法学相关资料合集（20篇英文）：在一国之内研究比较法

University of California, Los Angeles School of Law Public Law & Legal Theory Research Paper Series COMPARATIVE LAW WITHOUT LEAVING HOME: WHAT CIVIL PROCEDURE CAN TEACH CRIMINAL PROCEDURE, AND VICE VERSA DAVID A. SKLANSKY and STEPHEN C. YEAZELL Research Paper No. 05-9 This paper may be downloaded without charge at: The Social Science Research Network Electronic Paper Collection: http://ssrn.com/abstract=706601 COMPARATIVE LAW WITHOUT LEAVING HOME: WHAT CIVIL PROCEDURE CAN TEACH CRIMINAL PROCEDURE, AND VICE VERSA David A. Sklansky * & Stephen C. Yeazell ** This is a plea for comparative work in civil and criminal procedure. We do not argue here that American civil and criminal procedure should be counterpoised more frequently with their analogs overseas. Surely that is true, but both the need for and the difficulties associated with this kind of work are well understood. We argue instead for something at once more straightforward and more radical: regularly contrasting American civil and criminal procedure with each other. This is a plea for comparative work in our own backyards. It seeks to demonstrate that such work has benefits, illuminating the significance of overlooked features and providing a more stable base for reform. Civil litigation and criminal litigation in the contemporary United States occupy separate worlds. They employ different procedural rules, often before different judges in different courthouses, and with almost entirely unconnected bars, each of which views the other with an attitude verging on contempt. In law schools, civil and criminal process are taught in separate courses by scholars whose ranks rarely overlap and who read little of the work produced by their opposite numbers. 1 Many judges, of course, still hear a mixture of civil and criminal cases; that is the practice in federal court, as well as in many state and local courts, particularly outside of big cities. Aside from this point of contact, though, there is little else to suggest that the two dockets are part of the same legal system. Precisely because civil and criminal procedure differ so strikingly today, drawing comparisons between the two sets of rules can be difficult, and the utility of the exercise can seem doubtful. What, for example, is the modern, criminal equivalent of civil doctrines of standing? What are the civil analogs to probation and parole? Comparing civil and criminal procedure can seem like comparing tangerines and socket wrenches, or like suggesting that barbers and surgeons, who after all share some professional history, should exchange notes more often today. Maybe civil and criminal procedure are rarely compared today because, in fact, they are incomparable. We think otherwise. Our grounds are basically two. * Professor of Law, UCLA School of Law; Visiting Professor of Law, U.C. Berkeley School of Law. The authors are particularly indebted to Norman Abrams, Alison Anderson, Stephen Bundy, Ann Carlson, Pamela Karlan, Kenneth Karst, Daniel Klerman, John Langbein, Máximo Langer, William Rubenstein, David Shapiro, Eleanor Swift, and Charles Weisselberg for helpful criticism of earlier drafts; Charles D’Itri and the Hugh & Hazel Darling Law Library for research assistance; and the UCLA Academic Senate and the UCLA School of Law Dean’s Fund for financial support. ** David G. Price and Dallas P. Price Professor of Law, UCLA School of Law. 1 With some distinguished exceptions, of course. See, e.g., ROBERT M. COVER, OWEN M. FISS & JUDITH RESNIK, PROCEDURE 9-21, 428-53, 581-88, 606-11, 649-58, 764-76, 777-79, 788-829, 1147-67, 1573-79, 1580-83 1668-74 (1988); OWN M. FISS & JUDITH RESNIK, ADJUDICATION AND ITS ALTERNATIVES: AN INTRODUCTION TO PROCEDURE 14-26, 149-51, 645-48, 772-77(2003). 2 First, despite everything, civil and criminal procedure still have a lot in common. They are both, after all, systems of adjudicating—or otherwise resolving—disputes, and settling—or sidestepping—disagreements about historical facts. They both aim at fairness, accuracy, and efficiency—albeit in different mixtures. They share similar stages: pleading, discovery, trial or settlement, and appeal. They share the institution of the jury. They both have rules designed to protect the finality of judgments. Civil litigation, we will argue in this essay, has been essentially privatized, whereas criminal litigation is today more or less a government monopoly. But enough points of commonality remain to make systematic comparison appear worthwhile, even accepting the degree to which the criminal-civil divide has come to parallel the public-private divide. Second, there are reasons to doubt the wisdom of replicating the public-private divide in the world of legal process. Many of the complaints raised about civil litigation today can be understood as objections to how far it has been privatized—how little regard the process seems to have for the broader public interest. Conversely, some of the loudest criticisms of criminal process over the past couple of decades have been about the way the system ignores the “private” interests of victims. We certainly do not contend that civil and criminal cases have no important differences and should be treated the same. There are large differences between the two categories of cases, and comparisons between them, if drawn carelessly, can be dangerously misleading. We will pause occasionally in the following pages to point out some of those dangers. But we do think civil and criminal process can each learn things from the other—including a keener understanding of its own nature, and a healthy degree of skepticism about its own assumptions. In civil process today, important critiques are too often brushed aside with appeals to the notion that the public’s only interest is in rough fairness between the parties and in moving the docket along. The state’s role as a guarantor of fair and accurate results has become vestigial. Meanwhile criminal process has taken on some of the less attractive features of government bureaucracies, including a pronounced institutional inertia and a resistance to the involvement of interested third parties. S.F.C. Milsom has argued that a special genius of the common law, one that partly offsets its untidiness and inconsistency, is the ability to provide points of comparison and alternative solutions to problems that present themselves in more than one setting. 2 Civil and criminal procedure offer underappreciated opportunities for precisely this kind of cross-pollination. Misled by the sense that the two realms have entirely different goals, we have asked too infrequently whether they have anything to teach each other. We are more likely to contemplate borrowing practices from a foreign legal system than from the next courtroom over. There is nothing wrong with international comparison; we could use a good deal more of it. But we should not ignore the similar opportunities closer to home. 2 See S.F.C. Milsom, Reason in the Development of the Common Law, 81 LAW Q. REV. 496 (1965), reprinted in S.F.C. MILSOM, STUDIES IN THE HISTORY OF THE COMMON LAW 149 (1985). 3 Domestic comparison, in fact, may have some advantages over international comparative law. One of the notorious problems of the latter enterprise is that the systems under scrutiny are so embedded in their own respective cultures that transplants and borrowings—indeed, sometimes even meaningful comparisons—prove difficult. 3 American civil and criminal process, on the other hand, have grown from the same social and political soil. It should therefore be possible to draw comparisons with greater confidence and to contemplate borrowings with less worrying about what will be lost in "translation." 4 This is an important point, because by international standards the similarities between American criminal and civil process are probably more striking than their differences. Broadly speaking, the Continental legal tradition has distinguished even more sharply than the Anglo-American tradition between public and private, and between criminal and civil. To call for more cross-comparison between American civil and criminal process is thus to call, in a sense, for widening the divide between the American legal system, taken as a whole, and its Continental analogs. We think the trade-off is worthwhile, for reasons we will explain, but it should be recognized at the outset. In the pages that follow we seek to demonstrate what is lost when civil and criminal process are treated as incomparable, and what is gained when they are not. We first provide some historical context. Although the divide between civil and criminal process is quite old, the current contours of that divide are not. One needs to go back only a century or so to find a world in which the chasm was far narrower than it is today. Revisiting that world helps to underscore the contingency of our current thinking about civil and criminal process, and will provide the conceptual platform for the remainder of the essay. The second part of the essay describes four areas in which modern civil and criminal process address similar problems in starkly different ways—settlement, finality, discovery, and remedies for failed process. In each of these areas, we suggest, criminal process and civil process each has something to teach the other. Civil settlement practice might profitably borrow from the tradition on the criminal side that case dispositions need judicial approval; criminal practice might learn from the tradition on the civil side of involving judges in the negotiations leading up to settlement. The notorious complexity of double jeopardy law could be alleviated—and the oddity of some of its features made more apparent—by more frequent comparison with civil doctrines of former adjudication. Civil discovery might benefit from some of the limiting mechanisms of criminal discovery; criminal discovery might profitably emulate, to some degree, the symmetry of civil discovery. And lessons can be learned from comparing the law of 3 See, e.g., MIRJAN R. DAMA?KA, THE FACES OF JUSTICE AND STATE AUTHORITY (1986); John H. Langbein, The Influence of Comparative Procedure in the United States, 43 AM J. COMP. L. 545 (1995); Máximo Langer, From Legal Transplants to Legal Translations: The Globalization of Plea Bargaining and the Americanization Thesis in Criminal Procedure, 45 HARV. INT'L L.J. 1 (2004). 4 Langer, supra note 3, at 29-35. 4 malpractice on the civil side with criminal doctrines regarding effective assistance of counsel. The third and concluding part of the essay discusses two areas—evidence and professional ethics—in which civil and criminal rules are already more or less unified, and have been so for a relatively long time. In each of these fields, we argue, cross- fertilization between civil and criminal litigation has improved the rules applied in both sets of cases. Rules regarding expert testimony, for example, have benefited from the presumption that they should apply equally in civil and in criminal cases. The same is true of rules regarding the proper limits of zealous advocacy. A final caveat is in order before we begin. This is a speculative essay, not a comprehensive program for reform. We think that comparing criminal and civil procedure can generate helpful insights and highlight overlooked possibilities, and we give examples of such insights and possibilities in the pages that follow. We do not claim that every possibility we identify deserves to be pursued, nor that we have identified all areas that might profitably be pursued. What we do claim, and what we hope our examples help to show, is that American civil and criminal process have things to teach each other. I. HOW WE GOT HERE A. When All Process Was Private In civil justice today the state sometimes seems to have no role to play; in criminal justice the citizen sometimes seems to have no role to play. One branch of law seems to have become the exclusive province of private disputes between peers. The other branch of law has become a state monopoly, and many aspects of procedural design reflect the tension between the desire to minimize crime and public expense at the same time and the special concerns that arise when the state deploys its power against the citizen. The criminal-civil divide and the public-private divide both appear to be such fundamental features of the law that it is easy to forget for how short a time they have been conjoined in this manner. As recently as the nineteenth century—indeed, well into the twentieth century—civil and criminal proceedings were, in essence, alternative ways for aggrieved victims of wrongs to enlist the adjudicative machinery of the state in seeking redress. For most of the nineteenth century, moreover, these two ancient systems of redress seemed to be converging rather than diverging. The mechanism was cross- pollination, and the most important pollinators were lawyers. Lawyers themselves were an innovation that criminal practice borrowed from civil litigation. Opposing counsel were common in civil cases by the early eighteenth century. Then they were imported in stages into criminal trials, often via statutory and constitutional provisions that explicitly 5 allowed for counsel to appear in criminal prosecutions “as in civil cases.” 5 By the nineteenth century lawyers were common in criminal as well as civil cases, and often the same lawyers appeared in both kinds of proceedings. 6 Once lawyers had feet on both sides of the criminal-civil divide, they began to carry ideas back and forth between the two realms, and the divide started to shrink. Take, for example, the aggressive use of objections to shape and limit testimony—the foundation of the modern law of evidence. This practice appears to have arisen first in criminal trials and then to have been imported into civil cases by lawyers who appeared on both sides of the docket. 7 Because the practice crossed the procedural divide so quickly, modern evidence law, when it began to solidify in the late nineteenth century, drew few distinctions between civil and criminal cases—a fact to which we will return later in this essay. Today, of course, the civil bar and the criminal bar are largely separate, and the great majority of criminal lawyers are employed by the government—regardless of which side they represent. In the nineteenth century, by contrast, the private bar was more unified, and it handled the bulk of litigation, both civil and criminal. Criminal prosecution, let alone criminal defense, had not yet become a government monopoly. Before the Civil War, in fact, criminal prosecution was not even a government specialty. The system relied on private prosecution, with state officials involved only in a minority of cases. So overwhelming was the case load for such officials that citizens with an interest in seeing justice done often hired private prosecutors to marshal evidence and to prosecute the case. Understandably, such citizens were often the victims, or family members of the victims, of the crimes alleged. The real or suspected motive of vengeance provided arguments for the defendant, and, more occasionally, criticism from appellate courts. But private prosecution flourished well into the twentieth century, in large part because few public prosecutors were funded and staffed well enough to bring charges in the vast run of cases. 8 This practice meant that for many citizens the functions of criminal and civil justice were barely distinguishable. Both served as means for citizens to pursue 5 Eben Moglen, Taking the Fifth: Reconsidering the Origins of the Constitutional Privilege Against Self-Incrimination, 92 MICH. L. REV. 1086, 1126 (1994) (quoting N.Y. CONST. (1777)). 6 Stephan Landsman, The Rise of the Contentious Spirit: Adversary Procedure in Eighteenth Century England, 75 CORNELL L. REV. 497 (1990); John H. Langbein, The Criminal Trial Before Lawyers, 45 U. CHI. L. REV. 263 (1978); John H. Langbein, Historical Foundations of the Law of Evidence: A View from the Ryder Sources, 96 COLUM. L. REV. 1168 (1996) [hereinafter Langbein, Historical Foundations of the Law of Evidence]. 7 See T.P. Gallanis, The Rise of Modern Evidence Law, 84 IOWA L. REV. 499 (1999). It is not entirely clear why heavy reliance on evidentiary objections to testimony appeared first in criminal trials. Langbein speculates it had to do with the fact criminal trials involved more witnesses and fewer documents. See Langbein, Historical Foundations of the Law of Evidence, supra note 6, at 1201-02. Gallanis proposes instead that lawyers in criminal cases were forced to turn their attention to objections because “prisoner’s counsel was allowed to do little more than cross-examine the victim and the other witnesses supporting the charge”; he could not, in particular, address the jury. See Gallanis, supra, at 545. 8 Robert M. Ireland, Privately Funded Prosecution of Crime in Nineteenth-Century United States, 39 AM. J. LEGAL HIST. 43 (1995). 6 grievances about the behavior of others. In both processes the vigor with which the complaints were pursued and the timing and means by which the disputes were terminated lay in the hands of instigators. Allen Steinberg’s study of private prosecution in Philadelphia before the Civil War makes the linkage clear: During the first half of the nineteenth century . . . an elaborate system of criminal justice typified by private prosecution and dominated by the city’s alderman became entrenched in Philadelphia. . . . [L]aw enforcement became the responsibility of a community’s residents, and the mechanism they used was the one remaining from the eighteenth-century town—the private criminal suit instituted in the office of the neighborhood alderman. . . . Aldermen had no power to implement the final disposition of these cases, but the decision about how far along the criminal justice process a case would go rested primarily with these officials and the private parties. . . . 9 “There was little role for a public prosecutor in this system of law enforcement. The state had little direct interest in it, and when the state was involved,” no major friction existed between the citizenry and the alderman. “Although aldermen were officers of the state, their real dependence was on the private citizens who provided their fees.” 10 In such a regime, citizens and lawyers alike might perceive relatively little distinction between instituting civil and criminal litigation: in either, the complainant would initiate the case, would pursue it, and would end it when and if he reached a satisfactory settlement with his adversary. So deeply ingrained were these assumptions that even at mid-century people objected to the public prosecutor’s playing any role in prosecution, on the grounds that to do so would be to favor one citizen (the complaining witness, as we now might call her) over another. 11 Understanding this situation gives us more insight into an institution that now seems questionably useful, the grand jury. Today everyone seems to agree that a minimally competent prosecutor can get a grand jury to “indict a ham sandwich.” 12 Not 9 Allen Steinberg, From Private Prosecution to Plea Bargaining: Criminal Prosecution, the District Attorney, and American Legal History, 30 CRIME & DELINQ. 568, 572-73 (1984). 10 See id. at 575. 11 See id. at 577. 12 See, e.g., David Margolick, Law Professor to Administer Courts in State, N.Y. TIMES, Feb. 1, 1985, at B2 (quoting Sol Wachtler). Remarkably, even the United States Department of Justice now seems to have endorsed the view that the grand jury serves as little more than a rubber stamp. The endorsement came as part of the Department's defense of new powers granted to the FBI in the wake of September 11, 2001. Among those powers is the ability to obtain court orders requiring the production of documents based solely on an application from the FBI stating that "the records concerned are sought for an authorized investigation … to obtain foreign intelligence information not concerning a United States person or to protect against international terrorism or clandestine intelligence activities." USA PATRIOT Act of 2001 § 215, 50 U.S.C. § 1861 (2004). In defending this provision, the Justice Department has argued that since the orders must be "approved by a federal judge," they are "more difficult to obtain than ordinary grand-jury subpoenas, which can require the production of the very same records but without prior judicial approval." 7 so in the nineteenth century. Then, as in Blackstone’s day, the grand jury played a substantial role, receiving the “suit of any private prosecutor” and deciding whether it warranted further proceedings. 13 The grand jury screened the trivial or the unfounded complaints from those that had substance. Today, the grand jury’s role makes little sense because we assume that a state official has already screened cases for merit and substance. In a world of private prosecution, the function was both sensible and essential: in the two decades bracketing the Civil War, Philadelphia grand juries dismissed almost two-thirds of all charges brought before them, a statistic unthinkable today. 14 We live in a world so different it is hard to capture the operating assumptions of a century ago. The bureaucratization of criminal justice created executive officers who now do the work of apprehension, investigation, and screening that once took place principally within the judicial system. Apprehension and investigation are now the jobs of professional law enforcement—a sector that did not exist before the middle of the nineteenth century and that became fully established only in the twentieth century. 15 Screening is the work of law enforcement and professional prosecutors. Even criminal defense is now a responsibility largely assumed by the state—albeit with varying levels of zeal. By 1963, when the Supreme Court ruled that states were constitutionally obligated to give lawyers to indigent felony defendants, most states were already doing so: the effect of Gideon was to require the small number of holdouts to meet the by-then common practice. 16 Because the vast majority of criminal defendants are indigent—the figure is over 80% in state felony cases 17 —in most criminal trials today all of the participants other than the defendant, the jury, and some of the witnesses are on the government payroll. The question being debated, moreover, is whether the government should forcibly house and employ, at prison wages, the defendant as well. In a powerful sense the state now “owns” the entire apparatus of criminal investigation and adjudication. The state owns it because the state pays for it. In twenty-first-century Los James Comey, Deputy Attorney General, Letter to the Editor, Rights and the Patriot Act, N.Y. TIMES, Apr. 28, 2004, at A20. 13 4 WILLIAM BLACKSTONE, COMMENTARIES ON THE LAWS OF ENGLAND *300. 