62 Wash. & Lee L. Rev. 1637 Washington and Lee Law Review Fall, 2005 Symposium: Have We Ceased to Be a Common Law Country? A Conversation on Unpublished, Depublished, Withdrawn and Per Curiam Opinions Panel II: Moral and Ethical Dimensions of the Controversy *1637 JUDGES AS TRUSTEES: A DUTY TO ACCOUNT AND AN OPPORTUNITY FOR VIRTUE Sarah M.R. Cravens [FNa1] Copyright (c) 2005 Washington and Lee University School of Law; Sarah M.R. Cravens Ever in the making, as law develops through the centuries, is this new faith which silently and steadily effaces our mistakes and eccentricities. I sometimes think that we worry ourselves overmuch aboutthe enduring consequences of our errors. They may work a little confusion for a time. In the end, they will be modified or corrected or their teachings ignored. The future takes care of such things. [FN1] The title of this symposium asks whether we have ceased to be a common law country and proceeds to tie that question to the issue of publication of judicial opinions. Although an answer to that question depends a great deal on an understanding of what it means to be a "common law country," I will begin by saying that if we have not already ceased to be a common law country, we soon shall, unless there is a significant shift in the norms for production of judicial decisions. The related question I shall pursue in this essay is what might be one of the advantages of changing course toward a more perfect common law system, insofar as that change would involve the practice of publication of opinions. In this essay, I propose that a required practice of publication of all judicial opinions (as I shall define that practice below) would both better satisfy the judicial duty to account for management of the common law and provide a greater opportunity for the flourishing of judicial virtue and the revitalization of a meaningful common law tradition. It is important to note at the outset that, in light of the recent controversy over proposed Federal Rule of Appellate Procedure 32.1 and my own experience as a clerk for a federal appellate judge, the remarks in this paper are limited to the publication practices of federal appellate courts. There is certainly much of interest to be said about publication of opinions and the maintenance of a common law system with reference to the district courts and *1638 the Supreme Court and, indeed, about how the publication of opinions at those levels plays into the conversation at this symposium about appellate opinions, but all of that must be left to another occasion. [FN2] It is similarly important to set out from the beginning the meaning of the term "publication" as it is used throughout this piece. In using this term, my primary concern is not with the form in which opinions are made available-this is not a matter of paper versus electronic issuance of opinions. I use the term "publication" as a convenient shorthand to describe a proposed practice of written explanations (of whatever length) of judicial decisions that are made available to all members of the public (in whatever format) and come without any restraint on their future use or citability. As I will go on to discuss at greater length below, a requirement of this kind of publication of all judicial opinions presents an advantage that begins by resolving a problem of perception but has the potential to do much more for the substance of the common law tradition. There are, of course, many interpretations of what it means to be a common law system and even more theories about the social functions of the law in general and judging in particular. This essay will not attempt to solve those larger questions. Instead, it will focus on one social function of judging-that is, to maintain and improve the integrity of the corpus of the common law through the exercise of judicial virtue [FN3]-and it will attempt to elaborate on how publication may help maintain our common law system in its best state by allowing opportunities for virtue to flourish. [FN4] To answer the question posed by the symposium and the related question posed in this paper, it is necessary to provide at least a sketch of what makes up *1639 the corpus of the common law system at issue. For the purposes of this paper, I take the corpus of the common law to be the embodiment of the community's shared values as expressed in the full array of specific applications of the law by judges. [FN5] The corpus is primarily the laws and decisions themselves, but includes and is knit together by the reasoning supporting the decisions and the values underlying the laws. As Gerald Postema has much more eloquently expressed it: "Law is a framework of practical reasoning that anchors public justification of decisions and actions to past communal decisions and actions." [FN6] Judicial opinions in a common law system should make clear how and why the substantive holding or the application of the law in a given case fits into the corpus. That may mean something as simple as explaining why a particular statute applies or why a particular line of precedents is binding, or it may be as complex as explaining how and why an apparently relevant line of precedents would not be properly applied to a given case but how instead an entirely different conceptualization of the problem is more appropriate. Though only holdings are precedential in a common law system, the reasoning behind those opinions is what makes the corpus cohesive. The common law is an entity with more depth and complexity than a list of rules-it is a more intricate and open-textured web. The so-called "thick judicial virtues" to be discussed below play an essential role in holding the strands together with integrity. Publication and citability along the lines suggested in this essay would make possible a continual and mutually reinforcing cycle of virtuous judging and improvement of the strength and understanding of the common law tradition. Without a change in this direction, the current