litigants, less on courts altogether and more on disputes, wherever they occur.159 In a fascinating recent book in this tradition, political scientist Charles Epp examines what he calls the American “rights revolution” of the 1960s, along with similar “revolutions” in India, Britain, and Canada. He concludes that “the common emphasis on constitutional provisions and judges is exaggerated” as a causal explanation for these law reform campaigns.160 His comparative data undermine “judge-centered explanation[s],” but buttress the account I have offered here — as he puts it, law reform by litigation “depends on resources, and resources for rights litigation depend on a support structure of rights-advocacy lawyers, rights-advocacy organizations, and sources of financing.”161 156. See supra notes 10, 12, & 13 and accompanying text. 157. See sources cited supra note 13. 158. Robert M. Cover, The Supreme Court, 1982 Term — Foreword: Nomos and Narrative, 97 HARV. L. REV. 4, 11, 40 (1983). 159. Generative books setting out bottom-up visions are STUART A. SCHEINGOLD, THE POLITICS OF RIGHTS: LAWYERS, PUBLIC POLICY, AND POLITICAL CHANGE (1974), and Joel Handler’s Social Movements and the Legal System, supra note 82. Also relevant are two major works arguing that courts are ineffective agents of social change, Donald L. Horowitz’s The Courts and Social Policy, supra note 12, and Gerald N. Rosenberg’s The Hollow Hope, supra note 13. For a fairly recent guide to the debate and the sources, see, for example, Michael W. McCann, Reform Litigation on Trial, 17 L. & SOC. INQUIRY 715, 729-42 (1992) (reviewing ROSENBERG, supra). And for an excellent collection of articles situated in the “bottom-up” school, see CAUSE LAWYERING: POLITICAL COMMITMENTS AND PROFESSIONAL RESPONSIBILITIES (Austin Sarat & Stuart Scheingold eds., 1998). Recent examples of similar thinking inside constitutional theory are William N. Eskridge, Jr., Public Law from the Bottom Up, 97 W. VA. L. REV. 141 (1994), and Michael J. Klarman, Rethinking the Civil Rights and Civil Liberties Revolutions, 82 VA. L. REV. 1, 32 (1996) (calling for scholarship “identifying and elaborating the background historical forces that rendered possible the postwar revolution in civil rights and civil liberties jurisprudence” that undercuts “the myth of the Court’s countermajoritarian heroics”). 160. EPP, THE RIGHTS REVOLUTION, supra note 96, at 5. 161. Id. at 14, 18. 2032 Michigan Law Review [Vol. 97:1994 Perhaps because most of the scholars examining civil rights injunctive cases as a special field of inquiry have been lawyers, not social scientists, they have not situated their work either in or in opposition to the political scientists’ “decentered” scholarship.162 Feeley and Rubin are certainly familiar with the debate between “bottom-up” and “top-down” approaches; Feeley, a political scientist, has been a long-time participant.163 I therefore must add to my regret about their judge-centeredness an additional regret that they elected to forgo the opportunity to bridge the gap between literatures, by explaining the theoretical underpinnings of their approach. Still, if, as I have argued, litigated prison reform cannot support the conceptual framework Feeley and Rubin seek to impose on it, Feeley and Rubin’s provocative book nonetheless contains many insights into courts and cases. It is an important contribution both to the scholarly debates about litigated reform of prisons and other institutions and to discussion of the phenomenology of judicial doctrinemaking. While it would be foolish to “replace a theory of judicial control of the agenda with its mirror image, a theory of complete control by strategic litigators,”164 it is time, I think, to complicate the picture scholars draw of institutional reform litigation. The task is an important one: although litigated reform of governmental institutions is no longer exciting simply because novel, it remains a regular and consequential component of the interaction between the court system and the executive and legislative branches of state and local governments.165 In the area of jails and prisons, litigated reform 162. I borrow this term from McCann, supra note 159, at 730. 