14 See Steinberg, supra note 9, at 574. As we discuss later in this essay, the grand jury today does play a role largely unrelated to screening: it allows prosecutors to question witnesses under oath before charges are brought. It therefore serves as a partial substitute for the general unavailability of depositions in criminal cases. But it provides this substitute only for prosecutors, and without the check provided by the presence of opposing counsel during questioning. Once the grand jury is understood as essentially a discovery device, not a screening device, these stark differences from the rules governing depositions on the civil side become harder to justify. See notes 94-108 and accompanying text, infra. 15 See, e.g., David A. Sklansky, The Private Police, 46 UCLA L. REV. 1165, 1193-1217 (1999). 16 Gideon v. Wainwright, 372 U.S. 335 (1963). 17 See CAROLINE WOLF HARLOW, U.S. DEP’T OF JUSTICE, BUREAU OF JUSTICE STATISTICS SPECIAL REPORT: DEFENSE COUNSEL IN CRIMINAL CASES 1 (2000); STEVEN K. SMITH & CAROL J. DEFRANCES, U.S. DEP’T OF JUSTICE, BUREAU OF JUSTICE STATISTICS SELECTED FINDINGS: INDIGENT DEFENSE 1, 4 (1996). 8 Angeles, for example, roughly a quarter of the city and county budgets is spent on criminal process—excluding the courts. 18 B. Police as a Public Investment This story of private enterprise and public reaction had one important exception in the nineteenth century, an exception that prepared the way for our existing dual systems. The emergence of public police forces helped shape the conditions for modern legal practice—on the civil as well as on the criminal side. In a system that depended on evidence of disputed facts, any mechanism that collected these facts and made it easier for the parties to present them in court was important. In the latter nineteenth century English and American societies acquired such a system—professional police—though the chief motivation lay not in the desire to assist courts but in the wish to keep public order in growing cities. This is not the place for a history of policing. But the growth of professional law enforcement explains part of the divergence of the civil and criminal systems. Well into the nineteenth century, in both England and America, the apprehension of offenders and the prosecution of criminal cases was a private affair, in which government participated only by offering rewards. Modern police departments, which date from the 1829 creation of the London Metropolitan Police, specialized at first almost entirely in patrol. They replaced the medieval institutions of the constable, the night watch, and the hue and cry, but they left the business of detection to the private sector— initially, to the informal network of informers and "thief-takers" inherited from the 1700s, and later, particularly in America, to the private detective industry pioneered by Alan Pinkerton. 19 There were natural synergies, though, between patrol work and detection, and by the late nineteenth century urban police forces on both sides of the Atlantic were developing significant sidelines in investigation. The detective branches of public police departments greatly expanded in the early twentieth century, and they began to absorb the techniques and ambitions of companies like the Pinkerton agency—a process accelerated by the newly-formed Federal Bureau of Investigation, both directly and through the example that it set. Spurred in part by the FBI, American law enforcement at all levels of government gradually came to see its most important mission as controlling crime, not 18 CITY OF LOS ANGELES 2002-03 BUDGET 16 (25% for "crime control"—excluding prosecution, which is a county responsibility); COUNTY OF LOS ANGELES ANNUAL REPORT 2002-03, at 16-25 (20% for "public protection"—including offices of district attorney, public defender, sheriff, and coroner, as well as some firefighting services). 19 See, e.g., Sklansky, supra note 15, at 1195-1217. Charles Dickens gave us a memorable remnant of this world in Bleak House, where one of the plot lines revolves around the work of a privately retained detective, whose devotion to truth-seeking brings down the wife of the aristocrat who retained him. 9 simply maintaining order. And controlling crime meant not just catching criminals, but convicting them in court. 20 By the 1920s and 1930s, the police still spent most of their time keeping order. Even when that task required an arrest (which it usually did not), and when the arrest led to a court proceeding (which was also not the norm), little evidence gathering was required: the officer’s testimony would usually suffice. But in those cases that did require investigation and gathering of evidence, the police were available to supply and supplement the efforts of professional prosecutors. Moreover, whether they operated with or without warrants, the police offered an American version of compelled evidence gathering: they could search records, interrogate unwilling witnesses, collect physical evidence, and compel access to real and personal property. The fruits of these efforts fed the prosecutorial arm, which could then present the evidence thus gathered in court, in support of the charges filed. If we add to this system the important additional ingredients of constitutional constraints on evidence-gathering and public support of the defense bar, it is still the system we use today. C. Law Reform and the Parting of the Ways If the nineteenth century marked a period of innovation in criminal process, the twentieth century saw enormous change on the civil side. That change affected both formal procedural rules and the organization of the bar. Together, they reshaped civil practice over the course of a century, making it look almost unrecognizable either from the perspective of the civil bar at the start of the twentieth century or from the vantage point of the criminal practitioner at the start of the twenty-first. While criminal process was becoming a branch of government law, civil practice was heading in the opposite direction, toward entrepreneurship, initiative, and an understanding of law as a business. Robert Gordon has told the story of the early growth of the law firm, a practice development paralleling the growth of these firms’ business clients. 21 As interesting a story, at the other end of the private-bar practice spectrum, has been the growth of the “plaintiffs’ firm.” Long a despised poor relation of the Wall Street practice, 22 the plaintiffs’ firm has emerged as a successful business in the years following the Second World War. 23 Taking advantage both of the contingent fee and of deregulation of advertising for professional services, 24 this part of the bar created small 20 See, e.g., Sklansky, supra note 14, at 1216-17. 21 Robert W. Gordon, The Ideal and the Actual in the Law: Fantasies and Practices of New York City Lawyers, 1870-1910, in THE NEW HIGH PRIESTS: LAWYERS IN POST-CIVIL WAR AMERICA 51, 59-62 (Gerard W. Gawalt ed., 1984). 22 See JEROLD S. AUERBACH, UNEQUAL JUSTICE: LAWYERS AND SOCIAL CHANGE IN MODERN AMERICA 40-50 (1976); JEROME E. CARLIN, LAWYERS ON THEIR OWN: A STUDY OF INDIVIDUAL PRACTITIONERS IN CHICAGO (1962). 23 Stephen C. Yeazell, Refinancing Civil Litigation, 51 DEPAUL L. REV. 183 (2001). 24 See Bates v. State Bar of Arizona, 433 U.S. 557 (1980); Shapero v. Kentucky Bar Assoc., 486 U.S. 466 (1988). 10 firms whose members’ hourly compensation essentially matched that of their insurer-paid defense counterparts. 25 As the proportion of solo practitioners declined from a majority of the American bar, the resulting firms—mostly small—constituted a new, better- capitalized group of legal entrepreneurs. Freed to compete, they did so fiercely, sometimes without dignity, but with great effectiveness. The idea that major political figures would have devoted time to denouncing the excessive power of the plaintiffs’ bar would have struck most observers as ludicrous a few decades ago; now it appears to be a part of at least one political party’s platform. 26 For our purposes the significance lies not in whether the critique of the plaintiffs’ bar is well-taken, but in what it tells us about the entrepreneurial spirit that has developed within the civil litigation bar. The public side of the bar paid (quite literally) tribute to this development at the end of the 1990s, as state attorneys general hired—usually on a contingent fee basis—plaintiffs’-side law firms to pursue, on the states’ behalf, claims against tobacco companies. 27 The decision to hire these firms was probably an intelligent move: the state attorney-general staffs, though they do civil litigation, probably were not equipped to handle the challenges presented by the tobacco litigation. The plaintiffs’ firms, forged in decades of entrepreneurial activity, were. The second great change in civil practice came in the middle third of the twentieth century, as first federal, then state courts adopted discovery systems. Because these are such a familiar part of the procedural landscape, it is easy to miss their historical significance today. They put into the hands of private lawyers the power to compel sworn testimony and to require the other party and unaffiliated witnesses to disclose information and documents, allow access to relevant real evidence, and more. There are limits to this power, but from a comparative perspective it is broader and deeper than the powers exercised by private lawyers in any other legal system. 28 In many respects the lawyers representing the parties in civil actions have in their hands powers remarkably like those exercised by police investigators in criminal matters. Unlike criminal 25 HERBERT M. KRITZER, RISKS, REPUTATIONS, AND REWARDS: CONTINGENCY FEE LEGAL PRACTICE IN THE UNITED STATES (2004). 26 In 2003 President George W. Bush made twelve speeches or public statements in which he identified “trial lawyers” as a significant issue of public policy. See 39 WEEKLY COMPILATION PRESIDENTIAL DOCS. 73, 117, 109, 289, 321, 776, 1142, 1253, 1361, 1459, 1523, 1767 (2003); Shailagh Murray, Conservative Appeal: Trial Lawyers Lobby Discovers Unlikely Friends: Republicans, WALL ST. J., July 8, 2004, at A1. And that was before a former plaintiffs' lawyer, John Edwards, became the Democratic candidate for Vice President in 2004. 27 Robert Gehrke, Bill Would Place Caps on Tobacco Lawyers, Associated Press Online, Apr. 10, 2003, available in Westlaw, 2003 WL 18223861 (reporting legislation aimed at requiring lawyers hired by states to sue tobacco makers to reduce fees); Abdon M. Pallasch, Tobacco Lawyers Settle for $188M: Illinois Initially Owed Them Nearly $1 Billion After Suit, CHI. SUN TIMES, Dec. 18, 2002, at 14 (reporting on state AG’s refusal to honor original fee agreement and subsequent settlement of lawyers’ claim); Fees Thrown Out for Anti-Tobacco Lawyers, SAN DIEGO UNION-TRIB., September 8, 2002, at A8 (reporting trial court’s rejection of arbitrators’ fee award to lawyers representing California in suit against tobacco companies). 28 KUO CHANG HUANG, INTRODUCING DISCOVERY INTO CIVIL LAW 37-103 (2003); Geoffrey C. Hazard, Jr., From Whom No Secrets Are Hid, 76 TEXAS L. REV. 1665 (1998). 11 investigation, however, the cost of civil discovery is borne by the parties separately, rather than by society as a whole. This practice distinguishes American civil procedure not only from American criminal process, but also from private litigation in civil law countries, where judges typically decide whether and how deeply to probe the disputed facts. 29 The American system of civil discovery strives to align procedural opportunities with party incentives; much of the debate about the system revolves around whether the incentives lead to spending that is wasteful seen from a social perspective. For our purposes the focus lies on the privatization of the system of fact investigation. Where the public generally bore the cost of criminal fact investigation, the parties (and to the extent they were inconvenienced, non-party witnesses) bore the costs of civil discovery. That divergence had numerous collateral effects: the size of the U.S. judiciary stayed small relative to the size of many civil law nations; many in the civil bar who once tried cases became “litigators,” whose efforts focused on the pretrial phase and on the ensuing settlement; many of the disputes giving rise to civil litigation vanished from public sight because they occurred in lawyer’s offices and on motion days rather than in trial. Debates about these consequences, singly and together, rage. 30 Our focus remains on how the divergence allowed us to lose sight of the respects in which the two systems might learn from each other. These two paths—assimilation of the criminal bar largely into government service at the same time that civil practitioners were looking ever more like entrepreneurs—gave different feels to discussions of procedural change in the two areas. Criminal law today is practiced almost exclusively by government bureaucracies, while private practitioners dominate civil law, and relatively few lawyers cross the boundary between the two domains. This separation has made the two bars and the academics who study them rather unlikely to think about “procedure” as a unified field. It has made cross- comparisons and borrowings hard to imagine, let alone to carry out. And it has given different casts to reform efforts on the two sides of the procedural divide. Bureaucracies at their best are faithful, reliable, honest, efficient, and public-spirited; they are unlikely to be innovative. Entrepreneurs at their best are forward-thinking and inventive, but they tend to focus overwhelmingly on profit as a measure of success, and they are apt to be dismissive of what they can learn from government. Partly as a consequence of these different mindsets and partly because criminal process had its great innovative era in the nineteenth century, the changes in civil procedure in the twentieth century dwarf those in criminal procedure. Civil practice saw the growth of private discovery, the asymptotic decline in trials, the elaboration of joinder devices (including but not limited to the class action), the diminished significance of pleading, and the judicial blessing of a number of forms of private adjudication. Together, these changes have transformed civil practice. The changes on the criminal side have been more modest: criminal procedure has not fundamentally altered its shape in a century, except when constitutionally required to do so. That is, of course, a large 29 See HUANG, supra note 28, at 46-47. 30 See, e.g., Stephen C. Yeazell, Getting What We Asked For, Getting What We Paid For, and Not Liking What We Got: The Vanishing Civil Trial, 1 J. EMPIRICAL LEGAL STUD. 943 (2004). 12 exception. The constraints imposed on criminal procedure by interpretations of the Fourth, Fifth, Sixth, and, to a lesser extent, Eighth Amendments to the federal Constitution, have been important, real, and remarkable. But they have been won and lost in the courts, not in drafting committees or law reform commissions. And reliance on constitutional adjudication as a path to procedural change has certain stultifying effects, effects that grow especially pronounced when—as with criminal justice—public funds are on the line. 31 Not all reforms are advances. Criminal practitioners may have been right to resist some of the innovations pioneered in civil procedure. Some of those changes may have been changes for the worse, and even some reforms that were welcome on the civil side may not have made sense on the criminal side. In other cases, though, glances across the procedural divide may be a useful corrective to the institutional conservatism of criminal procedure. The civil bar, for its part, may have something to gain from the reminder that civil justice is more than an arena for earning profits. Some of the procedures on the criminal side of the divide may make little sense in civil cases; some of those procedures may in fact make little sense even in criminal cases. Elsewhere, though, criminal process may have more to teach civil process than is commonly imagined. We believe, in fact, that each side of the procedural divide has things to teach the other. We turn now to some of the possible lessons. II. NEGLECTED COMPARISONS ACROSS THE PROCEDURAL DIVIDE A. Settlement Most civil and most criminal cases end with a consensual resolution. 32 In civil litigation we call such resolutions settlements; on the criminal docket, they go under the 31 See David A. Sklansky, Quasi-Affirmative Rights in Constitutional Criminal Procedure, 88 VA. L. REV. 1229, 1279-85 (2002). 32 It is remarkably difficult to get robust statistics on the rate of civil or criminal consensual settlement. The most common mistake is to subtract the percentage of trials from the percentage of filings and thereby arrive at a figure of over 95% of settlements or plea bargains. That is an error because it disregards the circumstance that many civil cases end with dispositive adjudication before trial. Those cases end because of a judicial decision that concludes the case, not because the parties decide to control the risks of adjudication with an agreement. See Stephen C. Yeazell, The Misunderstood Consequences of Modern Civil Process, 1994 WIS. L. REV. 631 [hereinafter Yeazell, Misunderstood Consequences]. An analogous calculation on the criminal side would put consensual agreements at approximately 95%, but, again, this disregards all dispositions that were unilaterally terminated or that were otherwise dismissed. See U.S. DEP’T OF JUSTICE, BUREAU OF JUSTICE STATISTICS: COMPENDIUM OF FEDERAL JUSTICE STATISTICS, 2001, at 55 (2003). Our own estimates, for which we claim no great statistical sophistication, put both the rate of plea bargain and the rate of civil settlement at about 60% of filed cases—very high but not overwhelming. On the criminal side, for example, the Federal Bureau of Justice Statistics study tracked felony filings in the nation’s seventy-five largest urban counties during a recent year; of the 88% of the cases adjudicated by the end of the year, 61% resulted in guilty pleas. See U.S. DEP’T OF JUSTICE, BUREAU OF JUSTICE STATISTICS STATE COURT PROCESSING STATISTICS: FELONY DEFENDANTS IN LARGE URBAN COUNTIES, 1998, at 24 (2001). Regarding the comparable figures on the civil side, see Yeazell, supra, at 636-71. 13 name of plea bargains. 33 On the civil side settlements are a generally accepted and widely encouraged practice. The practice has some prominent academic critics, but by and large it remains broadly accepted by the bar, judiciary, and clients. 34 By contrast, the plea bargain, though equally dominant as a form of resolution, is widely criticized and often attacked as fundamentally illegitimate both by the public and many segments of the profession. The title of an essay by a prominent criminal law scholar, Plea Bargaining as Disaster, captures a widely held view. 35 Plea bargaining has its defenders, but even they rarely praise the practice as anything better than a necessary evil. The debate over plea bargaining is largely a debate about whether the evil is truly necessary. 36 The truth, of course, is that both plea bargaining and the system of civil settlement are likely to be with us for some time to come; the real question is what form they should take. And here the differences between the two systems are, if anything, even more striking. Civil settlement and plea bargaining each seem deeply embedded in their respective legal cultures. As a result, civil settlement occurs in an atmosphere of entrepreneurial creativity untrammeled by--and sometimes at odds with--the regularity implied by the rule of law. Criminal pleas, in contrast, occur in an atmosphere that, at least formally, values consistency, transparency, and regularity. We examine each system in turn, and then step back to draw some comparisons and to suggest some lessons each system could teach the other. First, the civil side. Settlement here is a private, largely invisible, contractual phenomenon; the judge plays a role only as a potential facilitator of private agreement. Civil parties settle at any point in the dispute, from before filing to after appeal, simply by agreeing to do so. Moreover, the terms to which they may agree are, with few exceptions, entirely a matter between the parties: the court “stands indifferent when the 33 It is another striking example of the disconnection between these procedural realms that “bargain,” the word one might associate with the commercial transactions that are a staple of civil litigation, applies instead to the criminal resolutions. 34 See, e.g., Samuel R. Gross & Kent D. Syverud, Don’t Try: Civil Jury Verdicts in a System Geared to Settlement, 44 UCLA L. REV. 1, 2-4 (1996); Judith Resnik, Whose Judgment? Vacating Judgments, Preferences for Settlement, and the Role of Adjudication at the Close of the Twentieth Century, 41 UCLA L. REV. 1471, 1474 (1994) (“On the one hand, on almost every occasion, judges and lawyers extol the virtues of settlement and the desirability of enabling private accommodations among litigants to end disputes without state-authored adjudication.”). Tellingly, at a symposium convened in December 2003 by the Section on Litigation of the American Bar Association (and attended by one of the authors), debate focused not on whether settlement was iniquitous but on particular settlement practices (e.g., should the magistrate judge conducting settlement conferences report to the trial judge on the parties’ positions). 35 Stephen J. Schulhofer, Plea Bargaining as Disaster, 101 YALE. L.J. 1979 (1992). 36 See, e.g., Albert W. Alschuler, Implementing the Defendant's Right to Trial: Alternatives to the Plea Bargaining System, 50 U. CHI. L. REV. 931, 1048 (1983); Stephanos Bibas, Plea Bargaining Outside the Shadow of Trial, 117 HARV. L. REV. 2463, 2544-45 (2004); Arthur Rosett & David A. Sklansky, Plea Bargaining, in THE OXFORD COMPANION TO AMERICAN LAW 604 (Kermit L. Hall ed., 2002); Stephen J. Schulhofer, Is Plea Bargaining Inevitable?, 97 HARV. L. REV. 1037, 1107 (1984); Ronald Wright & Marc Miller, The Screening/Bargaining Tradeoff, 55 STAN. L. REV. 29 (2002). For a rare defense of plea bargaining as actually desirable, see Frank H. Easterbrook, Criminal Procedure as a Market System, 12 J. LEGAL STUD. 289 (1983). 14 parties, for whatever reason commends itself to them, choose to settle a litigation.” 37 The overwhelming majority of settlements require only that the parties agree; the court is told only that the plaintiff is dismissing the case. Only in relatively exotic situations—settlements involving minors or absentee members of a class action, consent decrees involving future enforcement by the court— need a judge consider or approve the substantive terms of the agreement. 38 And even consent decrees receive meaningful review only in those exceptional suits thought to “affect[] the public interest” 39 —e.g., class actions, shareholder derivative suits, bankruptcy claims, environmental enforcement actions, etc. 40 Because “ordinary” civil cases 41 are thought to implicate “only private interests,” 42 the court in such a case “makes only the minimal determination of whether the agreement is appropriate to be accorded the status of a judicially enforceable decree.” 