common law system risks soon becoming fairly meaningless. To turn the focus from the corpus itself to the judges who work with it, the interpretation of the common law ideal of judging that lies at the foundation of this essay relies on an understanding of public ownership of the law. In this model, the common law is (metaphorically) the property of the public. The public has entrusted the care and maintenance of the corpus of the common law to the judiciary. As trustees of the law, judges have an obligation to uphold and maintain the corpus in individual cases in accordance with the underlying aims *1640 of the corpus. This also incorporates an obligation to maintain its integrity. [FN7] This obligation requires judges to take an involved role in ensuring the proper development of the law, rather than acting as mere umpires between the parties. [FN8] The practice and significance of publication of opinions bears directly on this obligation to maintain the clarity and integrity of the corpus. As trustees, judges must be held accountable for their management of the corpus. At a threshold level, there is what might be called a "thin" kind of accountability. "Thin accountability" refers, in part, to the transparency needed in order for the public to assure itself that the trustees are carrying out their obligations. This aspect of accountability is largely about perception and confidence on the part of the public. [FN9] When judges fail to account publicly for their opinions or make them public but uncitable in future instances, they undermine themselves as trustees. They shake the confidence of those who put the law in their care by suggesting that some opinions (and possibly, by extension, the decisions themselves) are less legitimate than others. This practice raises doubts about the management of the corpus, thereby undermining the institution of adjudication by introducing instability. [FN10] Another aspect of thin accountability comprehends the basic but essential qualities associated with judicial independence, incorruptibility, and basic adherence to the law. Thin accountability is a serious concern, but still not the one on which this essay focuses. For the purposes of this essay, I am more interested in establishing a "thick" judicial accountability. That is, not the issues of perception, compliance with a code of conduct, or even judicial candor, but the more substantial aspect of what a judge is supposed to be doing as trustee of the law. It is not about a simple recitation of actions taken, but about fulfillment of a deeper obligation. *1641 Thin accountability is what keeps a judge from reflecting her personal values in her decisionmaking. Thick accountability is an accountability to the tradition itself-an expectation of virtuous judging. Thick accountability ensures that judges reflect the values of the corpus. It is not so much about the substance of those values as it is about how those values come into the method of reasoning. That is, the judge's reasoning must continually be tied back to the corpus-with integrity and fit. [FN11] The constraint on the judge is the integrity of the corpus. This is where I find it useful to look to the virtue ethics tradition. An individual member of the public may care very little about the personal character of judges, but surely ought to care about whether judges reach just outcomes. Judicial virtue is not so much about the individual character of a judge per se, but is rather an avenue for reaching a just outcome. Judges uphold the tradition of the law (of justice) by exercising judicial virtue. It is this role that should more deeply concern a member of the public observing the trustee. A thin accountability without real judicial virtue is possible, but the ultimate goal ought to be thin accountability and thick judicial virtue-that is, an accountability for justice-in combination. To understand this thick accountability and how it intersects with judicial virtue, some background is in order. Virtue ethics is a theory in moral philosophy that takes its roots in the work of Aristotle, but has been more recently developing as an alternative to the more standard utilitarian or deontological theories of moral philosophy. [FN12] Virtue ethics takes as its goal the achievement of a certain kind of human excellence. It is currently being applied to normative legal theory to create a field of "virtue jurisprudence." [FN13] In the Aristotelian model, judicial virtues are entwined with a concept of justice. Judicial virtues are those active conditions required to render a virtuous (or just) decision. For Aristotle, one who makes use of virtue in relation to someone else does justice. That is, the result is justice, insofar as it relates to someone else. The active condition or characteristic standing on its own is virtue. [FN14] Justice *1642 consists of possession (and employment) of all virtues in combination. Aristotle's concept of justice as presented in the Nichomachean Ethics (particularly in Book Five) is a concept of balance. It is a balance between the extremes of excess of and deficiency in certain qualities and a balance that seeks an equalization of positions between parties. Aristotle distinguishes between distributive justice (seeking justice by balancing the distribution with merit) and corrective justice (seeking justice by equalizing parties' positions), [FN15] but both require judicial virtue in the balancing. There are, in Aristotle's vision, situations in which the law does not provide an answer and, indeed, situations in which the law provides a wrong or absurd answer. [FN16] In those situations, virtue will guide the judge to the just outcome through equity. [FN17] Justice, in this view, can be divided into the lawful and the fair. There might well be disagreement about the specific items on the list of judicial virtues, but there would likely be general consensus about some. Professor Solum has usefully divided the virtues into "thin" and "thick" judicial