163. See, e.g., Malcolm M. Feeley, Hollow Hope, Flypaper, and Metaphors, 17 L. & SOC. INQUIRY 745 (1993). 164. EPP, THE RIGHTS REVOLUTION, supra note 96, at 22; Wasby, supra note 105, at 352 (“Litigation for social change . . . is often reflexive and far from completely planned, with many constraints on the planning of litigation campaigns, many detours along the road to organizational goals and much flexibility of action by both the litigation organizations and individual staff attorneys.”). 165. There has, unfortunately, been a decided decrease in the amount of scholarly analysis of civil rights structural injunction cases in recent years, perhaps because of the oft-stated view that this type of lawsuit has become rarer. See, e.g., p. 145 (litigated reform of prison had by late 1980s “run its course”); Marcus, supra note 10, at 648. (“Chayes’s focus on public law litigation seems ill-conceived because the incidence of the kind of lawsuits he had in mind — school desegregation and prison conditions cases — was waning even as he wrote.”). It is true that the number of class action filings and of civil rights class action filings, brought both by prisoners and nonprisoners, followed a downward trend from their peak in the mid 1970s until the early 1990s (when the numbers leveled off, and even seem to have started increasing). See ADMINISTRATIVE OFFICE OF THE UNITED STATES COURTS, ANNUAL REPORT OF THE DIRECTOR OF THE ADMINISTRATIVE OFFICE OF THE UNITED STATES COURTS (annual volumes for the years 1972-1997). But because institutional reform cases frequently last for many years, all that we can conclude from decreases in new class action filings is that their number probably is not increasing as fast as it used to — not that it is decreasing. The empirical work necessary to evaluate trends in injunctive litigation has not been done, but the claim of decreasing significance seems to me incorrect. While it is certainly true that school May 1999] Judicial Policymaking 2033 continues apace:166 at last count, 27% of the nation’s state prisons, housing 39% of the nation’s state prisoners, and about 17% of the nation’s jails, housing about 40% of the nation’s jail inmates were under court order.167 New focuses of correctional litigation include the needs of women inmates,168 and inmates with disabilities.169 Despite decades of practical and scholarly experience with institutional reform litigation, its causes, its successes, and its failures remain little understood. But the practice now has sturdy roots and corresponding staying power, and understanding its origins and contours is important on its own and as a way of illuminating law as a source of social change generally. Scholarly interest in prison reform, and in institutional reform litigation more broadly, should be augmented by the recently endesegregation cases have grown and continue to grow rarer, see GARY ORFIELD, SUSAN E. EATON, & HARVARD PROJECT ON SCHOOL DESEGREGATION, DISMANTLING DESEGREGATION: THE QUIET REVERSAL OF BROWN V . BOARD OF EDUCATION (1996), other types of litigation continue to proliferate, in state and federal courts, as described in the text. 166. See ACLU National Prison Project, supra note 98 (describing extant state prison court orders); Sturm, Legacy and Future, supra note 13, at 639 (arguing for importance of prison litigation in the future). 167. Data are derived from BUREAU OF JUSTICE STATISTICS, 1993 JAIL CENSUS, supra note 148, and BUREAU OF JUSTICE STATISTICS, 1995 PRISON CENSUS, supra note 151. (My estimates make what seems to me the reasonable assumption that the 558 jails that failed to answer the court order questions in fact have court orders in roughly the same proportion as the remaining 2952 jails.) In fact, the prevalence of prison court orders increased slightly disproportionately (both as compared to the number of prisons, and in terms of covered inmate population) from 1990 to 1995. Data are derived from id.; BUREAU OF JUSTICE STATISTICS, U.S. DEPT. OF JUSTICE, CENSUS OF STATE AND FEDERAL ADULT CORRECTIONAL FACILITIES, 1990 (1993) (ICPSR 9908) (raw data available from the National Archive of Criminal Justice Data (visited June 1, 1999) <http://www.icpsr.umich.edu/NACJD/ archive.html>). 