43 Three decades ago, with public interest litigation and class action suits both appearing to be on the upswing, Abram Chayes could claim plausibly, if controversially, that civil suits affecting “only private interests” were a dwindling species. 44 But the notion proved remarkably durable that civil litigation is first and foremost an arena for the adjustment of private disputes—disputes in which the public plays no role aside from providing a neutral arbitrator. The durability of that notion explains some of the strong political and judicial opposition that emerged to the kind of “public law litigation” that Chayes celebrated. It explains, too, why an “active role for the trial court in approving the adequacy of a settlement” remains today “the exceptional situation, not the general rule.” 45 37 Heddendorf v. Goldfine, 167 F. Supp. 915, 926 (D.C. Mass. 1958) (Wyzanski, J.); accord, e.g., United States v. City of Miami, 614 F.2d 1322, 1330 (5th Cir. 1980). 38 See, e.g., FED. R. CIV. P. 23(e) (requiring court approval, after notice, of a settlement involving a class); CAL. CIV. PROC. CODE §372(a) (requiring judicial approval of settlement of claims involving minors) (West 2005). 39 Adams v. Bell, 711 F.2d 161, 170 n.40 (D.C. Cir. 1983) (en banc). 40 See, e.g., id.; Janus Films, Inc. v. Miller, 801 F.2d 578, 582 (2d Cir. 1986); United States v. Telluride Co., 849 F. Supp. 1400, 1402 (D. Colo. 1994); Friends of the Earth v. Archer Daniels Midland Co., 780 F. Supp. 95, 99 (N.D.N.Y. 1992). 41 City of Miami, 614 F. Supp. at 1330. 42 Janus Films, 801 F.2d at 582. 43 Id. 44 Abram Chayes, The Role of the Judge in Public Law Litigation, 89 HARV. L. REV. 1281 (1976). 45 City of Miami, 614 F. Supp. at 1331. If “[t]here is no indication of reduction in the volume or importance of Chayesian judicial activity,” Charles F. Sabel & William H. Simon, Destabilization Rights: How Public Law Litigation Succeeds, 117 HARV. L. REV. 1015, 1018-19 (2004), neither is there any indication of a significant increase—or of greater willingness to see public law litigation as the norm rather than the exception in civil process. 15 There is a public debate and some legislation on the extent of confidentiality in settlement agreements. 46 But even here the questions are narrow: should confidentiality be available in cases likely to recur or where the claim involves governmental wrongdoing or threats to public safety. 47 There is no serious dispute about the propriety of keeping confidential the amount of the settlement, and no suggestion that judges should approve their substantive fairness. A recent change in the Federal Rules of Civil Procedure nicely captures the image of civil litigation as a private activity. Although civil discovery is part of the judicial process, parties are explicitly forbidden from filing discovery documents with the court unless they are subsequently used in a motion or at trial. 48 Having privatized the civil fact investigation process, we are generally happy to have consensual outcomes be as private as the discovery that preceded it. Where civil settlements are overwhelmingly private, criminal settlements are always public. 49 Where civil settlements require, except in unusual cases, only the consent of the parties, criminal pleas always require the participation and approval of the judge. By comparison to civil settlements, criminal plea bargains are not “bargains” at all, but legislative proposals passed by two “houses” and presented to a third official who can approve or veto. As the analogy suggests, we treat pleas less like private agreements than like public acts of state, requiring the transparency and separation of powers we insist should accompany accountable government. Procedural rules reflect all these features. Two federal rules treat criminal and civil agreements: Rule 11 of the Rules of Criminal Procedure and Rule 16 of the Rules of Civil Procedure. Comparing their features enables one concretely to understand the fundamentally different conceptions surrounding analogous terminations of litigation. The civil provisions assume that counsel will adequately inform and protect clients; a malpractice suit is the remedy if the lawyer fails in this regard. 50 Many provisions of the 46 Richard A. Rosen, Confidentiality Agreements Become Increasingly Elusive: Several States Have Limited the Availability of Protective Orders, and Judges Are Now More Skeptical About Issuing Such Orders, NAT’L L.J., July 20, 1998, at B7; James E. Rooks, Jr., Let the Sun Shine In: ‘Sunshine’ Laws Do Not ‘Chill’ Settlements, Say Advocates of Open Courts, TRIAL MAG., June 2003, at 18. 47 See, e.g., Jack H. Friedenthal, Secrecy and the Civil Justice System in Civil Litigation: Discovery and Party Agreements, 9 J.L. & POL’Y 67 (2000); TEX. R. CIV. P. 76A, SEALING COURT RECORDS; Sunshine in Litigation; Concealment of Public Hazards Prohibited, FLA. STAT. ANN. § 69.081 (West 2004). 48 FED. R. CIV. P. 5(d) (“[D]isclosures . . . and the following discovery requests and responses must not be filed until they are used in the proceeding or the court order filing: (i) depositions, (ii) interrogatories, (iii) requests for documents or to permit entry upon land, and (iv) requests for admission.”). This list includes every widely used discovery device. The motive for this Rule change was relatively innocuous: were all discovery routinely filed the storage space required would be enormous. 49 At least eventually. To protect ongoing investigations, settlements with cooperating defendants may be sealed pending other indictments. 50 And one finds a professional literature regarding malpractice pitfalls in settlements. See, e.g., Jack L. Meligan, Using Plaintiff Structured Settlement Specialists, TRIAL MAG., Apr. 1999; Michael G. Daigneault & Jack Marshall, The Lawyer’s Role in the Structured Settlement Era: Duties, Rules, and Perils, FED. LAW., Oct. 1996, at 10. 16 criminal rule implicitly make the opposite assumption, requiring the judge repeatedly to inform the defendant of matters well known to any competent criminal lawyer. The civil rules cast the judge in the role—should she wish to assume it—of participant and even persuader, convening the parties and suggesting various possibilities of settlement and creative approaches to bridging disagreements between the parties. 51 The criminal rule forbids the judge from participating in such conversations, although (or because) she must bless its result when it is presented to her. 52 The initial stage of criminal agreements, then, is a matter entirely for the lawyers and the defendant; the judge is not to taint them by her presence until the moment comes when he may accept or reject the parties’ agreement. Civil settlements, on the other hand, are matters that warrant a judge’s presence, and perhaps his active encouragement, but not his assessment of their justice. 53 Moreover, nothing in the civil rule suggests any limits to the parties’ or judge’s creativity in reaching consensus, and fertile minds have created a wide range of settlement devices. Structured settlements, sliding scale settlements, high-low agreements, 54 and cede-backs of punitive damages 55 testify to the creativity and breadth 51 See, e.g., Judith Resnik, Mediating Preferences: Litigant Preferences for Process and Judicial Preferences for Settlement, 2002 J. DISP. RESOL. 155, 156 n.5 (citing Hubert L. Will, Judicial Responsibility for the Disposition of Litigation, in Proceedings of Seminar for Newly Appointed United States District Judges, 75 F.R.D. 117 (1976)); Judith Resnik, Trial As Error, Jurisdiction As Injury: Transforming the Meaning of Article III, 113 HARV. L. REV. 924, 943-949 (2000). 52 FED. R. CRIM. P. 11(e)(1) (instructing that “[t]he court shall not participate in any such discussions"). Most state rules mirror the federal rule in this respect, but a growing minority of states allow and even encourage judges to participate in plea negotiations. See Wright & Miller, supra note 36, at 89 & n.224. 53 See supra notes 37-43 and accompanying text. 54 Structured settlements provide for payouts over an extended period. For example, a tort defendant, instead of settling for a large, single cash payment, might purchase an annuity policy guaranteeing defendant lifetime payments of $5,000 a month. Such settlements enable minors or financially unsophisticated parties to assure themselves of lifetime support. See, e.g., Barbara D. Goldberg & Kenneth Mauro, Utilizing Structured Settlements, 658 PLI/LIT 31, 40-42 (2001). Sliding scale settlements, sometime known as Mary Carter agreements, make the amount of payment by one party contingent on the plaintiff’s recovery against remaining defendants. They assure the plaintiff some compensation, provide bridge financing for litigation against remaining defendants, and ameliorate the effects of joint and several liability regimes. Such agreements have been the subject both of debate and regulation. See, e.g., Dosdourian v. Carsten, 624 So.2d 241 (Fla. 1993); Lisa Bernstein & Daniel Klerman, An Economic Analysis Of Mary Carter Settlement Agreements, 83 GEO. L.J. 2215 (1995). See also CAL. CIV. PROC. CODE §§ 877, 877.5, 877.6 (West 2005). High-low agreements put floors and ceilings on damage judgments, allowing parties to proceed to trial over liability without risking disaster. For example, a plaintiff might agree to cap defendant’s liability at $500,000 in return for defendant’s agreement to pay at least $100,000 even if there were a defense verdict on liability. See, e.g., Andrea M. Alonso & Kevin G. Faley, High-Low Agreements: You Can Have Your Cake and Eat It, Too, 29 THE BRIEF 69 (1999). 55 One defendant in a punitive damages case with a number of plaintiffs, recognizing that a settlement with some of the defendants would not reduce the amount of any future punitive damage award, arranged with a group of these plaintiffs for them to collect their portion of punitive damages and then 17 of civil settlement practice. The criminal rule, by contrast, explicitly limits the range of bargaining. 56 Finally, for a criminal plea to be entered the defendant generally must not only admit guilt but also satisfy the judge that he both understands the consequences of his plea and has committed acts that would constitute the relevant offense. 57 The efforts that judges make along these lines are often criticized as inadequate, but there is no “factuality” inquiry at all on the civil side. Indeed, civil parties often expressly reserve the right to assert continued innocence of civil wrongdoing as a part of the civil settlement, and sometimes make such settlements in order to be able to make such assertions. Stepping back from details now, consider the different universes occupied by these practices. Civil settlements are an arena of great entrepreneurial creativity, welcomed by the rules and actively encouraged from the bench. Encouragement, in fact, is typically the judge's sole role in civil settlement. The rules have nothing to say about the content of civil settlements, and neither, in the vast of cases, does the judge who nominally presides over the litigation. The terms of a civil settlement need bear no relation to the outcome dictated by substantive law. Nor need they bear any resemblance to the settlement of a nearly identical case that being handled in the courtroom next door. The dark side of unconstrained creativity is arbitrary lawlessness. On the criminal side, for all the complaints about backroom deals, there is a strong background presumption that cases can be settled only in ways consistent with the public's interest in seeing substantive justice done. That is why there are complaints about the aspects of plea bargaining that remain hidden from view, while the backroom bargaining on the civil side rarely comes in for criticism. The notion that plea bargains must accord with substantive standards of justice finds expression in the restrictions on the subjects of bargaining, in the judge's responsibility for ensuring the appropriateness of a plea, and also, perhaps, in the prohibition against judicial involvement in the negotiation of the plea agreement. For if the public's interest is in the substantive appropriateness of a plea agreement, and not in the degree to which the agreement satisfies the private objectives of the offender and the victim, it makes a certain amount of sense to limit the judge's role to a scrutiny of the final terms of the agreement, and to keep the judge out of the process giving rise to the agreement. Suppose we set aside, though, the historical development of the two branches of our legal system. In so doing, we might consider whether the criminal system should be so completely governmental as the present regime implies, and whether the civil system return (“cede back”) those awards to the defendant. The Ninth Circuit upheld the practice against a challenge. In re The Exxon Valdez 229 F.3d 790 (9th Cir. 2000). 56 Fed. R. Crim. P. 11(e)(1). 57 Id. 11(b)(2)-(3). 18 should be as entirely a matter of private ordering. In fact, if we put aside the prevailing assumptions, we see signs in both areas that create doubt. In criminal law, the victims’ rights movement 58 is, among other things, a revival of the strong impulses underlying private prosecution—the insistence that the violation of criminal law involves a wronged party as well as a transgressed law. At a more philosophical level, one finds the same strands of concern in the literature arguing for a vision of criminal justice as restorative justice. 59 One need not believe that victims should set sentences or have a veto over plea bargains to think that the present regime could be more inclusive, more sensitive to the ways in which crime disrupts civil society as well as offending moral and penal codes. Moreover, the greater inclusivity might work to eliminate part of the suspicion of the bargain: if persons outside the bureaucracy were part of the transaction, it might achieve easier public acceptance. One might instantiate such a perception in an altered rule of criminal procedure blending Criminal Rule 11 with Civil Rule 16. A judge presiding over a criminal plea might participate in settlement discussion, allowing the lawyers and defendant to know whether the judge would accept the sentencing recommendations the prosecutor has agreed to make. 60 In a growing number of state courts, judges are in fact allowed to participate in plea discussions, and there is reason to believe that such participation makes the process fairer and more transparent. 61 If the discussion involved victims and families of victims, as well as public prosecutors and publicly salaried defenders, one can even imagine greater public inclination to accept the results, whether or not they were elaborately publicized. Indeed, advocates of restorative justice might imagine three- cornered arrangements, in which an act of restitution or remorse might replace part of what would otherwise be the sentence. Without stepping back into a world of private prosecution, one can posit gains from the appreciation of the extent to which criminal prosecution involves private as well as public values. The converse is equally true. Civil settlements might gain something from our acceptance of the proposition that civil justice remains a part of justice, even when 58 See generally Sue Anna Moss Cellini, The Proposed Victims’ Rights Amendment to the Constitution of the United States: Opening the Door of the Criminal Justice System to the Victim, 14 ARIZ. J. INT’L & COMP. LAW 839 (1997); Walker A. Matthews, III, Note, Proposed Victims’ Rights Amendment: Ethical Considerations for the Prudent Prosecutor, 11 GEO. J. LEGAL ETHICS 735, 738-39 (1998) (providing a brief summary of the history of the Victims’ Rights Movement); Paul Gerwitz, Victims and Voyeurs at the Criminal Trial, 90 NW. U. L. REV. 863, 868 (1996) (“[T]he ‘victims’ rights’ movement…has sought to enhance the place of the victim in the criminal trial process…[T]he contemporary victims’ rights movement has successfully advocated…[for] crime victims [to] be assured of restitution, compensation, and counseling…[to] be consulted before plea bargains are finalized, and…[for the consideration of] victim impact evidence…at sentencing.”). 59 See generally David Dolinko, Restorative Justice and the Justification of Punishment, 2003 UTAH L. REV. 319 (2003); Heather Strang, The Practice of Restorative Justice: Repairing the Harm: Victims and Restorative Justice, 2003 UTAH L. REV. 15 (2003); John Braithwaite, Restorative Justice: Assessing Optimistic and Pessimistic Accounts, 25 CRIME & JUST. 1 (1999). 60 FED. R. CRIM. P. 11(e)(2). In such cases the defendant will know the sentencing range, but in many cases that range could be broad indeed. 61 See Wright & Miller, supra note 36, at 88-91. 19 consensually arrived at. Such a proposition might find its roots in two recollections: that settlements replace trials, which are public, and that settlements result in large part from the discovery powers delegated to the parties by modern procedural rules. And, as in criminal law, one sees in the existing rules traces of unease at the near-total privatization of civil settlement. In civil antitrust actions brought by the United States, proposed settlements must be opened for public comment and then approved by the court as “in the public interest.” 62 The impulse that led to this legislation is clear: competition law is a fundamental part of the market economy, and one worries about too-quickly struck bargains that have ripple effects beyond the parties. This argument, however, has no clear limiting principle: most of substantive law has important public implications and potential externalities. One might concretize this perception in a modest exchange between Criminal Rule 11 and Civil Rule 16. At present, the civil rule gives the judge a role as convener of the conference and, should she wish it, of persuader. Her role on the merits of a settlement is limited to prediction: what might be a trial outcome and what might be the strong and weak points of the parties’ cases (something the judge will know as a result of discovery). Suppose first that the terms of proposed settlements were reported to the judge, at least in cases where suit had been filed. Further suppose that a civil judge considering settlement had the power a criminal judge possesses over a nolo contendere plea: “Such a plea shall be accepted by the court only after due consideration of the views of the parties and the interest of the public in the effective administration of justice.” 63 Like a nolo plea, a civil settlement typically admits no responsibility: it takes no position on the allegations of the complaint. In a criminal case, the judge has the power to refuse a nolo plea if she believes that the failure to accept responsibility will have unacceptable collateral consequences—typically in pending or predictable civil actions. One can imagine a similar power in civil cases. Are there circumstances in which a defendant must accept responsibility, or in which a plaintiff must litigate a case? One possibility might be in civil actions involving repetitive or multiple harms: claims of negligent product design, for example, or fraudulent lending practices. In cases of this kind, ought the legal system to insist on adjudication or more global settlements? As with the criminal judge’s power to reject nolo pleas, one need not imagine many exercises of such power. But even an occasional one would serve to assure that law and civil litigation bore a closer relationship. A second implication of recognizing the public character of private settlements draws on the transparency principle of the criminal process, and the relationship of transparency to fairness. In the criminal process, one can be fairly certain that even moderately well-counseled defendants will not plead to charges or sentence recommendations wildly outside those usual for their jurisdiction. Defense counsel and prosecutors alike know the going “price” of aggravated assault or a first-offender drug 62 15 U.S.C. § 16 (2004). 63 Fed. R. Crim. P. 11(b). 20 possession charge. 64 They know because pleas are public. In this limited sense criminal pleas are fair in ways that many civil settlements are not. Plea bargaining would be even fairer in this regard if, as we suggest below, criminal process took some lessons in other areas from civil process: if criminal discovery borrowed some tools from civil discovery, 65 and if inadequate assistance jurisprudence took some cues from civil malpractice law in assessing competence in settlement negotiations. 66 On the civil side, though, only large repeat players—insurers and recidivist defendants whose scale causes them to behave like insurers—generally have much information at all about patterns of settlement. That knowledge does not tend to disseminate, and several studies suggest that even experienced counsel, insurance adjusters and judges differ widely in their estimates of the “value” of the same cases. 67 Praising plea bargaining as transparent may seem perverse. One standard criticism of the practice is precisely that the bargaining takes place behind closed doors, hidden from public view. 68 We do not wish to minimize that concern, which would remain even if criminal discovery were broadened and standards of defense attorney competence in the bargaining stage were tightened. But civil settlements operate in a different world entirely, a world where even the results of the bargaining are undisclosed. We have, in civil litigation, a system that mimics the price discrimination in the sales of airline tickets—with the added feature that the settling parties don’t even know they are paying or accepting a price substantially different from those in the next courtroom. True believers in the legitimating power of consent may not be troubled, but many others might be: we do, after all, require that prices of publicly traded stocks be posted, and automobiles carry at least the maximum price. One can imagine a settlement registration system, in which the parties would be required to describe, at least generically, the terms of their settlement: “vehicular accident, fractured pelvis, $X in medical and lost wages claimed, settled for $Y”); the parties would not need to be revealed, just the nature of the claim and the terms of settlement. 69 The details of such a system would be controverted, 64 Cf. Bibas, supra note 36, at 2481 (explaining that defense lawyers “develop a feel for cases and can gauge the going rate for particular types of crimes and defendants…[I]nexperienced lawyers have yet to develop an intuitive sense of what a case is worth.”). 65 See notes 94-108 and accompanying text, infra; Bibas, supra note 36, at 2493-94, 2531-32. 