virtues. [FN18] The "thin" virtues are those he sets out as uncontroversial-incorruptibility and judicial sobriety, civic courage, judicial temperament and impartiality, diligence and carefulness, judicial intelligence and learnedness, craft, and skill. [FN19] Adopting this attractive distinction between "thin" and "thick," these "thin judicial virtues" are among the aspects I would incorporate into the expectations of my notion of "thin accountability." On the other hand, Solum's "thick judicial virtues" include somewhat broader and/or more abstract qualities or conditions, including the virtue of justice (from Aristotle's account of laws as nomoi comes the concept of the judge as "nominos" or, as Solum has it, "someone who grasps the importance of lawfulness and acts on the basis of the laws and norms of her community") [FN20] and the virtue of practical wisdom (from Aristotle's account of justice as fairness and from his focus on "epieikeia" (equity), Solum suggests a model of the judge as "phronimos" (one exhibiting fairness, wisdom, and prudence in judging)). [FN21] In Aristotle's model, virtues are developed through practice; they are not qualities or conditions innately possessed. Virtues are closely tied to the actions of the individual who exercises them. I propose that the more a judge is *1643 required to account publicly with written reasoning not only for why an opinion is legally correct but for why it is practically wise and just, and the more the judge is further required to incorporate that account into the larger tradition, the more the judge will be forced into the practice of regularly exercising judicial virtues and judicial virtues (or virtuous judging) will in turn flourish. "Nominos," as a thick judicial virtue, is about legal correctness in the longer view. "Phronimos," or practical wisdom, on the other hand, is about fairness-it is relevant in contexts in which a "legally correct" result would lead to an absurd and unintended consequence. Perhaps even in a context in which more than one outcome may be correct, the virtue of practical wisdom would lead the judge to choose the better outcome to preserve or improve the integrity of the corpus. To borrow from Dworkin, the virtuous judge in such a situation would, like Judge Hercules, search out the best "fit." [FN22] Aristotle proposes (and I would agree) that no set of regulations or decrees can do justice in every case. [FN23] The law will always be underdetermined to some degree. The thick virtues take decision-making beyond simply the legally correct. They entail a view of the laws as being like the Greek "nomoi" (that is, law that incorporates not just specific regulations, but deeper values). They take a longer view for greater anchoring and stability of the law. The common law may be seen as a record of the continual evolution and improvement of the integrity of the law and of our understanding of how law and virtue are intertwined. Judicial virtue (or virtuous judging, to put it another way) is intertwined with the virtue that is a part of the corpus, as expressed in virtuous decisions. The virtue found in the corpus itself is, on the one hand, the result of judges exhibiting judicial virtues and making virtuous (just) decisions. On the other hand, the virtues the judge is supposed to exercise arise out of the values to be found in the corpus. The expectation of the judge as trustee is that he will further the concept of justice that is found in the corpus itself. It is out of the values the common law is attempting to further that we develop our idea of what the judge is supposed to be doing. Thus, virtuous judging is a component of a larger cycle that defines, develops, and achieves justice. The term "virtue" may sound as though it relies too heavily upon the individual moral character of the judge. However, the moral character at issue is not necessarily personal to the judge. The moral character the virtuous judge employs is that on which the community has collectively agreed by establishing its corpus of laws in a particular way-it is specific to the role of the judge, not to the judge personally. Thus, I do not follow Aristotle's view of virtue *1644 completely, as I do not believe judicial virtue to be dependent on the personal character or morals of the individual who is judging. I view this kind of judicial virtue as required only when that individual undertakes the role of judge. I do not subscribe to an idea of the unity of virtues. An actor may, and I believe often does, exercise different virtues or vices in different roles. [FN24] Furthermore, in my view, thick accountability is not so much about the substance of the values per se, as it is about the activity of reasoning that those values require-that the reasoning must be continually tied back to the integrity of the corpus. However, the benefit of this thick accountability is largely lost if all of that reasoning is not made available and useable. Thick accountability can only be fully achieved if the accounts are published and given equal weight and an equal part to play in the tradition, without any limitation on either availability or future citability. This substantive legitimacy of judicial reasoning is essential. What Professor Luban describes as "reasoned elaboration and visible expression of public values" [FN25] is essentially what I see as a requirement for a thicker and more complex kind of accountability-that is, for thick judicial virtue in a meaningful common law system. [FN26] Along with many others, I believe it is essential in a common law tradition that judges do not make decisions in isolation but as a part of an interpretive tradition. [FN27] It is that tradition, not only of law, but of virtue, that I take as the foundation of the theory I explore in this piece. Virtue has to be linked to a community and its traditions, or it has no meaning and no value. A virtue must be understood by the community to be a virtue. As Alasdair MacIntyre and others