168. See, e.g., Ellen M. Barry, Jail Litigation Concerning Women Prisoners, PRISON J., Spring-Summer 1991, at 44; Ellen M. Barry, Pregnant Prisoners, 12 HARV. WOMEN’S L.J. 189 (1989); Women Prisoners of D.C. Dept. of Corrections v. District of Columbia, 877 F. Supp. 634 (D.D.C. 1994), modified, 899 F. Supp. 659 (D.D.C. 1995), revd. in part, 93 F.3d 910 (D.C. Cir. 1996), on remand, 968 F. Supp. 744 (D.D.C. 1997); United States v. Michigan, No. 9740053 (E.D. Mich. filed Mar. 10, 1997), and Nunn v. Michigan Dept. of Correction, No. 96CV-71416DT (E.D. Mich.) (consolidated cases about Michigan women’s prisons; currently pre-trial); United States v. Arizona, No. 97-476-PHX-ROS (D. Ariz. settled Mar. 11, 1999) (Arizona women’s prisons; currently pre-trial); Agreement between ACLU National Prison Project and Sheriff of Arlington County, Virginia (Feb. 1, 1996) (implementing measures to control sexual misconduct). 169. See Pennsylvania Dept. of Corrections v. Yeskey, 524 U.S. 206 (1998) (holding that the Americans with Disabilities Act covers correctional facilities); Clark v. California, No. C96-1486 FMS, (N.D. Cal. Oct. 1, 1996) (systemwide case about California prisoners with developmental disabilities; interim agreement entered Aug. 12, 1998) (for an earlier, reported, opinion, see 123 F.3d 1267 (9th Cir. 1997) (rejecting states’ 11th Amendment immunity claim)). Coleman v. Wilson, 912 F. Supp. 1282 (E.D. Cal. 1995) (systemwide case about California prisoners with serious mental illness), app. dismissed, 101 F.3d 705 (9th Cir.) (table op., available at 1996 WL 665551); Clarkson v. Coughlin, 898 F. Supp. 1019 (S.D.N.Y. 1995) (injunctive case brought by deaf inmates in New York prison system); see also Belbot & Marquart, supra note 131, at 309-13 (identifying from survey of state correctional lawyers areas of expected litigation increase, including housing, treatment, and job assignment of inmates infected with HIV and other physically and mentally disabled inmates). 2034 Michigan Law Review [Vol. 97:1994 acted Prison Litigation Reform Act (“PLRA”),170 which has significantly altered the rules of prison, jail, and juvenile facility litigation. Most notably, the PLRA radically curtails fee shifting,171 and mandates that, like contested orders, consent decrees must henceforth rest on a judicial finding both as to constitutional violation and as to appropriate scope.172 As a result of this second change, the PLRA requires parties who seek to settle prison cases to choose between “private settlement agreements,” the substantive scope of which may be as broad as the parties agree but which are unenforceable in federal court, or substantively narrower but enforceable court orders.173 The PLRA further requires “termination” of any order (uncontested or litigated) on motion by a party, if the ordered relief is more than two years old and no longer necessary to remedy a current and ongoing constitutional violation;174 this new rule is provoking nationwide relitigation of previously quiet cases, with mixed outcomes.175 As with efforts to understand the origins and history, analysis of the current trends in prison litigation will require careful examination of nonjudicial actors and the process and structure of the litigation; while the PLRA does alter the power of judges in correctional litigation,176 its effect on the forms of bargaining and on the bargaining positions of the parties is far greater.177 More generally, outside jails and prisons, there is a large amount of current litigation and ongoing court-ordered reform in the ar- 170. Prison Litigation Reform Act, Pub. L. No. 104-134, 110 Stat. 1321, §§ 801-810 (Apr. 26, 1996) (codified at 18 U.S.C. § 3626, 28 U.S.C. §§ 1346(b), 1915, 1915A, 42 U.S.C. §§ 1997 et seq.). 171. See 42 U.S.C. § 1997e(d); Martin v. Hadix, 119 S. Ct. 1998 (1999) (analyzing application of this provision). 172. See 18 U.S.C. § 3626(a)(1). 