66 See notes 109-151 and accompanying text, infra; Bibas, supra note 36, at 2542. 67 See Roselle L. Wissler et al., Decisionmaking About General Damages: A Comparison of Jurors, Judges, and Lawyers, 98 MICH. L. REV. 751, 811 (1999) (reporting that the “awards our jurors actually gave tended to be smaller than the awards the judges and lawyers predicted they would give. One might expect the expertise of judges and lawyers to include accurate predictions of what jurors will do. . . . Systematic errors call for explanations. . . .”); Murray S. Levin, The Propriety of Evaluative Mediation: Concerns about the Nature & Quality of an Evaluative Opinion, 16 OHIO ST. J. ON DISP. RESOL. 267, 289- 91 (2001) (reaching a similar conclusion on the basis of less elaborately analyzed data, reporting “dramatic” differences in the estimates of insurance adjusters, lawyers, and similar persons). 68 See, e.g., Bibas, supra note 36, at 2473, 2475; Rosett & Sklansky, supra note 36, at 606; Wright & Miller, supra note 36, at 10. 69 A more refined format would include the existence and amount of insurance coverage or the existence of a self-insured entity as defendant. 21 and we will not attempt to work them out here. The point is simply that such a system would borrow a valuable tool of transparency from the criminal system. Intermediate regimes and other borrowings are also possible. Suppose that for certain claims—perhaps those involving public safety and repeatable hazards—settlement had to be revealed to and approved, not by a judge, but by a public official charged with health and safety, acting as a guardian ad litem for public interests. Such a regime might decrease the number of settlements, and would almost certainly decrease the amount of many settlements, for the simple reason that one of the features purchased in most settlements is confidentiality. It might have the further consequence of increasing the incidence of certain forms of civil litigation—as the nature of particular repetitive complaints and their settlement value became better known. And it might secure an important public good—the assurance that civil justice was functioning in a way that warranted public confidence. Looking further ahead, one could imagine that such an increase in confidence might justify shedding of regulatory regimes designed to supplement or substitute for civil litigation—e.g., product safety oversight. These proposals are intentionally sketchy. Our chief objective is not to promote a specific agenda for reform of civil and criminal settlement, but rather to demonstrate the benefits of considering each system in light of the other. Each system has lessons to teach the other. The criminal system could use the infusion of creativity and responsiveness typical of the civil system. The civil system could gain much from remembering that civil justice is still a part of equal justice under law and that implicit in that aspiration is a commitment to like outcomes of like cases. B. Former Adjudication: Res Judicata and Double Jeopardy As important as settlement is, the data do not bear out the common wisdom that “everyone” settles or pleads. Our best estimate is that between a quarter and a third of civil and criminal filings end in an adjudicated disposition. 70 But adjudications in the civil and criminal context represent quite different states of rest, and the gulf between them has been growing. Roughly put, the principles of former adjudication on the civil side have reshaped themselves as civil process has evolved. On the criminal side, process has undergone major revolutions since the eighteenth century, but the principles of former adjudication have remained frozen. At early common law, a plaintiff suffering an adverse judgment on a writ of, say, trespass, was barred from bringing a second trespass writ on the same facts. He remained, however, free to bring a second action on the same set of facts on a different writ—say trespass in consimile casu. 71 If he had prevailed in the first suit, he could sue 70 See supra note 32. 71 FLEMING JAMES, JR., GEOFFREY C. HAZARD, JR., & JOHN LEUBSDORF, CIVIL PROCEDURE 674 (5th ed. 2001). 22 on the resulting judgment, but a new suit on the same claim would be barred. 72 Both positions matched the constraints of that procedural system: at common law a pleader could not combine two writs in the same complaint and had to plead facts that tracked the stylized allegations of the writ. Under those circumstances matching the breadth of preclusion to the breadth of the writ made sense. Because modern pleading allows combinations of claims unknown to common law, and because discovery and amended pleadings are both available, the typical stance taken today is that a lawsuit, successful or otherwise, bars a second suit on any claim arising from the same transaction or occurrence as the first. 73 Moreover, this preclusive effect attaches not only to trial verdicts and judgments, but to many earlier stages of litigation: summary judgments and even, in many jurisdictions, demurrers 74 create broad preclusive effect both as to claims actually litigated and to all related claims, whether or not asserted. 75 Only judgments that rest solely on jurisdictional grounds have a preclusive effect limited to the issues actually decided. Otherwise, any judgment—even one explicitly unrelated to the merits, such as a sanction for noncompliance with discovery orders—bars a subsequent action based on the same occurrence or transaction. Modern procedure on the civil side thus imposes broader disabilities, because it gives the parties broader procedural opportunities. As one treatise, stressing both available discovery tools and party responsibility, puts it: Modern pleading rules allow plaintiffs to allege claim formulations in the alternative and afford liberal opportunity to amend any formulation found insufficient on a demurrer or motion to dismiss a claim. There is no reason not to expect plaintiffs to avail themselves of these opportunities. Similarly, modern procedure, particularly discovery, allows plaintiffs abundant opportunity to develop all available evidence that could be used at trial. These opportunities thus enlarge the scope of what might have been litigated by the plaintiff in the first action and should correspondingly enlarge the scope of claim preclusion. 76 In criminal litigation, former adjudication plays a different role, with different rules and a different name: double jeopardy. Double jeopardy started very much as 72 There is some uncertainty about how the former defendant asserted such a bar. Some courts spoke of election of remedies. See Hitchin v. Campbell, 96 E.R. 487 (1772). Others spoke of the cause of action having “merged” in the judgment. 2 HENRY C. BLACK, A TREATISE ON THE LAW OF JUDGMENTS §674, at 811 (1891) 73 See RESTATEMENT (SECOND) OF JUDGMENTS § 24 cmt. a (1982). 74 Some states permit a refiling following the dismissal of an improperly pleaded complaint. See, e.g., Keidatz v. Albany, 249 P.2d 264 (1952). But the trend of modern preclusion law sweeps broadly: so long as a plaintiff has had adequate opportunity to amend pleadings and to reformulate a complaint, a dismissal on a demurrer will bar a new suit based on a revised complaint. See RESTATEMENT (SECOND) OF JUDGMENTS §27 cmt. d (1982); JAMES, HAZARD & LEUBSDORF, supra note 71, at 702. 75 See RESTATEMENT (SECOND) OF JUDGMENTS § 24 (1982). 76 JAMES, HAZARD & LEUBSDORF, supra note 71, at 702. 23 civil former adjudication did: in a world in which charges were narrowly and specifically drawn, and in which not much happened between indictment and trial. In that world, it made sense for jeopardy to “attach” only when things got underway in earnest--that is to say, with the start of trial. And it further made sense that the jeopardy bar would apply only to the “same identical crime,” as Blackstone put it. 77 The early law of double jeopardy reflected these features. So does the modern law. In fact the most striking characteristic of today’s double jeopardy doctrine is that it differs hardly at all from that articulated by Blackstone’s Its most remarkable feature has been its stasis in the face of large changes in the apparatus of criminal prosecution. Today, double jeopardy doctrine plays two roles: First, it serves as the analogue to civil doctrines of former adjudication, preventing the relitigation of charges and multiple punishments for the same crime. Second, it prevents certain forms of prosecutorial misconduct unrelated to either of the preceding functions, by requiring the dismissal of charges where the prosecutor has goaded the defendant into seeking a mistrial. 78 In both roles, double jeopardy law is notoriously inconstant and confusing. Among the more charitable of the terms commonly used to describe it is “quagmire.” 79 The Supreme Court itself has called double jeopardy doctrine “a veritable Sargasso Sea which could not fail to challenge the most intrepid judicial navigator.” 80 We will not attempt in this essay to return double jeopardy law to dry land. We do want to suggest, though, that at least some of the confusion surrounding this area of doctrine—in particular, the confusions associated with the role of double jeopardy in barring relitigation of criminal charges—might be alleviated by understanding and comparison. The understanding relates the constitutional background to the nineteenth-century changes in criminal process. The comparisons, of course, are those across the divide between civil and criminal procedure. When the rules for preventing relitigation of criminal charges are placed alongside the analogous rules in civil process, two differences immediately reveal themselves. The first involves the stage of the proceeding that triggers the doctrine, and the second involves breadth of protection against a subsequent charge. Both are best 77 4 WILLIAM BLACKSTONE, COMMENTARIES *330. 78 See Oregon v. Kennedy, 456 U.S. 667 (1982). We will largely ignore this second role, but we should note that none of the changes we suggest would prevent double jeopardy from continuing to serve it. 79 E.g., Commonwealth v. Burkhardt, 586 A.2d 375, 377 (Pa. 1991); Andrew L. Subin, The Double Jeopardy Implications of In Rem Forfeiture of Crime-Related Property: The Gradual Realization of a Constitutional Violation, 19 SEATTLE U. L. REV. 253, 268 (1996). Cf., e.g., Akhil Reed Amar, Double Jeopardy Law Made Simple, 106 YALE L.J. 1807, 1807-08 (1997) (complaining that double jeopardy law is “full of double talk” and “double takes”). 80 Albernaz v. United States, 450 U.S. 333 (1981). Justice Rehnquist, who wrote for the Court in Albernaz, elsewhere compared double jeopardy doctrine to a “Gordian knot.” Whalen v. United States, 445 U.S. 684, 702 (1980) (Rehnquist, J., dissenting). On these and other metaphors for double jeopardy law. See George C. Thomas III, Multiple Punishments for the Same Offense: The Analysis After Missouri v. Hunter, or Don Quixote, the Sargasso Sea, and the Gordian Knot, 62 WASH. U. L.Q. 79 (1984). 24 understood—and perhaps reformed—with reference to the historical development of the two processes. First, the triggering stage. In the civil context, former adjudication requires a judgment, but a judgment at any stage of the proceeding will do. That position reflects the opportunities for investigation (discovery) and maneuvering (amended pleadings, joinder) available to the plaintiff. Double jeopardy, by contrast, does not always require a judgment, but even a judgment will not always trigger its operation. Double jeopardy “attaches,” barring relitigation, only with the opening of the prosecution case at trial, and thus does not come into effect with dismissals prior to trial. Where there has been an adjudication before trial, double jeopardy does not prevent the prosecution from refiling after further investigation and stronger proof. Pretrial adjudication simply does not count in the world of double jeopardy. A charge dismissed by a judge or unindicted by a grand jury for want of evidence can be refiled if the prosecutor uncovers more evidence. For that matter, it can be refiled even if the prosecutor fails to uncover more evidence. 81 Waiting until trial for jeopardy to attach gives the police and prosecution several bites at the apple: a weak case or a missing witness can be corrected, and crime will be duly, if tardily, punished. Such a policy made a good deal of sense in a world in which criminal process closely resembled pre-modern civil process. Before professional policing, when the collection of evidence was essentially an amateur and private activity, a new prosecution might well involve a new complaining witness and could thus be thought of as involving a different “party.” Moreover, not too much happened between arrest and trial: bail proceedings and arraignment were not about testing evidence in the case. In a world in which trial was the main and only show, it made sense for doctrines of former adjudication to be linked to trial. A preliminary hearing that actually tested the evidence was not part of the procedural landscape. The persistence of such a doctrine in the face of professional police and the widespread use of the preliminary hearing, though, is a good deal more puzzling. Essentially it tells the prosecutor that she need not prepare so well or so thoroughly for a preliminary hearing as her civil counterpart must for a motion for summary judgment. It makes the preliminary hearing “preliminary” in two senses of the word. One can view the doctrine in part as a reminder of the extent to which the prosecution still relies on friendly witnesses, a small relic of private prosecution. Even in such a world, however, one could imagine postponement of the complaint and the preliminary hearing until such a witness appeared—with the concomitant doctrinal result that double jeopardy would apply to such dismissals. If an analogous rule seems sensible in civil process, it is worth asking whether it would make sense for criminal litigation as well. In this instance, comparison illuminates the question because the two doctrines started out in a parallel position. But as civil process broadened the investigatory powers and procedural maneuvering room of the parties, the doctrines of former adjudication gradually 81 Some states throw the defendant a lifeline after two dismissals. For example, in California, a twice dismissed charge cannot be refiled—unless it falls into one of several exceptions, such as “excusable neglect” in the case of violent felonies. CAL. PENAL CODE §§ 1387(a), 1387.1(a) (West 2005). 25 broadened. Double jeopardy, by contrast, whether because of the reluctance to remodel constitutional doctrines or because of concerns about law enforcement, remains in an essentially eighteenth-century mold. Comparison with the civil model is hardly dispositive, but it raises important questions. The second interesting comparison of double jeopardy and former adjudication concerns the scope of the preclusion imposed. Civil former adjudication doctrine covers both the claim brought and all transactionally related claims and legal theories. Currently, a claim arising out of a traffic accident must include damages for personal injury, lost wages, and property damage; an effort to bring one claim for property damages and a second for personal injuries will be barred. 82 This broad range of preclusion typically finds its justification not just in efficiency but in the availability of broad discovery and relatively easy amendment of pleadings and joinder. If the procedural system gives a party ample investigative tools and substantial procedural room to maneuver, it seems fair to ask that these opportunities not be deployed piecemeal. By contrast, double jeopardy still lives in the world of the writs: when double jeopardy applies, it covers only the precise crime charged, lesser included offenses, and offenses for which the crime charged is itself a lesser included offense. 83 Double jeopardy does not bar a subsequent prosecution of a crime containing a different element and lacking at least one of the elements of the previously charged offense, even though it arises from the same set of facts. Suppose a driver intentionally rams another car, causing both property damage and personal injury. Vandalism and battery each have an element not shared by the other: vandalism requires property damage, and battery requires force or violence against a person's body. 84 So the driver could be prosecuted first for vandalism, and later for battery, or vice versa. Prosecutors would be under no obligations to bring the two charges in one action. Can one justify the narrowness of the double jeopardy bar? The United States Supreme Court briefly thought not. In Grady v. Corbin 85 a drunk driver struck another car and killed an occupant of that vehicle. He was cited for and pleaded guilty to driving while intoxicated; the district attorney’s interoffice communications were so poor that the lawyer responsible for the traffic court did not know of the death and did not ask for a continuance. When Corbin was later charged with vehicular homicide, he asserted double jeopardy as a bar, and the U.S. Supreme Court agreed, taking a half step toward the “same transaction” test for double jeopardy, even as it denied it was doing so. 86 Even 82 See, e.g., Rush v. City of Maple Heights, 147 N.E.2d 599 (Ohio 1958). 83 See United States v. Dixon, 509 U.S. 688 (1992); Blockburger v. United States, 284 U.S. 209 (1932). 84 See, e.g., CAL. PENAL CODE §§ 240, 242 (West 2005). 85 495 U.S. 508 (1990). 86 Footnote 15 of the majority opinion sought to distinguish the Court’s ruling from a “same transaction” principle: Adoption of a “same transaction” test would bar the homicide and assault prosecutions even if the State were able to establish the essential elements of those crimes without 26 that cautious step was unstable, however. Three years later, in United States v. Dixon, the Court overruled Grady and explicitly returned to the “elements” test: so long as a new indictment contains any element not already contained in the original charges and drops at least one element of the earlier charges, the prosecution is not barred. 87 In neither Grady nor Dixon were the opinions particularly illuminating. The Court’s opinion in Grady, expanding the scope of the bar, alluded to the work load of prosecutors’ offices: Prosecutors’ offices are often overworked and may not always have the time to monitor seemingly minor cases as they wind through the judicial system. But these facts cannot excuse the need for scrupulous adherence to our constitutional principles. See Santobello v. New York, 404 U.S. 257, 260 (1971) (“This record represents another example of an unfortunate lapse in orderly prosecutorial procedures, in part, no doubt, because of the enormous increase in the workload of the often understaffed prosecutor’s offices. The heavy workload may well explain these episodes, but it does not excuse them”). With adequate preparation and foresight, the State could have prosecuted Corbin for the offenses charged in the traffic tickets and the subsequent indictment in a single proceeding, thereby avoiding this double jeopardy question. 88 The dissent in Grady agreed about what was at stake in the case—requiring “prosecutors to observe a rule we have explicitly rejected in principle: that all charges arising out of a single occurrence must be joined in a single indictment” 89 —but did not explain why such a rule was inappropriate. The majority opinion in Dixon, written by one of the Grady dissenters, was no more illuminating. The Court explained in Dixon that the older, elements-based rule had prevailed at common law, and that Grady had given insufficient weight to that common law understanding. If the exercise represented by this essay has value, we ought to be able to say more than that. The striking thing about modern double jeopardy doctrine is its failure to reshape itself in step with changes in the nature of criminal process. Had doctrine evolved with the changes in the rest of criminal process, we would preclude prosecution proving the conduct for which Corbin previously was convicted. The Court, however, has “steadfastly refused to adopt the ‘single transaction’ view of the Double Jeopardy Clause.” Garrett v. United States, 471 U.S. 773, 790 (1985). But see Jones v. Thomas, 491 U.S. 376, 388-389 (1989) (BRENNAN, J., joined by MARSHALL, J., dissenting) (maintaining that “the Double Jeopardy Clause requires, except in very limited circumstances, that all charges against a defendant growing out of a single criminal transaction be tried in one proceeding”). Id. at 524. 87 509 U.S. 688 (1993). 88 Grady, 495 U.S. at 524. 89 Id. at 526. 27 for all transactionally related offenses when there had been a judgment either at a preliminary hearing or at trial. On the face of things, one might imagine that such a doctrine would adapt itself nicely to criminal law. A complaint or police investigation usually focuses on a particular event: a robbery, murder, or traffic accident. Often the charge that will most naturally occur is obvious, but further investigation may shift perceptions and, as in Grady, some event occurring after the initial report may change the obvious charge—from drunk driving to homicide. But fairness to the defendant as well as administrative efficiency suggests that requiring the prosecutor to bring “all charges arising out of a single occurrence . . . in a single indictment” has much to be said for it, given the substantial increase in public funding of the investigative arm of the criminal justice system that has occurred since the double jeopardy clause was written. The widespread use of the preliminary hearing, at which evidence is actually presented and tested, similarly suggests that any such stage--not just the empanelling of a jury, should trigger preclusion. What is the argument on the other side, the argument too obvious for any of the Justices to discuss it? The argument that one doctrine is constitutional and the other is not does not carry much weight. The constitutional doctrine has its roots in common law, and as legal contexts alter, even constitutionalized understandings may sensibly shift in response, as they have in many areas. The Supreme Court itself has recognized as much in other contexts: when most felonies were no longer capital, for example, the use of deadly force to stop fleeing felons was no longer per se reasonable under the Fourth Amendment. 