have written before, what those virtues are for a given community is best understood through resolution of *1645 conflicts in a common social institution. [FN28] I would argue that our judges, as those in the position (and habit) of resolving conflicts in the common social institution of the courts, are well positioned to understand the virtues, just as they are well positioned to understand what the common law is, through the same mechanism of resolution of conflicts ina common social institution. Much, if not all, of the value in this understanding, however, is wasted if it is not made common to the community from which and for whom it was determined. Even if it is available for review but unavailable for citation, that second class status delegitimizes it and robs it of its value as an integral part of the community's understanding. Virtue is linked to tradition within a community. To be an integral part of the community's identity, there must be a sense of integrity and permanence to the maintenance and the development of a community's conception of virtue. In the same way, to be an integral part of the community's corpus of law, there must be an integrity and permanence in the way that corpus is maintained and developed. Dworkin's chain novel model is one useful analogy here. [FN29] Another analogy is Cardozo's vision of the silently evolving shape of the law and how that is a part of the integrity judges are supposed to uphold. [FN30] For a common law society, it is essential to have a theory of law that allows for continual improvement and part of that process is acknowledging and accepting that there will be errors along the way and that there is no benefit to be had in hiding them. The benefit can only be achieved and appreciated with the legitimacy and usefulness to be derived from published and citable judicial opinions. Further, this thick accountability, in combination with a requirement of citable publication, requires that a judge always be accountable to herself as well, so that in future cases she must consider both her previous decision and the reasoning behind it for the decision to become a fully legitimate part of the corpus. It is the judge's duty, as a trustee, to maintain the integrity of the corpus of the common law, and the way to do that is to exercise judicial virtue. I propose that a judge cannot engage in fully virtuous judging and thus cannot entirely fulfill her obligation of accountability as a trustee of the law unless she situates her reasoning in the community's tradition of virtue, because that tradition of virtue is the underlying or overarching aim of the law. To fully accomplish that aim, the judge must openly explain her virtuous reasoning so that the community understands it and so that it may be legitimately integrated into the *1646 tradition. Judicial reasoning that is merely available to look at, but unavailable for later citation, does not fulfill the obligation of the trustee because the authority or legitimacy of the decision, and the reasoning behind it, is undermined by both the perception and the reality of its second-class status. As a result, it will not be integrated into the tradition of the common law or the broader tradition of virtue underlying it. Being conveniently placed to divine, from the conflicts before them and the body of the common law they review, a community's conception of virtue, and having the authority to interpret and apply the common law to those conflicts, judges cannot be satisfactory trustees-cannot be truly virtuous judges-without making both public and legitimate the reasoning on which they rely. This is not just a matter of perception-the way these unpublished opinions are used (or rather, not used) risks a real disservice to the corpus. If the opinions are correct-and I have not yet been convinced of any legitimate reason why they should not be correct (at least any more so than published opinions), although it must be admitted that in the current practice they often may not be correct-the corpus should benefit from them. Opinions are a public good that comes from adjudication. [FN31] With different classes of opinions, however, adjudication is devalued because the public good is lost. One might criticize my approach from the outset on the grounds that it is simply unrealistic to expect every judge in every case to undertake to achieve greatness in this manner. This would indeed be a fair and accurate criticism, if I intended such an expectation. I do not, so I must explain some limitations on this requirement. First, it is important to accept that the law will always be underdetermined to a certain extent. It would be unreasonable to expect that every last detail and nuance could be authoritatively raised, resolved, and neatly tied up by any judge in a single opinion. Furthermore, it would be unreasonable to expect a judge to write a brilliant treatise encompassing all possible relevant points when most cases are fairly straightforward. The marginal value of such an undertaking would be next to nothing. Similarly, it should be noted that the requirement proposed here would not entail a reassessment of every relevant case to have gone before simply because they are all of equal weight, but rather would assume judges would continue in their usual (one hopes) practice of citing to the most recent, the strongest, the most accepted case in a line of precedents, and would allow themselves to narrow their discussion to those cases in which the nuance of the reasoning is most in line with the issue they are presently addressing. [FN32] There would be no time for *1647 such ambitious undertakings as reviewing every case conceivably relevant or attempting in every case to write a definitive treatise on the law, nor would there be any reasonable justification for such attempts. It is not unreasonable, however, to propose an expectation that judges will produce in every case some piece of writing giving some explanation of the court's