173. See, e.g., Austin v. Hopper, 15 F. Supp.2d 1210 (M.D. Ala. 1998), 28 F. Supp.2d 1231 (M.D. Ala. 1998) (each explaining the difference and approving a private settlement agreement in resolution of a class action). 174. See 18 U.S.C. § 3626(b)(1), (b)(3). 175. Compare, e.g., Ruiz v. Johnson, 37 F. Supp.2d 855 (S.D. Tex. 1999) (denying motion to terminate on alternative bases that the PLRA’s termination provisions are unconstitutional, and that conditions in the Texas prison system impose “current and ongoing” constitutional harm on inmates), with Parrish v. Atlanta Dept. of Corrections, 156 F.3d 1128 (11th Cir. 1998) (finding as a matter of law no “current and ongoing” constitutional violations at the Lauderdale County Jail, in Alabama). 176. For example, it removes from district judges the general authority to grant or approve “prisoner release orders,” which include certain population caps. 18 U.S.C. § 3626(a)(3). 177. See, e.g. 18 U.S.C. §§ 3626(a)(1), (g)(1), (g)(6), (g)(7) (allowing “private settlement agreements” that do not comply with new restrictions on entry of court orders in prison cases). May 1999] Judicial Policymaking 2035 eas of, for example, child welfare,178 mental health179 and mental retardation180 facilities, juvenile correctional facilities,181 public housing,182 and public school funding.183 And new areas of litigation are opening up.184 At the same time, several important private funding sources of reform litigation have recently cut back their 178. See, e.g., H.R. REP. NO. 101-395, at 193-204 app. V. (1990) (listing, inter alia, child welfare cases, by state); NATIONAL CTR. FOR YOUTH LAW, FOSTER CARE REFORM LITIGATION DOCKET (1998). 179. See Olmstead v. L.C., 119 S. Ct. 2176 (1999) (holding that unnecessary institutionalization of people with mental illness and mental retardation can be unlawful discrimination); Bazelon Ctr. for Mental Health Law, People with Disabilities and the Right to Adequate and Appropriate Public Services (Sept. 1996). 180. See Mary F. Hayden, Class-Action, Civil Rights Litigation for Institutionalized Persons with Mental Retardation and Other Developmental Disabilities: A Review, 21 MENTAL & PHYSICAL DISABILITY L. REP. 411 (1997). 181. For a very comprehensive listing, see YOUTH LAW CTR., COURT CASES AND AGENCY RULINGS ON SPECIAL EDUCATION IN JUVENILE AND ADULT CORRECTIONAL FACILITIES: A SPECIAL EDUCATION CLEARINGHOUSE (CASE SUMMARIES) (1996); NATIONAL JUVENILE DETENTION ASSN. & YOUTH LAW CTR., JUVENILE DETENTION AND TRAINING SCHOOL CROWDING: A CLEARINGHOUSE OF COURT CASES (1998); see also Michael J. Dale, Lawsuits And Public Policy: The Role of Litigation in Correcting Conditions in Juvenile Detention Centers, 32 U.S.F. L. REV. 675 (1998); MICHAEL J. DALE ET AL., REPRESENTING THE CHILD CLIENT, Chs. 1 and 2 (1998); Parent et al., Conditions of Confinement: Juvenile Detention and Corrections Facilities: Research Report 34, 39-40 (text and tbls.3-17 & 3-20) (reporting results of OJJDP, Children in Custody: Census of Public and Private Juvenile Detention, Correctional, and Shelter Facilities, 1987, 1989, and 1991: in 1987, facilities holding 11% of juveniles confined in public facilities operated under court order, including consent decrees; the number increased to 13% in 1989, and to 23% in 1991). 182. See Florence Wagman Roisman, Long Overdue: Desegregation Litigation and Next Steps to End Discrimination and Segregation in the Public Housing and Section 8 Existing Housing Programs, 4 CITYSCAPE 171, 194-96 (1999) (listing desegregation cases involving HUD). 183. See Peter Enrich, Leaving Equality Behind: New Directions in School Finance Reform, 48 VAND. L. REV. 101 (1995); Molly McUsic, The Use of Education Clauses in School Finance Reform Litigation, 28 HARV. J. ON LEGIS. 307 (1991). 