90 It would hardly be a stretch for the Court to conclude, along similar lines, that the bureaucratization of modern prosecution—and the addition, by statute, of countless new and discrete criminal prohibitions 91 —warrants reexamination of the scope of the double jeopardy bar. Beyond the realm of argument, one can speculate about considerations that may lie in the background of the Court’s reluctance to re-examine criminal former adjudication. The twentieth-century Supreme Court placed a number of constitutional constraints on police and prosecutorial behavior. Might it seem unfair for the Court to handcuff law enforcement, as some would see it, with a welter of new constitutional constraints, and then to decide that the police, as thus constrained, got only one bite at the apple? Or might, as the Grady majority suggested, the limited reach of the double jeopardy bar have something to do with the resources and organization of prosecutors’ offices? One could imagine that, because the state is usually paying for the defense as well as the prosecution, that re-prosecution creates no financial burden for the 90 Tennessee v. Garner, 471 U.S. 1 (1985). 91 See, e.g., William Stuntz, The Pathological Politics of Criminal Law, 100 MICH. L. REV. 505, 512-19 (2001). Professor Stuntz notes that relatively few criminal prohibitions are actually lesser-included versions of other prohibitions. For the most part, he points out, “criminal codes consist of a great many more sets of overlapping circles than concentric circles.” As a consequence, “defendants who commit what is, in ordinary terminology, a single crime can be treated as though they committed many different crimes—and that state of affairs is not the exception, but the rule.” Id. at 519. 28 defendants. 92 But a glance across the procedural divide suggests that this consideration, standing alone, is insufficient to justify the traditional rule. Plaintiffs’ lawyers can have limited resources and organizational difficulties, too. But in the civil system we do not waive former adjudication if the plaintiff offers to pay the costs of the first defense. The question, then, is why the same legal system should come to opposite conclusions about the doctrine depending on whether it arises civilly or criminally. Does the apparent irrationality of double jeopardy doctrine lie in the nagging suspicion that criminal law is less entirely “public” than we typically insist? If private interests—those of victims, for example—are involved and those interests inadequately represented by the police and prosecutor at a given time, should we allow a later representation of those interests by a different or more diligent prosecutor? 93 In the same way that we do not allow ordinary legislation to make itself impervious to repeal or amendment by a new majority, the present narrow scope of double jeopardy might be thought of as appropriate modesty concerning the effectiveness of public representation. Again, we do not propose here to predict what a reexamination with such questions in mind would ultimately produce. Our goal is more modest. We wish only to point out the manner in which a comparison across procedural regimes could help us begin to untangle one part of double jeopardy law by suggesting alternative approaches and throwing into doubt some old assumptions. C. Investigation and Discovery In exploring settlement and former adjudication, we have stressed the information-gathering of the parties. The compressed trial is a hallmark of U.S. procedure, both civil and criminal. Because both systems must allow for the possibility of a jury, and because a jury is thought to require presentation of evidence in a 92 Such an argument is ahistorical: double jeopardy doctrine developed before the state assumed the burden of paying for criminal defense of felonies. 93 One civil analogue to the restrictive scope of charge preclusion on the criminal side appears in a relatively obscure feature of civil issue preclusion. The Supreme Court has held that the United States government is not bound to the rule of non-mutual issue preclusion, which applies to other parties, because it would be administratively infeasible to require the United States to appeal every litigative loss on an issue of statutory or constitutional construction, lest it be forever bound by the preclusion rules. See United States v. Mendoza, 464 U.S. 154 (1984). The Court’s grounds were administrative: the rules which make sense in an individual context do not when the enterprise in question is as large as the United States. Perhaps the same might be said about charge preclusion on the criminal side. The difficulty, though, is that prosecutorial authorities are typically focused on a given defendant once he or she has come into the system: it would not seem to be burdensome to ask the prosecutor to delay the complaint until he has assembled enough information to be sure that he knows all the available charges arising from the acts under investigation—particularly if, as is currently the case, he can file a superseding indictment if additional charges come to light. The existing rule seems to invite premature complaints and preliminary hearings, with the prosecutor assured that a failure to assemble the case carefully will have no permanent consequences. 29 compressed space, 94 essentially all factual material conceivably relevant to the case must be uncovered before trial begins. Both systems use investigation—the interviewing of witnesses and examination of premises—for part of this investigation. But the means differ substantially. For civil adjudication this phase of informal investigation precedes a phase of compelled disclosure and then compelled discovery in which the court orders the parties to produce documents and respond to questions under oath, in a broad—some say over- broad 95 —phase of factual investigation that often lays the groundwork for settlement or summary judgment. The guiding principle is “no surprises”: no surprise witnesses, and no surprise testimony. In one of a number of cases that could be cited for this proposition, the First Circuit ordered a new trial in a medical malpractice action because the defendant introduced letters from the plaintiff that had not appeared on a pretrial exhibit list. No one disputed the pertinence of the letters. The objection was rather that the surprise production of the evidence constituted “trial by ambush” and warranted reversal. 96 A criminal litigator would not know what to make of such a point. Criminal defense lawyers (and their clients) live and die by the surprise witness, and both prosecutors and defense lawyers are regularly surprised by witnesses’ appearances or statements. This difference flows from the different tools of investigation employed in the two systems and by the courts’ stance toward the existing doctrines. Criminal discovery relies on each party’s own investigations and subsequent disclosures by that party, rather than on the mutual adversarial probing characteristic of civil process. To do this investigating, the prosecution of course uses police and similar officers. In the vast run of cases, law enforcement officers with broad search powers conduct virtually all the preliminary investigation, often before any lawyer becomes involved. In run-of-the-mill cases neither side may go much beyond the investigation conducted by the police. The defendant relies either on a violation of constitutional restraints on search and interrogation to suppress evidence, or on in-court challenges to the credibility of prosecution witnesses. In other cases—particularly cases involving wealthy defendants—the defense will employ private investigators to conduct its own version of the police investigation, trying either to bolster its challenges to the credibility of a prosecution witness or to generate independent evidence. In so doing, the defense will be aided by knowing whatever the government thinks might be exculpatory. Under existing law the prosecution must disclose to the defendant any exculpatory evidence. 97 The defendant has no equivalent obligation to disclose inculpatory evidence—and it is not clear whether such an 94 One might question the second assumption, developed in a period when the average juror was illiterate, but it is at the moment a given of jury trial, even those that stretch over many months. See MIRJIAN R. DAMA?KA, EVIDENCE LAW ADRIFT 58-73 (1997). 95 See, e.g., Hazard, supra note 28. 96 Klonoski v. Mahlab, 156 F.3d 255 (1998). 97 Brady v. Maryland, 373 U.S. 83 (1963). 30 obligation would survive constitutional scrutiny were it imposed. Depending on the applicable statutes and rules, the defense may be required to reveal its intention to pursue one of several affirmative defenses—insanity, an alibi, etc.—but its obligation almost never stretches much further than that until shortly before trial. At that point, both sides generally must reveal to one another a list of witnesses, documents, and other material they intend to produce at trial, together with any written statements of those persons. On this basis both sides go to trial. In more complex criminal cases this process of investigation may begin to look somewhat more like the civil system—but only on the prosecution side. The prosecutor may convene a grand jury to subpoena documents and, more importantly, to require witnesses to answer questions under oath. Grand jury questioning has some of the features of a civil deposition, though notably without the defendant’s lawyer having an opportunity to cross-examine or to raise objections. Comparing the two systems broadly, one can discern two models: a two-stage model involving mutual independent investigation followed by disclosure shortly before trial and a three-stage model involving investigation, disclosure, and then compelled pretrial testimony. 98 Civil process uses the second model, the three-stage inquiry. Criminal process uses both. It employs the two-stage model in most cases, but supplements it with compelled testimony, under the sole control of the prosecutor, in a significant category of cases. Civil litigators who venture into criminal cases tend to be stunned and often outraged by their inability to depose government witnesses or even to file interrogatories or requests for admissions. We should take note of their outrage, for it may have some basis. Lawyers who cross over from civil to criminal litigation (or vice versa) often engage in precisely the kind of cross-comparison of the two systems we are urging— although the comparisons they draw are often partisan, and they may stop drawing comparisons altogether once they become acclimated. We think these comparisons are warranted and that both sides have something to learn. We begin with how the study of criminal discovery might benefit civil process. What is most remarkable about the criminal discovery system is that it works as well as it does. The constitutional obligation of prosecutors to turn over exculpatory information is largely self-enforced: the decision whether particular information is “exculpatory” is made in the first instance by prosecutors, and it is often a decision that courts never have an opportunity to review. Horror stories regularly surface of prosecutors failing to disclose information that, when it later to comes to light, seems obviously exculpatory to sober outside observers. But there is no doubt that many prosecutors—probably most—regularly go well beyond their constitutional and statutory obligations of discovery, providing defendants with virtually all information uncovered 98 The most recent version of the Federal Rules of Civil Procedure might be said to contemplate a four-stage process: investigation, disclosure, compelled discovery regarding “claims or defenses” of the parties, followed in a few cases with further discovery concerning the “subject matter” of the suit, a phrase not defined but apparently involving broader inquiries than “claims or defenses.” 31 by law enforcement. Nor is it apparent that the absence of depositions significantly impairs the fairness of most criminal trials. Might this regime, enhanced with closer judicial supervision, serve as a model for the reform of the civil system? Indeed, has it already so served, though without explicit recognition? The dark side of the civil system’s insistence on pretrial knowledge is that it often results in overkill. A lawyer with access to discovery tools has to use them, sometimes to excess. Although the data suggest that most cases proceed with only modest discovery, 99 in some outlier cases discovery becomes a monster whose costs exceed outside observers’ estimates of need or rationality. The last two decades have seen several changes in procedure designed to rein in discovery in such cases. 100 In so doing the civil system has borrowed, perhaps unwittingly, 101 from the criminal system. What if it borrowed more? Suppose civil discovery typically stopped at the second phase—investigation and uncompelled fact-gathering followed by an exchange of evidence, with each side presenting a summary of its own case. At that point it would be possible for a judge to evaluate whether the civil equivalent of the grand jury—the compelled disclosure regime of document production and depositions—was warranted. In a number of cases, probably those roughly paralleling those in which the grand jury is used criminally, it would be. In others, however, the absence of judicial permission to proceed to this second phase would probably not leave the parties in a materially worse position than they now occupy, but would save both significant cost. The 2000 Amendments to Federal Rule of Civil Procedure 26 took a very small step in this direction. First, they require the parties, after their own initial investigation but before seeking discovery from the other side, to “disclose”—to identify witnesses and documents supporting their own claim or defense. 102 This represents an unacknowledged, symmetrical adaptation of the criminal prosecution’s disclosure of exculpatory evidence: here the plaintiff must put “inculpatory” evidence on the table, and the defendant exculpatory. 103 The second half- 99 David M. Trubek et al., The Cost of Ordinary Litigation, 31 UCLA L. REV. 72, 89-90 (1983) (reporting than in most normal-stakes litigation there were only one or two “discovery events” and they were not disputed). 100 Rule 26 of the Federal Rules of Civil Procedure was amended in 1983 to require discovery requests be signed by the propounding lawyer and that the lawyer's signature certified both a basis for the discovery sought and a representation that the discovery in question was not excessive given the case. More recently, 2001 amendments limited the scope of discovery available as a matter of course, allowing parties to discover only material relevant to pleaded claims and defenses (narrowing the rule from the pre- revision "subject matter" of the lawsuit). The same set of amendments presumptively limited the number of interrogatories and length of depositions. 101 The advisory committee notes to the relevant revisions make no mention of the criminal system. 102 In addition, the defendant must reveal the existence and limits of insurance coverage and the plaintiff a calculation of damages. 103 The hope, expressed by the drafters, was that it would eliminate some of the early and routine discovery requests. It may also reflect the thought that this mutual laying down of the cards would bring about early settlement negotiations. 32 borrowing from the criminal system was the taking-back into judicial hands the decision about whether the broadest sort of discovery inquiries—those involving not just the “claims [and] defenses” but the “subject matter” of the suit ought to be discoverable. This change may represent a very small step towards the recognition that public as well as private interests are at stake in civil litigation. One could imagine the civil system’s repaying such borrowings by lending criminal procedure some of the symmetry that characterizes civil discovery. Suppose we gave the targets of grand jury investigations qualified rights to examine witnesses called to testify and to name additional witnesses who should appear. We would then inhabit a world in which, as now, the grand jury is not often used, but in which, when it was used, it would afford roughly symmetrical opportunities to prosecution and defense. Implementing such a proposal would require making some difficult judgments: who should count as a target, when and how that determination should be made, how to handle targets without lawyers, how to accommodate concerns about witness safety and the integrity of ongoing investigations, etc. But the gains from making grand jury proceedings more even-handed might be worth the costs. It would alleviate the often- criticized asymmetrical quality of the grand jury, which, in an altered process might be less likely to indict the proverbial ham sandwich. With defendants having the opportunity to present and examine witnesses, the “protection” of a grand jury indictment, thought by the framers of the Constitution to be worth embodying in an amendment, might again become real. 104 Many states now create an equivalent opportunity in their preliminary hearings, and it is unlikely that either the republic or effective law enforcement would end if this practice were extended to the grand jury. There would even be advantages for prosecutors, who then often would be able to introduce the grand jury testimony of witnesses who no longer were available at the time of trial. 105 For this improvement in the defendant’s situation, it would, however, be fair to ask for a form of discovery in return. Prosecutors and defense attorneys alike, for 104 Even more modest borrowing from civil discovery might do much to restore the grand jury's role as a check on prosecutorial power. Since 1978, New York has allowed targets of grand jury investigations to bring their lawyers with them when testifying before the grand jury—a privilege that civil litigants have long taken for granted when testifying at depositions. The result has been sharp increases both in the number of targets choosing to testify before the grand jury and in the number of cases in which grand juries vote not to indict. See William Glaberson, New Trend in Grand Juries: Meet the Accused, N.Y. TIMES, June 20, 2004, at A1. 105 Until the Supreme Court’s recent decision in Crawford v. Washington, 541 U.S. 36 (2004), there was uncertainty regarding when, if ever, the Confrontation Clause of the Sixth Amendment prohibited the prosecution from introducing grand jury testimony at trial, despite the defendant’s inability to cross- examine witnesses before the grand jury. Crawford makes clear that the absence of cross-examination makes grand jury testimony flatly inadmissible as substantive evidence, although it still may be used for impeachment if the witness testifies differently at trial. If, on the other hand, defendants were able to cross- examine witnesses before the grand jury, the Confrontation Clause would not bar the government from introducing, at trial, the grand jury testimony of witnesses who later became unavailable. Nor would the hearsay rule bar such evidence, at least as long as the defendant had not only an opportunity but a “similar motive” to cross-examine the witness before the grand jury (similar, that is, to the motive the defendant would have at trial). FED. R. EVID. 804(b)(1); United States v. Salerno, 505 U.S. 317 (1992). 33 example, are often directed, by statute or court order, to disclose witness and exhibit lists before trial. It is an open secret, though, that criminal defense attorneys generally can disregard this rule with impunity. In the words of one experienced observer of criminal proceedings (herself, we should note, a former federal prosecutor), “when it comes to discovery, the prosecutors play chess and the defense plays street hockey.” 106 As a result, in the average criminal trial, the defense has the prosecution list of witnesses, but the prosecution has to guess at the defense strategy. No one thinks this asymmetry is constitutionally required. If defense lawyers are to have access to the grand jury, defense lawyers should be obliged to produce trial witness and exhibit lists and witness statements—and the obligation should be given some of the same teeth that failure to disclose or discover has in civil cases. Importing this much of civil discovery practice might improve both the fairness of trial and the accuracy of outcomes without unduly burdening defendants. The net result of the paired borrowings just described would be slightly to expand the scope of discovery in some criminal cases and substantially to contract the scope of discovery in some civil cases. The qualifications to this description flow from the gap between the availability and the use of procedural devices in both systems. Most criminal cases do not involve significant use of the grand jury, 107 and most civil cases do not involve extensive discovery. 108 The effect of these changes might be substantial in a small portion of cases, but in those cases it would be largely beneficial. Because much of civil discovery in big cases is redundant, presumptive limiting and tighter judicial control might create a better situation for both parties, and surely for the administration of justice seen more generally. Similarly, asymmetrical access to the grand jury is significant only in complex cases—some of which (e.g., procurement fraud) resemble civil litigation closely enough to make the asymmetry especially anomalous. In smaller cases the changes in criminal process here envisioned would have their chief effect in the regular enforcement of existing rules that would require both sides to give the other a glimpse into trial strategy before the jury is sworn; that seems an undangerous idea, and one likely to improve both justice and efficiency. At present we operate wildly different discovery regimes. In one, state officials have wide power to question and search, but there is no pretrial opportunity to ask cross- examined questions under oath. In the other system, virtually no state officials participate, but virtually every piece of information uncovered is subject to the adversarial process, both in arguments about whether it should be discovered, and then concerning its significance. There is reason to think both systems do some things better than the other, and that both might profitably borrow some practices from the other. The 106 Conversation with Prof. Laurie L. Levenson, Loyola Law School, Los Angeles, California (June 12, 2002). 107 This statement holds substantially true even in the federal system which formally requires a grand jury indictment. Most indictments make only slight use of the grand jury's powers: a statement of a law enforcement officer is presented and the jury indicts. Daniel Richman, Grand Jury Secrecy: Plugging the Leaks in an Empty Bucket, 36 AM. CRIM. L. REV. 339, 342 (1999). 108 See Trubek et al., supra note 99. 