reasoning. [FN33] This reasoning need not be an exhaustive account of the relevant area of law and how it fits into the corpus. In very hard or new cases, of course, something so elaborate might be the ideal, but the proposal here is simpler. When restricted to finding a realistic approach, we should choose an approach that pushes judges towards, and not away from, the ideal. To require publication as defined above tends to force the court toward an ideal of virtue and integrity for the corpus. To allow unpublication tends to permit less virtuous judging and therefore tends away from the integrity of the corpus. The corpus, in its ideal state, embodies justice. It is this ideal of justice that the common law should be structured to aim for. As Holmes put it over a century ago, "a body of law is more rational and more civilized when every rule it contains is referred articulately and definitely to an end which it subserves, and *1648 when the grounds for desiring that end are stated or are ready to be stated in words." [FN34] There are several objections one might raise to the proposal set forth in this essay. A few have been briefly addressed already, but a few more merit acknowledgement and response. I have not yet, for example, addressed what is often the most loudly voiced objection from the judges themselves. Some judges object that they are already overworked. Though (as discussed in more depth by others at this symposium) many judges do not necessarily agree with this response, one solution to this practical difficulty is to fill judicial vacancies with greater dispatch and to further increase (and, in turn, quickly fill) the number of seats on appellate benches. To the extent that judges fear a requirement such as the one presented here would wildly expand the work they already do, I would hazard that the bulk of the work behind what I suggest is already taking place. The additional step of equalizing the status of the opinions produced does not (or at any rate should not) require additional work. It requires only that judges dismiss their worry about every word that might be cited back to them. In those instances in which an appellate court might otherwise have produced only a statement of the disposition rather than even an unpublished opinion, this proposal would add the work of providing some amount of written explanation. However, this explanation need not be a work of literary genius. If a judge can explain a decision aloud, as she might do to one of her law clerks, for example, and transcribe and edit this explanation as needed, the work will be done. Indeed, while acknowledging certain objections to the involvement of other court employees in researching, drafting, and editing, [FN35] as long as the judge has ultimate authority over the decision reached and the reasoning for it, I believe further objection is unfounded. The added benefit that might come of a judge alone taking full control for writing every word of every opinion she signs is likely negligible. With sufficient oversight and review by the judge, the involvement of clerks and staff attorneys may in fact present an opportunity for better written opinions than if all the work were left to the judge alone. The judge must not abdicate her role as trustee, but this does not require that she draft every word from scratch. Some critics (judges among them) object that currently unpublished opinions are less thoroughly considered and therefore bear a greater danger of being wrong, so that publishing them would introduce confusion and error into *1649 the corpus. My response to this objection may come across as flippant, but I mean it quite seriously: judges are going to make mistakes. They are human. There will be hard or new cases that present real difficulties and they must produce decisions. This is their function and anything less is an abdication of their trusteeship. There is no reason that the course of the law may not be corrected later on. As noted by Justice Cardozo: There is no assurance that the rules of the majority will be the expression of perfect reason when embodied in constitution or in statute. We ought not to expect more of it when embodied in the judgments of the courts. The tide rises and falls, but the sands of error crumble. [FN36] In the case of judgment orders or memorandum dispositions (in which no reasoning, but only outcomes, are provided), the avoidance of any explanation is, in my view, an outright abdication of the judicial role. In a very useful example, two authors have presented a paradigm in which an appellate panel perceives itself to be unequal to properly deciding a complex case. [FN37] The question much discussed in the balance of their article is whether it would be advisable for such a panel to issue a decision without any explanation, in order to avoid making bad law. [FN38] Although I might cast my response in somewhat different terms, I agree with these authors' ultimate conclusion that it would be improper to issue a simple judgment order in such a case. The judge, as trustee of the law, may not choose to manage the corpus in ways she cannot account for. Judges are given latitude to seek counsel from amici and from scholars, they may seek additional briefing from the parties themselves, and they are set no time limit but that of their own consciences. If they are so uncertain of their decision as to be fearful of error in explaining it, I suggest they ought not to issue such a decision at all. A judge must be prepared to account for his decisions and accept that he may err. On the other hand, if the objection about possible error in what are now unpublished opinions is truly a concern about carelessness, or incomplete consideration given to a certain set of decisions because those cases are deemed to be less worthy of full consideration, again I can see no rational justification for such an objection. Accepting such an *1650 objection is certainly a failure to achieve thick accountability and perhaps even