184. For example, the most recent wave of school funding litigation is only a few years old. See Molly S. McUsic, The Law’s Role in the Distribution of Education, in LAW AND SCHOOL REFORM: SIX STRATEGIES FOR PROMOTING EDUCATIONAL EQUITY 88, 102-105 (Jay P. Heubert ed., 1999). And there is growing interest in injunctive litigation for remedying conditions at nursing homes, and systemic civil rights abuses in police departments. Recent nursing home litigations include: United States v. Tucker House II, Inc., No 96-1271 (E.D. Pa. consent decree entered Mar. 6, 1996); United States v. Chester Care Center, No. 98-CV139 (E.D. Pa. consent decree entered Jan. 14, 1998); United States v. City of Philadelphia, No. 98-4253 (E.D. Pa. Settlement entered Aug. 14, 1998); United States v. District of Columbia, No. 95-0948 TFH (D.D.C. stipulated order entered July 6, 1995, modified Nov. 7, 1995, further relief granted Dec. 22, 1995, further relief granted Feb. 23, 1996, dismissed May 14, 1997). Recent police cases include: Wilkins v. Maryland State Police, No. MJG-93-468 (D. Md. settlement decree entered Jan. 5, 1995); Thomas v. County of Los Angeles, No. CV 905217 (C.D. Cal. settlement Jan. 31, 1996); NAACP, Philadelphia Branch v. City of Philadelphia, No. 96-6045 (E.D. Pa. settlement entered Sept. 4, 1996); United States v. City of Pittsburgh, No. 97-0354 (W.D. Penn. consent decree entered Apr. 16, 1997); United States v. City of Steubenville, C2-97-966 (S.D. Ohio consent decree entered Sept. 3, 1997); see also 42 U.S.C. § 14141 (1994 statute allowing U.S. Attorney General to sue to “eliminate” a “pattern or practice of conduct by law enforcement officers . . . that deprives persons of [federal] rights, privileges, or immunities”); Marshall Miller, Police Brutality, 17 YALE L. & POLY. REV. 149 (1998) (comprehensive analysis of 42 U.S.C. § 14141). 2036 Michigan Law Review [Vol. 97:1994 support,185 and opponents of the litigation continue to contest the appropriateness of litigated law reform and courts as forums for regulation of governmental institutions.186 The PLRA may be just the first in a series of statutes aiming to limit reform litigation.187 Institutional reform litigation is not a judicial movement but a political practice.188 How courts began, and whether they continue, to be an arena for such litigation; how the litigation looks; and whether it succeeds or fails are functions not simply of judicial will and role, but of the goals, resources, and actions of many groups and actors, filtered through the rules of litigation. If scholars are going to be useful observers and analysts of this universe of cases, we must free ourselves from our long-bred urge to talk only about judges and open our eyes instead to the full range of participants and forces at work. 185. See Sturm, Legacy and Future, supra note 13, at 643 & n.14 (describing cutoff in funding of prison litigation by the Edna McConnell Clark Foundation, and decrease in funding of public interest litigation by the Ford Foundation). 186. Cf. JEAN STEFANCIC & RICHARD DELGADO, NO MERCY: HOW CONSERVATIVE THINK TANKS AND FOUNDATIONS CHANGED AMERICA’S SOCIAL AGENDA (1996). 187. See Judicial Improvement Act of 1999, S. 248, 106th Cong., 145 CONG. REC. S701 (daily ed. Jan. 19, 1999); Judicial Improvement Act of 1998, S. 2163, 105th Cong, 144 CONG. REC. S6188 (daily ed. June 11, 1998). This is not to say, however, that Congress will be able to shut down categories of litigation it doesn’t like. Prison reform litigation is proving unexpectedly resistant to the most ambitious congressional efforts to radically change its nature. See, e.g., William J. Taylor, Apocalypse Not: The Impact of the Prison Litigation Reform Act on Settlement in Prison Conditions Cases (unpublished manuscript) (finding that predictions that PLRA would all but eliminate settlement of prison reform cases are proving false; parties and courts continue to enter settlements and decrees). 188. See, e.g., John Denvir, Towards a Political Theory of Public Interest Litigation, 54 N.C. L. REV. 1133 (1976); Diver, Judges as Powerbrokers, supra note 12, at 45 (discussing institutional reform lawsuits as “component[s] of the continuous political bargaining process that determines the shape and content of public policy”); Xavier de Souza Briggs & Robin A. Lenhardt, After the Gavel Falls: Race, Community Politics, and Suburban Housing 3 (Mar. 1998) (unpublished manuscript) (analyzing “complex political processes that do so much, ‘after the gavel falls,’ to render courts and their agents effective or ineffective agents of change”).