34 borrowings we have suggested would reduce somewhat the level of discovery in some civil cases, while increasing the adversarial participation in discovery in the criminal system. We think both results would be salutary. But our larger point, which does not depend on that assessment, is that comparisons of discovery rules across the procedural divide can provide a degree of perspective likely to assist those who seek to improve the functioning of both systems. D. Remedies for Failed Process The exploration thus far has tracked what one might call “ordinary procedure”— in which the respective systems work more or less according to design—with our inquiry focused on what we can learn from the differences in such normal operations. We turn now to a comparison that is out-of-the-ordinary in two respects. First, it explores the two systems’ reactions to the collapse and the failure of ordinary procedure. Second, and unlike the preceding examples, we lay less stress on possible borrowings—though we suggest a couple—than on the insights that such a comparison yields. Occasionally procedural systems fail in ways that ordinary process cannot fix. A judgment is entered. The court has correctly identified the applicable law and has properly applied that law to the facts as they appeared. But a second view suggests that something has gone seriously wrong. An unsuspected witness returns from the upper Amazon with the will in her luggage. A newly-available scientific process definitively establishes that the convicted defendant was not present during the critical events. Or, a lawyer charged with representing his client failed to appear at the hearing on a dispositive motion and judgment was entered against the client. All legal systems must address situations like these, balancing the desire for accuracy with the need for finality. In general, procedural collapses come in three varieties, which we shall treat in ascending order of conceptual difficulty. The civil and criminal systems approach these varieties with two kinds of remedies: specific and substitutionary. The specific remedies reopen or set aside the flawed judgment. The substitutionary remedies instead seek damages from those responsible for the failures. The specific remedies display a rough symmetry across the civil-criminal divide; the substitutionary remedies do not. The symmetry and the asymmetry both repay exploration. Perhaps the easiest cases involve both new evidence and a finding that one of the parties possessed that information during the original proceeding but failed to fulfill his legal obligation to disclose it. For example, the prosecutor has breached his constitutional or statutory obligation to disclose exculpatory evidence to the defendant. 109 109 The constitutional basis for this obligation flows from Brady v. Maryland, 373 U.S. 83 (1963). A second, statutory obligation in federal courts flows from the Jencks Act, 18 U.S.C. § 3500 (2004), and is reflected in FED. R. CRIM. P. 26.2. 35 Or, in a civil case, the plaintiff has requested certain documents 110 in defendant’s possession, but the defendant has failed to produce them. If the evidence is significant, both the criminal and civil systems respond alike—by setting aside the judgment and ordering a new trial. These cases are easy for the courts because the misconduct of the side relying on the judgment obliterates the weight that might otherwise be given to finality. In criminal cases the courts reach this result either on habeas corpus (and its statutory modifications) or by ordering a new trial; where the strength of the exculpatory evidence is in doubt, courts weigh heavily the culpability of the prosecutor in withholding the information. 111 In civil cases courts reach the same result under rules or statutes permitting the reopening of judgments, 112 even when they have to engage in some torture of the applicable statute to reach this result. 113 Next in order of difficulty are the cases where the evidence is new and significant, but where the other party bears no responsibility for the failure to uncover the information. Here fault lines begin to appear in both systems, with courts forced to choose between accuracy and finality. Courts first seek a way out by inquiring into the prior diligence of the party seeking relief; failures in adversarial diligence justify the denial of relief. 114 When this move does not work, the courts’ next step is to apply a time limitation: the civil Rule sets a one-year limit on such motions; the criminal Rule a 3-year limit. 115 Both the existence of these limits and the longer time for criminal than civil 110 FED. R. CIV. P. 26(a) requires parties to identify or produce certain documents without a request; FED. R. CIV. P. 34 permits the parties to require production of any document relevant to a claim or defense of any party. 111 WAYNE R. LAFAVE, JEROLD H. ISRAEL & NANCY J. KING, CRIMINAL PROCEDURE 950-51 (3d ed. 2000). See, e.g., United States v. Mele, 462 F.2d 918 (2d Cir. 1972) (ordering new trial although new evidence was not conclusive, because prosecution’s suppression was deliberate). 112 See, e.g., Rozier v. Ford Motor Co., 573 F.2d 1332 (5th Cir. 1978) (reopening judgment for Ford’s failure to produce engineering report suggesting a defect in gas tank of similarly designed vehicle). 113 The torture is sometimes necessary because FED. R. CIV. P. 60(b)(2) and (3), which provide for reopening judgments on the basis of newly discovered evidence and for fraud are both subject to a one-year statute of limitations. The sections not subject to this time limit include other provisions, for “void” judgments or for “any other reason justifying relief.” Ordinarily, one would read these sections to exclude fraud, since that is separately specified. But when courts are convinced that the opponent has engaged in intentional concealment, they strain to put the case in the categories not subject to the time limit. JAMES, HAZARD & LEUBSDORF, supra note 71, at 786-87 (“The most important variables for determining whether relief from the judgment will be granted are the proof of fraud..., the opportunity for having detected it, and the consequences of its commission.”). 114 The text of the civil Rule explicitly requires that the new evidence “by due diligence could not have been discovered” in time for ordinary processes to account for it. FED. R. CIV. P. 60(b)(3). The criminal rule does not contain this text, but the cases are uniform in requiring such a showing. The United States Code Service annotations to FED. R. CRIM. P. 33 (Mathew Bender & Co. 2004) contain two pages of single-spaced case summaries, almost all denying relief on this ground. 115 This limit in the criminal system is in fact longer than the Rule would suggest. FED. R. CRIM. P. 33 limits only the time within which the convicted defendant can bring a motion. It does not, of course, limit the power of the executive to grant clemency or pardons. In some celebrated cases release and exoneration has come many years after the original conviction. See, e.g., Henry Weinstein, DNA Frees Man Jailed for 22 Years: After Spending Nearly Half His Life Behind Bars, a Louisiana Man Is Cleared by New 36 motions display the uneasy tension between our wanting to get it right and our wanting to get it over. Not surprisingly, both systems look carefully not just at the novelty of the evidence, but at its significance: how likely is it to change the result if a new trial is ordered? 116 Finally, we come to the most difficult category, in which the failure of process is due neither to the other side nor to the fortuitous appearance of new evidence, but to the party’s own lawyer. As we shall explore more fully in the next section, 117 both civil and criminal litigators must meet professional standards of competence, 118 and the legal system gives clients remedies for lawyers’ failure to meet the standards: a damage action for malpractice; and a motion to set aside the judgment. 119 In theory these remedies might be cumulative: A wrongly convicted defendant could sue for his release and then bring a civil action against his former lawyer for damages suffered during the imprisonment resulting from the lawyer’s negligence. 120 A civil party could similarly move to have a judgment set aside—and then sue her former lawyer for the costs of doing so. In practice, these two paths often represent alternatives: a client who can be relieved of a judgment may suffer few compensable damages, 121 and, even if he has suffered some damages, may find the setting-aside of the judgment so important as to make any possible damages trivial. Conversely, at least some clients— those who have suffered only monetary harms—may be completely compensated by money damages and thus have no incentive to disturb a judgment. 122 Taking malpractice first—because its apparent symmetry reveals the fault lines sharply—one finds an interesting discrepancy. Clients represented badly either in civil or criminal matters have, in theory, a malpractice remedy. On the civil side, malpractice doctrine has a familiar tort structure: breach of duty and causation. Defining duty, courts ask what a lawyer of “reasonable competence [would do] under similar Testing of Evidence in a 1981 Rape, L.A. TIMES, Sept. 20, 2003, at A11; Carol Morello, Va. Inmate Imprisoned 21 Years Released a Day After DNA Tests, WASH. POST, Feb. 13, 2003, at B1. 116 See JAMES, HAZARD & LEUBSDORF, supra note 71, at 788-89 (on civil cases). See, e.g., United States v. Aponte-Vega, 230 F.3d 522 (2d Cir. 2000) (relief warranted only if new evidence would probably lead to acquittal). 117 See text at notes 178-199 infra. 118 MODEL RULES OF PROF’L CONDUCT R. 1.1 (2003) (“A lawyer shall provide competent representation to a client.”). 119 On the criminal side a convicted defendant can, either on direct appeal or on habeas, seek to have his conviction reversed or set aside if counsel fails to meet Sixth Amendment standards of “effectiveness.” On the civil side in the federal system such a motion is brought under FED. R. CIV. P. 60. States all have analogues; see, e.g., CAL. CIV. PROC. CODE §473(b) (West 2005). 120 For reasons indicated below, the list of recipients of this pair of remedies is not long. See nfra notes 128-131 and accompanying text. 121 A default judgment set aside might be the most common example. 122 One example would be a plaintiff whose lawyer had failed to file the complaint within the statute of limitations but who had recovered the amount sought in the original complaint from the lawyer’s malpractice carrier. 37 circumstances.” 123 For litigators a standard treatise lists recurrent areas of liability involving failures to investigate, evaluate, plead properly, make appropriate motions, present witnesses, inform clients of the implications of settlement offers, and to meet deadlines. 124 These standards hardly require perfection, 125 but they do hold lawyers to account for lack of effort, 126 and an increasing number of successful claims involve lawyers’ failure adequately to counsel clients about risks and possible consequences of various strategies. 127 Beyond showing such a breach of duty, the malpractice client must also show cause, generally interpreted to mean that, but for the breach, the client would have had a successful outcome to the litigation. This requirement yields the “case within a case” structure of malpractice actions. The client, having shown that his lawyer was careless, must then present the original case as an element of the malpractice action, to show that he would have prevailed in the first case. Not surprisingly, leading scholars of this field tell us: “[c]ausation issues in malpractice cases are frequent and often difficult.” 128 Even with this hurdle, however, in recent decades malpractice actions against civil practitioners have grown in number and success, to the point that “the threat or actuality of malpractice liability is viewed by many scholars of the legal profession as probably the most significant regulator of lawyer behavior.” 129 On the criminal side malpractice does not and could not perform this function, largely because the standard for criminal malpractice adds two elements not present in civil side malpractice actions. In most states the criminal client/plaintiff must not only show that the lawyer was negligent, but also get his underlying conviction set aside and prove his actual, factual innocence of the charge involved. 130 These additional requirements defeat most claims. The justifications for these additional hurdles are not 123 GEOFFREY C. HAZARD, JR., SUSAN P. KONIAK & ROGER C. CRAMTON, THE LAW AND ETHICS OF LAWYERING 156 (3d ed. 1999). 124 5 RONALD E. MALLEN & JEFFREY M. SMITH, LEGAL MALPRACTICE 404-05 (5th ed. 2000). 125 Compare Lucas v. Hamm, 364 P.2d 685 (Cal. 1961) (lawyer’s failure to draft trust satisfying Rule Against Perpetuities not malpractice) with Smith v. Lewis, 530 P.2d 589 (Cal. 1975) (holding that lawyer’s failure to investigate spouse’s entitlement to community property interest in government pension was adequate basis for malpractice verdict). 126 See, e.g., Gleason v. Title Guarantee Co., 300 F.2d 813 (5th Cir. 1962) (reliance on telephone conversation with title insurer, rather than lawyer’s personally checking title malpractice); Waldman v. Levine, 544 A.2d 683 (D.C. 1988) (finding that failure to consult with medical expert before advising client to accept $2,000 settlement in obstetrical case was malpractice). 127 5 MALLEN & SMITH, supra note 124, at 524 (“Principally within the last decade, attorneys have been challenged not only for the adequacy of their investigation but for the propriety of their evaluation and advice.”). 128 HAZARD, KONIAK, & CRAMTON, supra note 123, at 166. 129 Id. at 154 (citing several studies to this effect). 130 3 MALLEN & SMITH, supra note 124, at 812-23. See, e.g., Barner v. Leeds, 13 P.3d 704 (Cal. 2000) (allowing action against public defender after client released from jail pursuant to finding of factual innocence of bank robbery); Heck v. Humphrey, 512 U.S. 477 (1994) (same conclusion under Federal Tort Claims Act). 38 convincingly articulated. 131 Some courts have referred to causation or proximate causation, stating that counsel’s incompetence could not have caused the wrongful conviction of a person who was in fact guilty—presumably on the grounds that such a conviction would not be “wrongful” 132 Other courts simply but opaquely say that public policy forbids a guilty person from recovering damages. 133 At a doctrinal level it is hard to know what to make of such decisions, whose justification not only verges on circularity but diverges from courts’ reasoning in civil litigation. We return to this topic after an examination of the clients’ specific remedies, where we see a more parallel structure. The specific remedy on the civil side is the action to set aside a judgment; on the criminal side it is typically a habeas corpus action claiming ineffective assistance of counsel. 134 Although one would not guess it from reading the relevant texts, 135 both in structure and in application the two fields are similar: parties are asked to supply a reason for thinking the judgment unreliable, and, in addition, to satisfy what might be called a “miscarriage of justice” requirement. By far the most common reason for granting relief under civil Rule 60(b)(1) is to reopen a default judgment, where the defendant has failed to answer the complaint. 136 By contrast, once the parties and their lawyers begin to engage, the courts’ threshold for upsetting judgments becomes dramatically higher. 137 Ineffective assistance cases on the criminal side follow a parallel track. Courts in these cases ask whether the lawyer’s behavior was so deficient as to deprive the defendant of 131 3 MALLEN & SMITH, supra note 124, at 813 (describing and collecting cases). 132 Claudio v. Heller, 463 N.Y.S.2d 155 (N.Y. Sup. Ct. 1983). 133 Bailey v. Tucker, 621 A.2d 108 (Pa. 1993); Wiley v. County of San Diego, 966 P.2d 983 (Cal. 1998). 134 Direct appeal lies, of course, for errors on the face of the record. In theory, those errors can include the most common of the habeas claims—ineffective assistance of counsel—but in practice such claims usually require more development of a factual record (why did defense counsel do or fail to do such and such?; how long did s/he investigate or prepare?) than the trial transcript will allow. 135 On the civil side the unreliability standard flows either from common law or its codification. Substantively, these claims allege that someone’s (usually the lawyer’s) neglect or the adversary’s fraud led to an erroneous judgment. The underlying Rule is widely viewed as codifying (somewhat unsatisfactorily) a series of decisions under such obscure common law writs as coram nobis. See 11 CHARLES A. WRIGHT, ARTHUR R. MILLER & MARY KAY KANE, FEDERAL PRACTICE & PROCEDURE 228 (2d ed. 1995). On their face, the statutory grounds look quite hospitable: “mistake, inadvertence, surprise, [and] excusable neglect,” are all listed as grounds for opening the judgment. On the criminal side, unreliability claims rest on an allegedly unconstitutional process leading to conviction. One of the most common of such claims— if only because it “explains” counsel’s failure to raise other claims—is that defendant did not receive the effective assistance of counsel guaranteed by the Sixth Amendment. In practice, courts apply these apparently divergent bodies of law in quite similar ways. To appreciate how the convergence occurs, one needs to focus on how the courts in civil cases have interpreted the “mistake, inadvertence, surprise” applications. Where the “mistake” has led to the default or non-appearance of one of the parties, the courts have been quite generous in reopening the judgment. 136 See, e.g., 11 WRIGHT ET AL., supra note 135, at 230; JAMES, HAZARD & LEUBSDORF, supra note 71, at 784 (“[R]elief from default judgments constitutes the most frequent occasion for resort to the Rule and . . . courts consistently show solicitude to these applicants.”) 137 See 11 WRIGHT ET AL., supra note 135, at 282-91. 39 “the basic elements of a fair trial.” 138 “The benchmark for judging any claim of ineffectiveness must be whether counsel’s conduct so undermined the proper functioning of the adversarial process that the trial cannot be relied on as having produced a just result.” 139 Paralleling the Rule 60 decisions granting relief because the lawyer in a civil case has failed to engage in the most elementary ways, Sixth Amendment opinions finding “per se” ineffectiveness of criminal defense focus on similar factors—counsel who were either unavailable or so restricted in their scope that they were prevented from fulfilling basic functions. 140 The focus on failed adversarial processes makes sense. Both civil and criminal processes depend on engaged counsel to yield either fair or accurate results. If a judgment, either civil or criminal, presupposes some probing of the other side’s case, verdicts lacking basic adversarial preconditions for such probing should not stand. 141 Looking beyond the nature of these failures, both bodies of law demand a showing that the lawyers’ lapses have resulted in real harm. On the civil side the question—applied by the courts but unstated in Rule 60—is whether the party seeking relief from judgment has a meritorious claim or defense. 142 On the criminal side the parallel requirement is “prejudice,” a test requiring the court to decide whether it is reasonably likely that, without the lawyers’ lapse, a different result (i.e., other than conviction) would have been reached. 143 In both instances, the courts are asking the litigants to show, besides a basic failure in the system of adversarial presentation, the likelihood that a better lawyer would have achieved a different outcome. What can one learn from this concealed set of parallelisms? First, the civil cases reveal an aspect of procedure almost submerged by the procedural changes in the twentieth century: the state’s responsibility for just outcomes. As we have argued above, civil process’s privatization has largely eliminated the sense that the state has a role in civil justice other than that of honest broker. If that role were entirely abandoned, though, there would be few cases setting aside default judgments as a result of lawyer negligence. The courts’ response would be as swift as it is conceptually clear: sue your lawyer for malpractice. Instead, we see a remarkable willingness to allow the harmed client to start over when the basic premise of party responsibility—lawyers’ 138 Strickland v. Washington, 466 U.S. 668, 684-85 (1984). 139 Id. at 686. 140 United States v. Cronic, 466 U.S. 649 (1984). See also LAFAVE, ISRAEL & KING, supra note 111, at 604 (describing the per se cases, in which “’counsel was either totally absent or prevented from assisting the accused during a critical stage of the proceeding.’”). 141 See, e.g., Peralta v. Heights Medical Center, 485 U.S. 80 (1988) (holding unconstitutional Texas statute requiring that one who had not received notice of suit and had therefore defaulted, show meritorious defense: “[H]ad [appellant] had notice of the suit, . . . he might have . . . worked out a settlement. . . . or [sold] his property himself in order to raise funds rather than suffer it sold at a constable’s auction.”). 142 See JAMES, HAZARD & LEUBSDORF, supra note 71, at 783 (describing cases under FED. R. CIV. P. 60(b), governing motions for relief from judgments in the federal system). 143 See LAFAVE, ISRAEL & KING, supra note 111, at 634. 40 engagement—has failed. This willingness reveals an interesting residual impulse to do justice, not just to conclude cases, on the civil side. Equally important, neither a Sixth Amendment ineffective assistance claim nor a civil motion to set aside a judgment require judges to decide how an ordinarily competent lawyer would have behaved. Instead they are to ask whether some serious miscarriage of justice has occurred. 144 A number of observers have argued that the courts have set Sixth Amendment standards too low—finding, for example, that a sleeping lawyer nevertheless rendered effective assistance of counsel. 145 Without entering fully into the competency debate, one can ask whether the Rules and the courts have been readier to relieve clients of the mistakes of privately retained civil counsel than of government-paid defense lawyers in similar cases. Beyond this question, the comparative exercise suggests that courts in Sixth Amendment cases might gain from looking over the remedial wall, not at the specific but at the substitutionary remedy—for civil malpractice. Increasingly the asserted deficiency in civil malpractice cases has concerned not the lawyer’s performance at trial but the advice the lawyer gives to his or her client. Given the prevalence of non- trial dispositions of criminal matters, it is fair to ask not whether the Sixth Amendment standard is too low but whether its focus on trial is too narrow. Should the “effectiveness” cases include more of the counseling function now prevalent as an element of civil malpractice? In a world in which most criminal defendants plead, it is important for them to understand what they are doing, and there are some well-developed guidelines on the civil side to which one might refer. 