a failure to achieve thin accountability. Failure to publish an opinion or even just a decision will not keep it from being wrong. Failure to publish the reasoning behind it, however, may very well restrain future judges in their efforts to correct the corpus. Perhaps this sounds counterintuitive, but I am in earnest here. The more explanation a judge gives in a difficult case, the more quickly and easily (and indeed more certainly) a future judge may understand how and why a decision was reached. With that knowledge, the future judge may save a great deal of time. The future judge may more easily find the flaw and correct the course of the corpus-thus the future judge may be better able to restore integrity to the corpus when the previous judge had not found the best fit. Indeed, it is the judge's duty as trustee to consider and find that best fit in reasoning, rather than blindly follow precedent. [FN39] This fit is important in any case. It may be simpler in some cases than in others to find the right fit, but that is not a reason to draw inconsistent lines between what will and will not be integrated into the common law. In the end, it is judicial decisions that are of weightiest consequence, and those will take effect whether the reasoning behind them is published or not. [FN40] Thus, judges, as trustees, ought to focus not on writing a few opinions of particular literary greatness that will enhance their reputations, but on the everyday task of making considered decisions and providing good explanations of their *1651 reasoning so that those decisions may be properly understood and integrated into the tradition. I suggested above that a simple explanation would suffice in many if not most of the cases in which published opinions are not currently produced. One might well object that this practice will ultimately require additional time because the panel of judges must agree not just on the disposition, but on the reason for the disposition. In my view, that this issue should be brought to light is all the better for the state of the common law. If three judges cannot agree about why a disposition should be so, certainly that is an indication that the law is insufficiently clear and thus that more time and more discussion is in order. A similar objection (to which I have a similar response) is that, in the current practice of appellate courts, the unpublished status of an opinion is often a means of compromise. When judges cannot agree on the disposition, much less the reasoning for it, an end to the argument may often be reached by guaranteeing a dissenting judge that the case will not be published, therefore, eliminating the harm of the perceived error in judgment for all except the actual parties to the case and eliminating at least one judge's trouble of drafting a dissenting opinion. Again, I see little benefit and notable risk to the corpus of the common law in such compromise. Requiring publication may result in further argument and reaching of consensus or it may result in the writing of more dissents. [FN41] Either would, I think, be a beneficial outcome for the common law. One further objection might be made to my proposal that so many of these currently unpublished or unexplained dispositions could be so easily or simply explained. If they are so straightforward, it might be argued that their inclusion in the corpus of the common law would add no value. However, it would add value in both substance and perception. In substance, there is no way to predict what apparently simple line of reasoning or what nuance of reasoning in applying, extending, or denying extension of a line of precedent, may become important. In perception, it resolves the problem of drawing lines between what is and is not important for the corpus. The unpredictability of the course of the law may be one reason to avoid drawing lines, but so is the very simple perception that judges may cherry-pick cases they have some motive for publishing or unpublishing. Treating all cases alike in giving written reasoning without restraint on the future use of that reasoning solves the threshold problem of perception and also expands opportunities for better understanding *1652 the substance of the corpus of the common law, therefore improving the maintenance of the corpus. As regards the extra time this may add to the process of adjudication, I remind the reader that I limit my remarks here to the federal appellate courts. The time concern is of much greater moment at the level of trial courts. Furthermore, one in search of a resolution of a dispute is not forced to resort to the courts. One must choose the priority required for the resolution of the dispute. If that priority is justice, then speed cannot be of equal concern. If the priority is speed, other avenues of resolution are open. In conclusion, without a uniform requirement of publication of equally citable written explanations of decisions of the federal appellate courts, we will cease to be a common law country. Such a requirement would help to resolve not only the problems of perception and confidence that plague the justice system in this country, but would provide an opportunity for a flourishing of judicial virtue that would revitalize the corpus of the common law as an embodiment of agreed ideals of justice by means of a thick accountability of judges. The position was perhaps best stated over two centuries ago: Uniformity . . . cannot be expected, where the judicial authority is shared among such a vast number of independent tribunals, unless the decisions of the various courts are made known to each other . . . . In a government which is emphatically styled a government of laws, the least possible range ought to be left for the discretion of the judge. Whatever tends to render the laws certain, equally tends to limit that discretion; and perhaps, nothing conduces more to that object than the publication of reports . . . . [FN42] [FNa1]. Assistant Professor, University of