146 With this perspective let us return to the most striking anomaly in this area—the standards for criminal malpractice actions. Given the rough parallelism of the specific remedies, the discrepancy in the malpractice standards is striking. We do not require civil malpractice plaintiffs to prove that, factually and morally, they deserved to prevail— only that they would, in fact have won their original lawsuits. Nor do we require them to have their judgments set aside as a precondition of bringing a malpractice action. 147 How 144 Id. at 602 (“In tying the concept of effective assistance to the functioning of the adversary process, the Court clearly rejected a measurement based solely on a comparison of counsel with his or her peers.”);JAMES, HAZARD & LEUBSDORF, supra note 71, at 783 (“Rule 60(b) does not explicitly state that an applicant for relief must show anything more than the existence of one of the circumstances referred to in the Rule. . . . [T]he decisional law [however] requires [additional showings].”); Anderson v. Cryovac, Inc., 862 F.2d 910, 923 (1st. Cir. 1988) (requiring the party moving to set aside a civil judgment to “show that the misconduct [of his adversary] foreclosed full and fair preparation or presentation of its case.”). 145 Jeffrey Levinson, Note, Don't Let Sleeping Lawyers Lie: Raising the Standard for Effective Assistance of Counsel, 38 AM. CRIM. L. REV. 147 (2001) (collecting critiques of Sixth Amendment standards as applied). See also, Leonard Post, Indigent Defense Services Blasted: National Review Has Been Launched, NAT’L L.J., July 12, 2004, at 1. 146 There are standards of professional practice on the criminal side which could offer guidance, too, but they have been largely ignored by courts assessing Sixth Amendment challenges to guilty pleas. See Gabriel J. Chin & Richard W. Holmes, Jr., Effective Assistance of Counsel and the Consequence of Guilty Pleas, 87 CORNELL L. REV. 697 (2002). 147 Indeed, were the civil judgment reopened, it would often serve as a complete or partial defense to a civil malpractice action, by enabling the defendant-lawyer to show that there were no lasting consequences of his malpractice. 41 does one explain a counterintuitive barrier to the criminal defendant whose lawyer has failed him? It seems particularly indefensible to say, on one hand, that the Constitution requires the state to supply counsel for indigent criminal defendants—but that the counsel thus supplied need not meet ordinary standards of professional competence. And recall that habeas actions based on the Sixth Amendment provide no answer because, as we have seen, they ask not whether the counsel has been ordinarily competent but whether there was a complete collapse of adversarial engagement. Several possibilities suggest themselves. First, one can see the doctrine as a holdover from the days when the state not only supplied no counsel to indigents, but in fact forbade representation to criminal defendants. The rule would in this respect be analogous to the ease with which civil judgments resulting from total lawyerly collapse are set aside: both doctrines remind us of a time when the state played quite different roles in the two systems. A second perspective emerges if one bears in mind the different financial structures of civil and criminal representation. Today the public pays for most criminal felony defense. One can understand the no-malpractice-remedy cases as an indirect damage rule: because the lawyer’s employer—the state—has supplied the defense for free, the plaintiff/client has suffered no out of pocket damages from an incompetent defense. Therefore the law will recognize only claims for egregious miscarriage of justice, not just the failure of ordinary competence. In this respect the standards for criminal malpractice resemble charitable immunities that protected many providers of medical and similar services fifty years ago. 148 One story is thus of simple time lag: the legal system often takes a number of years to adapt itself to changed social circumstances. Charitable and municipal services existed for many years before courts imposed tort liability; Gideon v. Wainwright, 149 requiring government to provide free felony counsel to indigents, is barely fifty years old, and the right even to paid criminal counsel is just over two centuries old. Whatever may be the origins of this immunity of criminal counsel to malpractice liability, there are some obvious arguments that it should go the way of charitable and municipal immunity generally. It is odd that medical providers hired by the county to care for indigent patients (sometimes including prisoners) should be subject to ordinary medical malpractice standards, but that county-paid providers of legal services should not. It is particularly odd when one considers that the private criminal bar—the approximately 20% of defense lawyers paid directly by their own clients—can shelter under the same wings of malpractice immunity. As the preceding discussion suggests, we do not think there are iron-clad arguments for aligning standards in legal malpractice cases brought by civil and criminal 148 See PROSSER AND KEETON ON THE LAW OF TORTS 1069-71 (W. Page Keeton et al., eds., 5th ed. Hornbook Series Law. ed. 1984); Bradley C. Canon & Dean Jaros, The Impact of Changes in Judicial Doctrine: The Abrogation of Charitable Immunity, 13 LAW & SOC’Y REV. 969 (1979). 149 372 U.S. 335 (1963). 42 clients. 150 Our argument is rather that the reasons for the different standards deserve more careful sorting and discussion, bearing in mind the characteristic weaknesses of litigation finance on both sides of the civil-criminal divide. On the criminal side, the defense bar is chronically underfunded. Allowing convicted clients to collect malpractice judgments might, in the long run, force better public funding and higher professional standards, as medical malpractice does in county hospitals. But in the short run it would even further deplete the resources available to fund criminal defense. On the civil side, the facts of the default and excusable neglect cases give one a strong suspicion that the lawyers who have failed to investigate, blown the deadlines, or suffered the dismissal have similarly failed to pay their malpractice insurance premiums. 151 Again, the long-run remedy—forcing lawyers to insure and forcing clients more thoroughly to investigate the lawyers they hire—has its attractions. But one understands a court’s yielding to the impulse to relieve the client of the burden of professional incompetence. In both instances we see courts choosing short-run considerations over long-run structural discipline; one can quarrel with the choices, but they are surely understandable. We believe only that thoughtful confrontation of analogous situations across the courtroom hallway has the potential for making both systems stronger. III. WHERE BORROWING IS ALREADY ROUTINE We have not exhausted the areas in which criminal procedure and civil procedure each may have lessons to teach the other. It could be fruitful to explore, for example, whether there is justification for the striking differences between the central role of the jury in calculating civil damages and the almost nonexistent role of the jury in non-capital criminal sentencing. 152 But we hope we have made clear by now why we think the procedural divide should be less impassable. We hope we have also made clear that we do not favor merging the two sets of procedural rules into one; what we favor is a regular practice of cross-comparison. We seek not to tear down the wall between civil and criminal procedure, but only to broaden the opportunities for cultural exchange. Our grievance is not with boundaries but with a certain form of provincialism. 150 To the possible justifications of the existing doctrine one might add the prospect of an endless supply of convicted criminal plaintiffs, most convinced that a better lawyer would have got them off and with a good deal of time on their hands to file such suits. But if this is the underlying concern, one might forbid pro per representation of criminal malpractice actions (though not for Sixth Amendment claims, many of which are filed pro per): requiring a convicted criminal defendant to convince a lawyer to take his case on a contingency basis might provide a good working assurance of a prima facie case. 151 See Carter v. Albert Einstein Medical Center, 804 F.2d 805, 808 (3d Cir. 1986) (“Although an action for malpractice is a possibility when a lawyer’s negligence results in dismissal, that remedy does not always prove satisfactory. It may be difficult for the client to obtain and collect a judgment for damages…”); James v. United States, 215 F.R.D. 590 (E.D. Cal. 2002) (setting aside civil judgment because of uninsured attorney’s gross negligence). 152 This is an inquiry rendered all the more pressing by the Supreme Court's recent series of decisions invalidating sentencing enhancements triggered by facts found by the trial judge. See United States v. Booker, ___ U.S. ___, 125 S.Ct. 738 (2005); Blakely v. Washington, ___ U.S. ___, 124 S. Ct. 2531 (2004); Apprendi v. New Jersey, 530 U.S. 466 (2000). One of the more striking features of the ruling Booker, which struck down the federal system of mandatory Sentencing Guidelines, was how far a majority of the Court was willing to go in crafting a remedy to avoid giving federal juries a larger role in sentencing. 43 To test our thesis, it would be nice to be able to examine fields of procedural law in which borrowing across the criminal-civil divide is already routine, where there is a rebuttable presumption that the rules for criminal and civil cases should be the same. It would be nice to be able to see whether such a presumption in fact has proven useful in the ways we have suggested that it would Fortunately, there are at least two such fields: the rules of evidence and the ethical codes governing the conduct of lawyers. And although both evidence law and the regulation of lawyers remain far from perfect, there is reason to believe that each field has been strengthened rather than weakened over time by bridging the criminal-civil divide. A. Evidence Law Ever since evidence law emerged as an integrated, systematic body of doctrine in the late 1800s and early 1900s, it has treated criminal and civil cases largely alike. Indeed, evidence law to a significant extent was itself a product of treating criminal and civil cases alike. 153 James Bradley Thayer, the “American protagonist of the modern law of evidence,” 154 conceived his field as entirely trans-substantive, and specifically as transcending the criminal-civil divide. 155 His student John Henry Wigmore was even more emphatic about this point. “The rules of admissibility,” Wigmore wrote toward the beginning of his monumental treatise, “are in general the same for the trial of civil and of criminal causes. Not only in practice, but in principle and in spirit, there is no occasion for a distinction. . . . There is but one system of rules for criminal trials and for civil trials.” 156 Wigmore criticized the very use of the phrase “Criminal Evidence,” because it “tended to foster the fallacy that there is some separate group of rules or some large number of modifications.” 157 Wigmore called that fallacy “inveterate,” 158 but, if so, his influence was such that he successfully silenced it. When efforts to codify evidence law began in earnest in the mid-twentieth century, the codifiers never proposed separate sets of rules for criminal and civil cases, and apparently never even considered doing so. The American Law Institute proposed a Model Code of Evidence, not separate codes of civil and criminal evidence. The National Commissioners on Uniform State Laws developed “Uniform Rules of 153 See supra text accompanying notes 6-7. 154 Charles Frederic Chamberlayne, The Modern Law of Evidence and its Purpose, 42 AM. L. REV. 757, 758 (1908). 155 See, e.g., JAMES BRADLEY THAYER, PRELIMINARY TREATISE ON EVIDENCE AT THE COMMON LAW (1892). 156 1 JOHN HENRY WIGMORE, A TREATISE ON THE ANGLO-AMERICAN SYSTEM OF EVIDENCE IN TRIALS AT COMMON LAW § 4, at 16-17 (2d ed. 1923). 157 Id. at 17. The concept of a separate body of evidence rules for criminal trials was associated most notably, perhaps, with Francis Wharton, whose Treatise on the Law of Evidence in Criminal Issues first appeared in 1880. Even Wharton, though, had begun his treatise by explaining that “[s]ubject to exceptions to be hereafter specifically noticed, the tests for the admission of evidence are the same in criminal as in civil trials.” Id. § 1, at 1. 158 1 WIGMORE, supra note 156, at 17. 44 Evidence.” These initiatives culminated, of course, in the Federal Rules of Evidence— not in federal rules of civil evidence and federal rules of criminal evidence. Indeed, one of the arguments pressed by backers of the Federal Rules of Evidence was precisely that the new rules would help resolve inconsistencies—taken as self-evidently undesirable— between the evidence law applied in civil cases and in criminal cases. 159 On the state level as well as the federal level, evidence law today is dominated by codes that apply to both civil and criminal cases. Evidence treatises almost all follow Wigmore in treating evidence law as “but one system of rules.” Law school courses do the same. Lawyers, judges, and scholars all think of evidence law as a single set of doctrines, treating civil and criminal cases for the most part interchangeably. 160 Of course there are exceptions to this general pattern of evidence rules spanning the civil-criminal divide. Some evidence rules apply only in criminal cases or only in civil cases. 161 But these are exceptions, and they are viewed as exceptions. The background assumption, almost never challenged, is that in general the “rules determining the competency of evidence should apply across the board, whether the case is on the civil or criminal calendar.” 162 Indeed, departures from this approach frequently are criticized as unprincipled, in part because they do depart, and in part because—not coincidentally—they also tend disproportionately to be the products of unadorned politics, rather than the deliberations by the committees of judges, scholars, and 159 See, e.g., COMMITTEE ON RULES OF PRACTICE AND PROCEDURE, JUDICIAL CONFERENCE OF THE UNITED STATES, RULES OF EVIDENCE: A PRELIMINARY REPORT ON THE ADVISABILITY AND FEASIBILITY OF DEVELOPING UNIFORM RULES OF EVIDENCE FOR THE UNITED STATES DISTRICT COURTS 26 (1962) (“One of the present difficulties is that the Civil Rule and the Criminal Rule are based on divergent theories.”); PROPOSED FEDERAL RULES OF EVIDENCE 1 (John R. Schmertz, Jr., ed. 1974) (“In general, [Civil] Rule 43(a) led federal courts to follow state evidence law in the absence of a federal statute or equity practice supplying a different rule. . . . Rule 26 of the Rules of Criminal Procedure instructed the federal courts to continue to develop their own criminal evidence rules from the common law as applied ‘in the light of reason and experience.’ For many years, therefore, the federal courts applied one set of rules on the criminal side and a distinct and nonuniform set in civil actions.”). 160 The most recent edition of Wharton's Criminal Evidence begins on a strikingly defensive note: "Why write a book devoted solely to criminal evidence? Although the same rules of evidence generally apply in criminal and civil cases, some rules rarely encountered in civil cases show up routinely in criminal cases. Some rules apply only to criminal cases. Occasionally states may have a separate set of evidence rules for criminal cases. In addition, constitutional concerns abound in criminal cases and often interact with the rules of evidence." FRANCIS WHARTON, 1 WHARTON'S CRIMINAL EVIDENCE § 1.1, at 1-2 (Barbara E. Bergman & Nancy Hollander, eds., 15th ed. 1997) (emphasis added). 161 See, e.g., FED. R. EVID. 404(a)(1) & (2) (listing exceptions to character evidence ban that apply only in criminal cases); id. 709(b) (imposing special restriction on expert psychiatric testimony in criminal cases); id. 803(b)(2) (restricting hearsay exception for dying declarations to civil cases and homicide prosecutions); id. 804(b)(3) (imposing special limitations on declarations against penal interest offered by a criminal defendant). 162 Logerquist v. McVey, 1 P.3d 113, 127 (Ariz. 2000). 45 practitioners that dominate modern rulemaking, in evidence as in other fields of procedure. 163 Not all variations in evidence law across the criminal-civil divide are explicit. Judges are widely believed to apply the character evidence ban, for example, less strictly in civil cases than in criminal cases. There has long been a perception that hearsay rules, on the other hand, are applied less strictly in criminal cases, or at least in certain categories of criminal cases. 164 The same is almost certainly true of the rules restricting expert testimony 165 —although here, too, the divergence seems to be narrowing, for reasons to which we shortly will return. There are significant differences, in brief, between the evidence law applied in civil cases and the evidence law applied in criminal cases. But reports of the “virtual disintegration of a unitary evidence code” 166 are premature. Evidence law has remained unified because the rebuttable presumption has remained that rules of evidence should apply “across the board.” Special rules for civil or criminal cases, whether express or unstated, are customarily viewed with skepticism. And that skepticism can be and has been a powerful force for reform. The most striking recent example pertains to expert testimony. A decade ago the Supreme Court directed trial judges to begin screening expert testimony for “reliability,” 167 a mandate since codified in Federal Rule of Evidence 702. The new requirement was announced and elaborated by the Court in a series of civil cases, and it responded to concerns specifically focused on civil litigation—the widespread complaints about “junk science” in the courtroom. But the Court drew no distinction between civil and criminal cases. It would have been awkward to do so, because the Court justified the requirement of “reliability” as a gloss on the provisions of the Federal Rules of Evidence 163 See, e.g., Anne Lawson Braswell, Note, Resurrection of the Ultimate Issue Rule: Federal Rule of Evidence 704(b) and the Insanity Defense, 72 CORNELL L. REV. 620 (1987) (describing and criticizing legislation that added FED. R. EVID. 704(b)); Daniel J. Capra, A Recipe for Confusion: Congress and the Federal Rules of Evidence, 55 U. MIAMI L. REV. 691 (2001) (same); Eleanor Swift, Does It Matter Who Is In Charge of Evidence Law?, 25 LOY. L.A. L. REV. 649 (1992) (describing and criticizing popular initiatives that altered evidence rules in California criminal trials); Peter W. Tague, Perils of the Rulemaking Process: The Development, Application, and Unconstitutionality of Rule 804(b)(3)'s Penal Interest Exception, 69 GEO. L.J. 851 (1981) (describing and criticizing political process that resulted in the limitations imposed by FED. R. EVID. 804(b)(3) on the "penal interest" exception to the hearsay rule). 164 See, e.g., Myrna S. Raeder, Cost-Benefit Analysis, Unintended Consequences, and Evidentiary Policy: A Critique and Rethinking of the Application of a Single Set of Evidence Rules to Civil and Criminal Cases, 19 CARDOZO L. REV. 1585, 1601 (1998). The Supreme Court recent reinvigoration of the Confrontation Clause in Crawford v. Washington, 541 U.S. 36 (2004), considerably tightens the constitutional standards for introducing hearsay against a criminal defendant, but only if the hearsay is deemed "testimonial"—a term the Court has left, for now, largely undefined. 165 See, e.g., id. at 1602; D. Michael Risinger, Navigating Expert Reliability: Are Criminal Standards of Certainty Being Left on the Dock?, 64 ALB. L. REV. 99 (2000). 166 Raeder, supra note 164, at 1604. 167 Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993). See also Kumho Tire Co. v. Carmichael, 526 U.S. 137 (1999); General Electric v. Joiner, 522 U.S. 136 (1997). 46 governing relevance and expert testimony, and those rules, like most other provisions of the Federal Rules of Evidence, themselves drew no distinction between civil and criminal cases. Nonetheless lower courts, perhaps reading between the lines of the United States Reports, have assessed the “reliability” of expert testimony in civil cases much more rigorously than in criminal cases. Until quite recently, forensic science evidence— fingerprinting, ballistics, handwriting identification, hair analysis, bite mark identification, etc.—received something of a free pass, even when there was little or no proof of its reliability. 168 The double standard for the admissibility of expert testimony led some observers to grow skeptical of both the possibility and the wisdom of unified rules of evidence. 169 But the resignation was unwarranted. The formal equivalence of treatment of expert testimony in civil and criminal trials is slowly and belatedly making itself felt in the courtroom. Forensic science evidence that has gone unquestioned for decades is suddenly under scrutiny. For example, several trial judges have excluded or significantly limited expert testimony by handwriting examiners—something that virtually never happened up until a few years ago. 170 Even fingerprint examiners are being forced to testify with more care and to qualify their claims of 100% accuracy. 171 It is hard not to see this as an improvement. It is also hard to believe it would have happened so rapidly were it not for the presumption, however loose, that rules of evidence announced in civil cases should also apply in criminal trials. 172 One might think the argument for such a presumption unusually strong in the case of evidence rules. To a great extent those rules govern processes of inference, and inference is something that one might think should operate basically the same in civil and criminal adjudications. Wigmore, in fact, reasoned along just these lines. “The relation between an Evidentiary Fact and particular Proposition,” he explained, “is always the same, without regard to the kind of litigation in which that proposition becomes material to be proved.” 173 But in truth the rules of evidence do much more than seek to promote 168 See, e.g., Raeder, supra note 164, at 1602; Risinger, supra note 165; Michael J. Saks, Banishing Ipse Dixit: The Impact of Kumho Tire on Forensic Identification Science, 57 WASH. & LEE L. REV. 879 (2000). 169 See Paul C. Giannelli, Scientific Evidence in Civil and Criminal Cases, 33 ARIZ. ST. L.J. 103 (2001); Raeder, supra note 164, at 1604-06. 170 See, e.g., United States v. Fujii, 152 F. Supp. 2d 939 (N.D. Ill. 2000). 171 See United States v. Plaza, 188 F. Supp. 2d 549 (E.D. Pa. 2002); Michael Specter, Do Fingerprints Lie?, NEW YORKER, May 27, 2002, at 96. 172 Richard Lempert has suggested more broadly that, over time, the rigor with which federal courts screen scientific evidence in civil and criminal cases will converge, because the de facto standard will toughen in criminal cases and loosen in civil cases—and that both changes will be for the better. See Richard Lempert, Remarks at the 16th Conference of the International Society for the Reform of Criminal Law (Dec. 8, 2002). 173 1 WIGMORE, supra note 156, § 4, at 16; see also id. at 18 (quoting Rex v. Burdett, 4 B. & Ald. 95, 122 (1820)) ("If the rules of evidence prescribe the best course to get at truth, they must be and are the same in all cases, and in all civilized countries."). 47 accurate inferences. They protect privacy (think, e.g., of the physician’s privilege), guard against harassment (think of spousal immunity), advance judicial efficiency (think of the rules limiting proof of collateral matters), and often seem more concerned with fairness than with accuracy (think of the hearsay exception for admissions of adverse parties). 174 They share, in short, many of the goals of procedural law more broadly. None of this is to say that the Federal Rules of Civil Procedure and the Federal Rules of Criminal Procedure should be replaced with a single “Federal Code of Procedure,” paralleling the Federal Rules of Evidence. But it is to say that evidence law provides additional reason to believe that routine cross-comparison might be profitable between civil and criminal procedure more broadly. The example of evidence law is also instructive in another regard. Unified rules of evidence are very much a common-law phenomenon; they are not found in civil-law countries. 175 Indeed, they are more and more a United States phenomenon. In the United Kingdom, legislation increasingly addresses the topics of civil and criminal evidence separately, and criminal evidence is increasingly taught and discussed as a distinct subject. 176 The lack of a unified law of evidence in civil-law countries reflects the sharper disjunction generally in Continental legal tradition between civil and criminal adjudication. 177 In this respect the United Kingdom and the Commonwealth countries more generally have been converging with Europe. Much of the increasing differentiation between civil and criminal evidence in the United Kingdom reflects the decline of the civil jury, a decline seen throughout the Commonwealth. 178 More and more, then, the American unified law of evidence is something of a global oddity. That fact may make some readers justifiably wary of our proposal to make the rest of our procedural law look a little more like our evidence law. There is a risk of exchanging one kind of provincialism for another, more literal kind. As we noted at the outset of this essay, the more we narrow the conceptual divide between civil and criminal procedure, the more we widen the divergence between the American legal system and those found elsewhere. And, of course, it is possible that the rest of the world knows something that we do not. Perhaps Continental legal tradition is right to draw such a 174 See, e.g., Michael L. Seigel, A Pragmatic Critique of Modern Evidence Scholarship, 88 NW. U. L. REV. 995 (1994); cf. FED. R. EVID. 102 (directing that the Federal Rules of Evidence "shall be construed to secure fairness in administration, elimination of unjustifiable expense and delay, and promotion of growth and development of the law of evidence to the end that the truth may be ascertained and proceedings justly determined"). 175 See DAMA?KA, supra note 94, at 10, 11 n.9. 176 See WILLIAM TWINING, RETHINKING EVIDENCE 203, 208 (1990); A.A.S. ZUCKERMAN, THE PRINCIPLES OF CRIMINAL EVIDENCE (1989) 177 See DAMA?KA supra note 94, at 105-20. 178 See id. at 109 n.66; ZUCKERMAN, supra note 176, at 3. Wholly apart from its implications for evidence law, the jury is itself an important site of convergence between the legal systems of civil-law countries and those in the Commonwealth nations, and one that underscores the deep divide in civil-law tradition between criminal and civil adjudication. The Commonwealth has been jettisoning civil juries just as Europe has begun to experiment with criminal juries. 48 sharp divide between civil and criminal adjudication. Perhaps the gradual drift of American law in that direction should be celebrated rather than deplored. Perhaps. But we think it at least equally likely that our own legal tradition, which for centuries saw civil and criminal adjudication as closely related processes of dispute resolution, has significant strengths that we should hesitate to discard. And our Constitution gives the civil jury a much more secure foothold here than elsewhere in the world. That alone ensures that civil and criminal trials will share more similarities here than elsewhere and powerfully limits the possibilities for convergence with other legal systems, regardless how we develop our procedural rules. For the foreseeable future, then, the most fruitful opportunities for cross-fertilization of our procedural systems will probably be found in our own backyard. B. Regulation of Lawyers Legal ethics, like the rules of evidence, treats civil and criminal cases largely alike. The ABA’s Model Rules of Professional Conduct, for example, draw few distinctions between criminal and civil litigation. One provision of the Model Rules does set forth “special responsibilities of a prosecutor,” 179 and a handful of other provisions single out criminal defense attorneys. Unlike civil litigators, criminal defense attorneys may not charge contingent fees; 180 criminal defense attorneys, but not civil litigators, are explicitly authorized to “defend [a] proceeding as to require that every element of the case be established.” 181 These are isolated exceptions, though. For the most part, the Model Rules treat civil and criminal lawyers the same. Obligations of competence, diligence, communication, confidentiality, candor, professional independence, impartiality—in short, the whole, interrelated set of ideas about what a lawyer is supposed to be—apply across the board. And in this respect the Model Rules are typical of modern rules principles of professional responsibility. “With minor exceptions, most professional codes do not differentiate between lawyers’ roles in criminal and civil cases.” 182 This remains the case despite persistent and spirited suggestions that civil litigators and criminal defense attorneys play very different roles and may appropriately be subject to very different obligations. 183 These suggestions persist because they are so 179 MODEL RULES OF PROF’L CONDUCT R. 3.8. (2004). 180 Id. R. 1.5(d)(2); see Pamela S. Karlan, Contingent Fees and Criminal Cases, 93 COLUM. L. REV. 595 (1993). 181 MODEL RULES OF PROF’L CONDUCT R. 3.1. (2004). 182 Fred C. Zacharias, Reconceptualizing Ethical Roles, 65 GEO. WASH. L. REV. 169, 169 (1997). 183 For particularly influential arguments along these lines, see DAVID LUBAN, LAWYERS AND JUSTICE: AN EMPIRICAL STUDY 59-66 (1988) [hereinafter LUBAN, LAWYERS AND JUSTICE]; Monroe H. Freedman, Professional Responsibility of the Criminal Defense Lawyer: The Three Hardest Questions, 64 MICH. L. REV. 27 (1966); David Luban, The Adversary System Excuse, in THE GOOD LAWYER: LAWYERS' ROLES AND LAWYERS' ETHICS 83 (David Luban ed., 1983) [hereinafter Luban, The Adversary System Excuse]; Murray L. Schwartz, The Zeal of the Civil Advocate, 1983 AM. B. FOUND. RES. J. 543. 49 plausible. They draw on the same intuition underlying the rigid separation of civil procedure and criminal procedure as fields of jurisprudence and scholarship: the notion that criminal trials and civil trials differ fundamentally, not only in their operation but also in their ends. 184 Thus it is said that criminal trials aim to protect the individual from the state, whereas civil trials serve only to decide which of two “civilians” is entitled to control assets. 185 Just as this distinction justifies a host of other procedural differences between civil trials and criminal trials, so also—the argument goes—it justifies allowing (and perhaps requiring) criminal defense attorneys to engage in forms of super-zealous advocacy that are properly placed off-limits to civil trial lawyers. Such forms of advocacy are sometimes said to include, for example, suggesting that a witness is lying when the lawyer knows the witness is telling the truth, or cooperating with a decision by the client to commit perjury. Even if civil litigators are forbidden to engage in tactics like these, the tactics have been defended as permissible (and even admirable) when carried out by criminal defense attorneys, on the ground that the latter, unlike the former, are participating in a process aimed not at “legal justice” but rather at “the preservation of the proper relation between the state and its subjects.” 186 This argument for separating criminal litigation ethics from civil litigation ethics has been properly rejected, and the grounds for the rejection are instructive. One reason for the continued joinder of criminal and civil rules of professional ethics, of course, may be simple inertia. But there has been another, more principled reason: a sharp functional divide between civil and criminal trials has been found, correctly, to be untenable. Both serve multiple purposes, those purposes overlap, and the “prime function” in each case can persuasively be characterized as the same: “‘doing justice’ in the course of settling disputes.” 187 Civil sanctions can be just as harsh and just as stigmatizing as criminal sanctions. 188 And criminal defendants cannot credibly be characterized as uniquely vulnerable. 189 184 See Schwartz, supra note 183, at 548, 556. 185 Id. at 554; see also LUBAN, LAWYERS AND JUSTICE, supra note 183, at 59-60, 63; Luban, The Adversary System Excuse, supra note 183, at 91-92. 186 Luban, The Adversary System Excuse, supra note 183, at 92; see also Schwartz, supra note 183, at 553-54. 187 Donald T. Weckstein, The Civil Advocate and the Multifaceted Functions of Dispute —Some Domestic and Cross-Cultural Perspectives, 1983 AM. B. FOUND. RES. 577, 579-80. It seems in retrospect particularly hard to defend the notion that criminal trials serve only to defend individuals against the state, and that "[i]t inflicts no tangible harm on anyone when a criminal evades punishment," Luban, The Adversary System Excuse, supra note 183, at 92; cf. Schwartz, supra note 183, at 553 (suggesting that "the basic purpose" of criminal procedure is "to avoid one type of error"). As William Simon has noted, this is a picture in which "victims do not appear." See William C. Simon, The Ethics of Criminal Defense, 91 MICH. L. REV. 1703, 1708 (1993). 188 Weckstein, supra note 187, at 579. See also Fred C. Zacharias, The Civil-Criminal Distinction in Professional Responsibility, 7 J. CONTEMP. LEGAL ISSUES 165, 172 (1996). 189 See Simon, supra note 187, at 1707; Zacharias, supra note 188, at 177. 50 This is not to deny that there are rough differences, on average, between criminal cases and civil cases—nor that ethics opinions regarding the boundaries of zealous advocacy may properly have, on average, a different feel in criminal cases than in civil cases. 190 But it is to say that the differences within the two classes of cases are often greater than the differences between them—sufficiently so that it is unhelpful to think of “criminal trial ethics” and “civil trial ethics” as separate things. One advocate of separate ethical rules for criminal defense attorneys has tried to save the argument by distinguishing not between criminal and civil trials per se, but rather between trials that fit within the “criminal defense paradigm” and those that fit within the “civil suit paradigm.” 191 The more he has thought about it, the more abstract these categories have become. Initially he called for treating “some noncriminal matters, such as administrative hearings” the same as criminal cases. 192 He later expanded the “criminal defense paradigm” to include “any litigation in which zealous advocacy on behalf of relatively weak clients” is necessary to protect them from “powerful institutions,” and suggested that “certain quasi-criminal matters, such as government antitrust suits and hearings before various regulatory agencies” should either be treated under the “civil suit paradigm” or “assimilat[ed] to the criminal defense paradigm, with the private entity in the role of the state and the government in the role of the private party.” 193 Still later he suggested similar treatment for rape prosecutions, on the ground that these cases pit the state against the “powerful institution” of “patriarchy.” 194 It is hard not to feel that this last move “proves too much.” 195 If a defense attorney in a rape prosecution cannot invoke the “criminal defense paradigm,” it may be time to change labels. But there is a good deal to be said for the underlying effort to recognize the differences between criminal and civil trials while allowing the rules governing each of these categories of litigation to borrow from the rules governing the other. This is precisely the enterprise that we advocate for the rules of civil and criminal procedure more broadly. And in the field of legal ethics, as in the field of evidence law, we think the rebuttable presumption that rules should apply across the board has led to deeper and more systematic thinking about what the rules should be, and how, if at all, they should vary according to context. 196 190 Compare, e.g., In re Complaint as to the Conduct of A., 554 P.2d 479 (Or. 1976) (suggesting that civil lawyers may properly be disciplined for knowingly eliciting truthful but misleading testimony), with Mich. Ethics Op. No. CI-1164 (1987) (advising that a criminal defense attorney may properly elicit misleading testimony so long as it is truthful), available at http://www.michbar.org/opinions/ethics/ numbered_opinions/ci-1164.html. 191 LUBAN, LAWYERS AND JUSTICE, supra note 183, at 65-66. 192 Luban, The Adversary System Excuse, supra note 183, at 92-93. 193 LUBAN, LAWYERS AND JUSTICE, supra note 183, at 65. 194 David Luban, Partisanship, Betrayal and Autonomy in the Lawyer-Client Relationship: A Reply to Stephen Ellmann, 90 COLUM. L. REV. 1004, 1028-30 (1990). 195 Simon, supra note 187, at 1710. 196 When, for example, the Ethics Committee of the Michigan State Bar advised a criminal defense attorney that the attorney could properly elicit testimony that was truthful but misleading, the committee 51 It warrants reiteration that we are not arguing that the rules should be the same in criminal cases and civil cases, only that differences should be explained and continually reexamined. The field of legal ethics, in fact, illustrates not only the virtues of thinking across the criminal-civil divide, but also the hazards of forgetting that sometimes the rules on the two sides of the divide should differ. For the rules of professional ethics applicable to civil litigators generally apply not only to criminal defense attorneys, but to prosecutors as well. 197 And here the refusal to recognize the differences between criminal and civil litigation has sometimes threatened absurd results. Civil litigators, for example, are forbidden to communicate, or have their agents communicate, with parties represented by other lawyers, except with the knowledge and permission of the other lawyers. Applying this rule in its full rigor to prosecutors could immunize any criminals wealthy enough to keep attorneys on retainer from any covert or overt investigatory contacts. Nonetheless courts have been markedly hostile to the suggestion that the rule should not apply to prosecutors—not because the rule itself seems sensible as applied to prosecutors, but rather because it has seemed unacceptable to exempt prosecutors from the ethical rules governing other lawyers. 198 Part of the problem has been the failure to develop meaningful and workable ethical principles for the distinctive tasks carried out by prosecutors, including in particular the exercise of charging discretion. 199 Forgetting the differences between criminal and civil litigation is no wiser than ignoring the similarities. In the main, though, our systems of criminal and civil procedure suffer from too little cross-comparison, not from excessive uniformity. And the field of legal ethics, like the field of evidence law, offers a useful model for greater cross-comparison. CONCLUSION One can tell two simple stories about criminal and civil process and how they have changed over the past century or so. One story is a contrast between petrification and metastasis. Criminal procedure appears to be in a state of arrested development. It is frozen roughly into the shape it had in 1800. In many jurisdictions the preliminary hearing has been added—but not assimilated into the structure. By contrast, civil procedure has gone into hypertrophy. The subject of multiple waves of change, its shape and general features would scarcely be recognizable to a lawyer from the early Republic. In the process it has gained some interesting features but threatens to become entirely unmoored from the rest of the legal system. felt obliged to stress the distinctive challenges placed on criminal defense attorneys by the absence of interrogatories and depositions. See Mich. Ethics Op. No. CI-1164, supra note 190. 197 For federal prosecutors this is now a matter of statute. See 28 U.S.C. § 530B (2004). 198 See, e.g., Rory K. Little, Who Should Regulate the Ethics of Federal Prosecutors?, 65 FORDHAM L. REV. 355, 361-62, 367-68 (1996); David A. Sklansky, Starr, Singleton, and the Prosecutor’s Role, 26 FORDHAM URB. L.J. 509, 537-38 (1999). 199 See Sklansky, supra note 198, at 530-38. 52 A second story contrasts the increasing privatization of civil process with the increasingly public status of criminal process. Civil litigation has privatized most of the costs of civil disputing, and with the costs most of the control. Criminal procedure, by contrast, has socialized substantially all of the costs of criminal investigation, prosecution, and defense. Both developments have entailed losses. Civil litigation reform often involves the effort to recapture some of the benefits of neutral state supervision—without any of the associated costs; not surprisingly, most such efforts don’t work. Criminal procedure threatens to become a form of administrative regulation, of intra-governmental regulation, like civil service or government procurement rules. Both stories lack nuance. And they ignore the clear triumphs of the American legal system, including its remarkable ability to hold the powerful to account while treating the weak with some respect. But these assessments also constitute an invitation. Precisely the contrasts between the two systems—both of which are distinctively native growths—enable us to ask more fundamental questions. Can we use the divide to examine afresh the features of each system, with openness to borrowing and reconsideration? Criminal procedure is not only about punishment; civil litigation is not only about dispute resolution. Both systems embody conflicting impulses and enough cultural and historical layers as to make any sentence that begins “The goal of X is . . .” either vacuous or demonstrably false. Still, convergence of the two systems would submerge much that is not only distinctive but inevitable and valuable in each. Nor do we believe that the two systems constitute an old curiosity shop in which the seeker can browse idly until one finds an odd procedure or two to consider taking home. The process of comparison and selection will not yield much unless it is more methodical and thoughtful than random shopping. What is called for is an independent and fluid re-examination of existing practices, using the other system as a point of perspective and leverage. Professionals tend to live in intellectual silos: having acquired over many years a good deal of knowledge, we also find that it is easy to take the world defined by that knowledge as given and inevitable. The value of a comparison from a place close to home is that deprives us of the too-easy dismissal of comparisons from far afield. We cannot say, “Of course, that’s interesting, but we have a Constitution and the jury system that prevent us from considering such a thing.” The nearness of the comparison is an advantage, and it is that advantage whose exploitation we hope to stimulate. Each system allows us a broad perspective on the other. The sparse public formality of the criminal system allows us to wonder whether the increasing privatization of the civil system is all to the good. The elaborate and symmetrical pretrial development of facts on the civil side let us ask whether we have too quickly assumed that defense participation in criminal discovery would be fatal to the system. Conversely, the ability of the criminal system to function as well as it has without crippling discovery lets us ask whether further pruning on the civil side might yield good results. More technically one can ask whether the rules of former adjudication—including those of double jeopardy— ought to take account of changes in the nature of adjudication. In each of these areas, 53 many of the differences between civil and criminal procedure may be fully justified. But evidence law and the rules of professional ethics both suggest the advantages of a certain degree of skepticism in this regard. Over the past century civil and criminal processes have steadily drifted apart. One increasingly resembles private enterprise, the other looks more and more like government bureaucracy. The divergence threatens to submerge part of each system. The element of private redress in the criminal system is in danger of disappearing as criminal law becomes the exclusive preserve of the government. The public values represented by the substantive law are jeopardized by the disappearance of civil adjudication from public view. Like real submersion of solid objects, these metaphoric submersions create secondary effects. One can see the rise of the victims' rights and restorative justice movements as reactions to increasingly exclusive governmental control of the criminal process. On the civil side, one can see the growth of punitive damages in a similar light—as plaintiffs seek through the civil system the kind of public vindication they might once have pursued as private prosecutions. The severance of criminal and civil process also entails undoubted gains. Each “federalized” system gains freedom to experiment without having to accommodate the other. But, as the federal image should suggest, extreme developments threaten the value of the larger enterprise. The invitation extended by this essay is not that a particular change ought to be adopted in either system, but that the two systems maintain a healthy and respectful awareness of each other—and sufficient humility to look to each other for insight and guidance.