Akron School of Law. It was not so long ago that I attended this same annual symposium as a student, so it is a tremendous honor to have been invited back as a participant. I am indebted to Professors David Caudill and Joan Shaughnessy, and to the Frances Lewis Law Center for including me. Many thanks as well to Brad Wendel and Brian Murchison for thoughtful challenges and suggestions at various stages in the development of this paper. [FN1]. Benjamin N. Cardozo, The Nature of the Judicial Process 179 (1921). [FN2]. For such a discussion, see generally Richard B. Cappalli, The Common Law's Case Against Non-Precedential Opinions, 76 S. Cal. L. Rev. 755 (2003) (exploring the question of publication with reference to a variety of state and federal courts at all levels). [FN3]. I do not propose that this is the only or most important social function of judging, only that it is an interesting and important one to explore for the topic of this symposium. [FN4]. The virtue to which I refer is the Aristotelian concept of virtue as it relates to justice, found in the Nicomachean Ethics, to be discussed at greater length below. The further this piece proceeds into a discussion of virtue jurisprudence, the clearer it will become that the ideas presented here err on the side of ideals, rather than practicalities. At this point, I shall say only that I acknowledge this limitation and present these theories in light of a proposal to aim for those ideals, conscious that they may not be met, but hopeful that judges might thus err on the side of virtue rather than vice. Much has been written elsewhere about the more practical and empirical aspects of this subject, so I indulge in an experiment here of more theoretical than practical use. For examples of more empirical and more practical work on the subject, see generally William L. Reynolds & William M. Richman, An Evaluation of Limited Publication in the United States Courts of Appeals: The Price of Reform, 48 U. Chi. L. Rev. 573 (1981); Cappalli, supra note 2. [FN5]. In itself, this is by no means a new or unique conception of the common law. For other discussions of how it relates to the publication of judicial opinions, see David Luban, Settlements and the Erosion of the Public Realm, 83 Geo. L.J. 2619, 2626 (1995) (discussing adjudication as "a reasoned elaboration and visible expression of public values"); Howard Slavitt, Selling the Integrity of the System of Precedent: Selective Publication, Depublication, & Vacatur, 30 Harv. C.R.-C.L. L. Rev. 109, 113-116 (1995) (discussing judicial role of espousing public values in context of publication of opinions). [FN6]. Gerald J. Postema, Integrity: Justice in Workclothes, 82 Iowa L. Rev. 821, 851 (1997). [FN7]. See generally id. [FN8]. As I have written elsewhere, for example, I believe judges have an obligation-for the sake of the clarity and integrity of the corpus-to redirect the reasoning when parties to an appeal simply miss what would be a better line of reasoning. See generally Sarah M. R. Cravens, Involved Appellate Judging, 88 Marq. L. Rev. 251 (2004). In such cases as these, in which the court takes such an involved role, publication is all the more important so that the judge's personal contribution may be fully comprehended. [FN9]. Although it is a very interesting related point, I shall not attempt to include a discussion of judicial candor in this essay. For a useful discussion of judicial candor and the ways in which it relates to these ideas of accountability, perception, and confidence, see generally Scott C. Idleman, A Prudential Theory of Judicial Candor, 73 Tex. L. Rev. 1307 (1995). [FN10]. Incidentally, this practice also has the potential to remove conflicts from the pool that it might have benefited the corpus to have resolved officially. These results are, of course, rendered all the worse in the somewhat rarer instances of de-publication of opinions. [FN11]. Postema proposes a similar idea, namely that "fidelity gives explicit shape to the historical dimension of justice." Postema, supra note 6, at 850-51. [FN12]. For a more complete treatment of Aristotelian virtue and the development of virtue ethics, see generally Aristotle, Nicomachean Ethics (Joe Sachs trans., Focus Publishing 2002); Rosalind Hursthouse, On Virtue Ethics (1999); Virtue Ethics (Roger Crisp & Michael Slote eds., 1997); Alasdair MacIntyre, After Virtue (1981). [FN13]. See generally Lawrence B. Solum, Virtue Jurisprudence: An Aretaic Theory of Law, (Working Paper 2004), available at http:// www.law.columbia.edu/center_program/legal_theory/papers/fall04? (last visited Oct. 10, 2005). [FN14]. Aristotle, supra note 12, Book 5, Ch. 1, at 1130a3-15. [FN15]. Id. Book 5, Ch. 2, at 1130b30-1131a1-4. [FN16]. Id. Book 5, Ch. 10, at 1137b12-33. [FN17]. Id. [FN18]. Solum, supra note 13, at 8, 12-13. [FN19]. Id. at 8-12. [FN20]. Id. at 13-14. [FN21]. Id. at 14-15; see also, Roger A. Shiner, Aristotle's Theory of Equity, 27 Loy. L.A. L. Rev. 1245, 1247 (1994) (explaining Aristotle's discussions of equity and his use of epieikeia). [FN22]. Ronald Dworkin, Law's Empire 238-40 (1986). [FN23]. Aristotle, supra note 12, Book 5, Ch. 10, at 1137b14-15. [FN24]. Thus, a judge can have completely unique personal values, as long as when the judge puts on his robe, he ties his reasoning specifically to the corpus through the exercise of judicial virtue-that is, seeking justice by reasoning in a way that reflects the virtues underlying the corpus. [FN25]. Luban, supra note 5, at 2626. [FN26]. Ideas about adjudication as a public good or as structural transformation are useful here. See generally id.; Owen M. Fiss, The Supreme Court 1978 Term-Foreword: The Forms of Justice, 93 Harv. L. Rev. 1 (1979) (discussing adjudication as structural transformation). [FN27]. Anthony Kronman, for example, notes: When a judge decides a case he does so not as a single individual standing in Olympian isolation, but as a member of a professional community which is itself the bearer of a specific and moral tradition . . . . The tradition of thought within which the judge is situated, and within whose horizon he encounters his task constrains him in the discretionary decisions that he makes. Anthony Kronman, The Problem of Judicial Discretion, 36 J. Legal Educ. 481, 483 (1986). See generally Karl Llewellyn, The Common Law Tradition (1960); Postema, supra note 6. [FN28]. MacIntyre, supra note 12, at 194-96. [FN29]. Dworkin, supra note 22, at 228-32. [FN30]. Cardozo, supra note 1, at 179. [FN31]. See generally Fiss, supra note 26; Luban, supra note 5. [FN32]. Cardozo considered this issue decades ago, remarking that: In these days there is a good deal of discussion whether the rule of adherence to precedent ought to be abandoned altogether. I would not go so far myself. I think adherence to precedent should be the rule and not the exception. I have already had occasion to dwell upon some of the considerations that sustain it. To these I may add that the labor of judges would be increased almost to the breaking point if every past decision could be reopened in every case, and one could not lay one's own course of bricks on the secure foundation of the courses laid by others who had gone before him. Cardozo, supra note 1, at 149. Indeed, this is another point at which it is useful to recall that it is judges who ultimately bear the responsibility for considering which prior case law is pertinent or indeed binding precedent in any case before them. The worry about increased cases for consideration is, I think, overstated. It will never be to a party's advantage to cite cases that are not pertinent, and judges ought to trust that their colleagues will have the practical wisdom to separate out the cases in which the reasoning and the outcomes are to be considered. [FN33]. In more than one conversation about this paper I have been confronted with a pragmatic challenge as to what appellate judges should be required to say by way of explaining a decision that the district court got right enough to leave alone. In those cases in which the district court itself provided a written explanation of its decision, of course I see no objection to an explanation simply incorporating that reasoning by reference (e.g., "For the reasons expressed by the court below, we affirm.") if the district court opinion was published, or by adopting and reprinting it if it was not published by the district court. This well satisfies the obligation of the trustee. When, by contrast, the district court has reached a decision requiring balancing of factors or other use of discretion and provided little or no reasoning to explain its ultimate decision, I do not think the appellate court may leave its opinion at a bald statement that it found no error in the outcome reached by the district court. The appellate court must have a reason for reaching a conclusion as to whether the district court committed error. That reasoning should be expressed. [FN34]. Oliver Wendell Holmes, The Path of the Law, 10 Harv. L. Rev. 457, 469 (1897). [FN35]. William M. Richman & William L. Reynolds, Elitism, Expediency, and the New Certiorari: Requiem for the Learned Hand Tradition, 81 Cornell L. Rev. 273, 286-92 (1996). [FN36]. Cardozo, supra note 1, at 177; see also Anthony T. Kronman, Precedent and Tradition, 99 Yale L.J. 1029, 1034-35 (1990) (discussing Hobbes' remarks on same subject: "No man's error becomes his own law, nor obliges him to persist in it. Neither, for the same reason, becomes it a law to other judges, though sworn to follow it." (quoting Leviathan 181 (M. Oakeshott ed.) (1st ed. 1651))). [FN37]. See Mitu Gulati & C.M.A. McCauliff, On Not Making Law, 61 Law & Contemp. Probs. 157, 157-59 (1988) (presenting hypothetical difficult case in which appellate courts might seek to avoid drafting reasoning behind decision). [FN38]. See generally id. [FN39]. Luban explains: Often, the judge's resolution of a legal problem-for example, choosing between two inconsistent but reasonable interpretations of statutory language-will unravel a past political compromise. Part of the judicial craft consists in unraveling as little as possible in addressing a case. Every judicial interpretation may let the political genie back out of the bottle, and so adjudication becomes a Janus-faced affair: when a judge enunciates a rule, she may be simultaneously settling a new political problem and reopening an old one. But that is no cause for regret: it is an essential part of the dynamics of a healthy polity. In adjudication, the law-the residue of political action-receives elaboration and reasoned reconsideration in the light of subsequent cases and controversies that have revealed its weak points. This process is as much a part of political vitality as free elections and legislative debate. Adjudication provides occasions for law, and therefore the politics it codifies, to assume tangible form. Take away those occasions and an essential dimension of human experience, political action, will either be extinguished or else degenerate into interminable and unresolved conflict. Luban, supra note 5, at 2637; see also Holmes, supra note 34, at 469 ("It is revolting to have no better reason for a rule of law than that so it was laid down in the time of Henry IV."). [FN40]. For lively discussion on this point, see Stephen R. Barnett, In Support of Proposed Federal Rule of Appellate Procedure 32.1: A Reply to Judge Alex Kozinski, The Fed. Law., Nov./Dec. 2004, at 32. [FN41]. Incidentally, I would extend the requirement of publication of reasoned explanations to dissents. Dissenting without an opinion is a wasted opportunity from the view of the corpus and does not discharge the dissenting judge's obligation of either thick or thin accountability as a trustee. [FN42]. 1 William Cranch, Reports of Cases Argued and Adjudged in the Supreme Court of the United States iii (Frederick C. Brightly ed., Banks Law Publishing Co. 3d ed. 1911) (1804), quoted in Slavitt, supra note 5, at 109.