Emory Law Journal

Precedent, Reliance, and the Judicial Function: A Response to Prof. Kozel
Timothy P. Terrell Professor of Law, Emory University School of Law.

Abstract

In this response,Professor Terrell argues thatProfessorKozel bears a particularly heavy burden of proof. Hefinds that ProfessorKozel'sthesis is, on the one hand, both bold and potentially far-reaching, but on the other, counterintuitive and inconsistent with our most basic understandings of the judicial role in our legal system. Professor Terrell evaluates Professor Kozel's recent article, and identifies several points of disagreement.

Introduction

I have been asked by the staff of the Emory Law Journal to respond, via the newfangled medium of an e-journal, to one of its previously published articles. The purpose of this exercise, I am told, is to generate more rapidly an academic dialogue concerning the issues identified in the initial article. I have agreed to participate in this effort, but reluctantly, for two reasons.

One is that the article I am to assess—Prof. Randy Kozel’s Precedent and Reliance1Randy J. Kozel, Precedent and Reliance, 62 Emory L.J. 1459 (2013). —is evidently one in a string of pieces by him on the general topic of precedent, some published, 2Prof. Kozel cites three other articles he has published on the topic of judicial precedent. Seeid. at 1469 n.50, 1489 n.136, 1505 n.234. others contemplated. 3In the section of his article in which he discusses the rather narrow range of his thesis in the present article, Prof. Kozel explains his limited field of view with the comment, “Alas, there is only so much one can do within the confines of a single article.” Id. at 1470. This suggests to me that he has more to say, and perhaps that will now include responding to my response. The task I have been given here, however, is to produce a short discussion of Prof. Kozel’s major points in his Emory Law Journal article, rather than present a comprehensive examination of the full range of his ideas. I have therefore not reviewed all possible relevant literature, as legitimate scholarship ordinarily expects.

This possible myopia then leads to my second reason to hesitate: I so thoroughly disagree with Prof. Kozel’s analysis of, and suggestions for, a new judicial approach to precedent that I feel I must be missing something. My negative response may not then generate dialogue but rather end it. In defense of my remarks, however, I would argue that Prof. Kozel bears a particularly heavy burden of proof: His thesis is, on the one hand, both bold and potentially far-reaching, but on the other, counterintuitive and inconsistent with our most basic understandings of the judicial role in our legal system. He had therefore better be very convincing to be impressive. I do not believe he has met that burden. His thesis turns out to be quite narrow in application, with too many implications left underdiscussed. Consequently, both the problem that prompts his article and his recommendation for addressing it remain unclear to me.

His thesis, as I understand it, is that courts—at least high-level courts—ought to act like legislatures. When such judges reconsider their existing decisional precedents, they should, like elected politicians, explicitly assess the “forward” social and economic impacts of a change in legal direction, 4Id. at 1461 (“[T]he question is less about whether past reliance should be protected and more about how departures from precedent are likely to prove disruptive going forward.”). rather than (or perhaps in addition to) the “backward” imposition on those who may have expected legal stability and thus relied to their detriment on existing caselaw. 5Id. The author believes that the current explanation and justification for varying judicial attitudes toward stare decisis is insufficiently and inappropriately grounded in these concerns for reliance interests. 6Prof. Kozel contends that his alternative approach will put the doctrine of stare decisis on “firmer ground.” Id. at 1461. Hence, a better approach for a judge to employ, in deciding whether to respect precedent or dump it, is to determine the degree and severity of possible future social and economic disruptions that a change in decisional law may impose. 7As Prof. Kozel notes, the final portions of his article that address the practical issues of implementing his approach provide only a “sketch[] [of an] . . . outline of an answer.” Id. at 1496.

I am not persuaded by this idea and its development in the article, for a host of reasons. Although the article is organized appropriately—starting with a criticism of the “reliance” assumption, then defining the scope of the article’s discussion of judicial precedent, followed by longer commentaries on deficiencies in justifications for stare decisis and thus the need for a “forward”-looking approach—I will not organize my response around the article’s structure. I will instead start with more detailed issues that struck me as vitally important—what might be labeled “procedural” concerns—that were not apparently thought very serious by Prof. Kozel. These then lead ineluctably to more fundamental—“substantive,” I suppose—problems. In proceeding in this fashion, I will not attempt to respond to every aspect of the article with which I disagree (or agree)—just those points I think most relevant to whatever additional attention Prof. Kozel may give this topic in his future work.

I. On Precedential “Procedure”

Which judges?Which cases? In the article’s second section, Prof. Kozel expressly limits the range of types of judges to which his thesis will be relevant. Rather than turn loose all judges to consider “forward” social impacts, he focuses only on the highest courts that are reconsidering only their own prior decisions—what he labels “horizontal” 8Id. at 1469. precedent in contrast to a lower court’s responsibility to respect higher court decisions (“vertical” precedent 9Id. ). But why? If the focus of his thesis is the disruptive effect of changing legal directions, why shouldn’t all judges matter? 10Prof. Kozel later acknowledges that his future-cost thesis applies to all government branches, not just the judiciary. Id. at 1502. If so, then shouldn’t it apply to the entire judiciary? Certainly a lower court decision that rejects a higher court precedent does not in and of itself “change” the general law immediately, but it sure did change the law applicable to the parties before it. Now all sorts of micro-level costs are being imposed—at the very least, appeals and the like. Indeed, a lower court decision to follow a controversial higher court decision imposes similar costs. Why have all these instances of social impact been removed from scrutiny?

The answer would be, I suspect, that high-level judicial decisions would be the ones that most clearly involve both “backward” reliance and “forward” future cost imposition, so they are appropriately the focus of the theory being espoused. But note the irony: The judges who are closest to the fact-finding responsibility of a court are removed from social cost–measuring responsibilities, while the judges furthest from that process—who must, by and large, accept the facts of a case as delivered to them—are required to reach significant and difficult additional fact-based determinations.

“Legal” errors and “factual” costs. This troubling irony ties directly to another linked to the judicial context: Appeals are limited, at least technically and formally, to the identification of a “legal” error made by a lower court that needs to be rectified. This seems to be Prof. Kozel’s assumption as well, because the focus of his analysis is the possible, and dramatic, overturning of precedent, 11Id. at 1470. and thus the appellate court’s decision not to respect stare decisis. But having brought to this court a “legal” mistake—that is, the lower court’s use of a “bad” precedent—the appellate court would now be required, as part of its analysis of how to respond, to consider facts associated with respecting or dumping that precedent. Does Prof. Kozel’s proposal assume that these considerations were developed and argued in the court below?

How judges? My concern over what an appellate court is supposed to do leads inevitably to questions about how it is supposed to do it. What steps, exactly, should a judge take to determine satisfactorily the relevant future social costs associated with her decision? I find the problems here endless. Judicial contexts ordinarily involve “evidence” or “showings” or something by the parties within the context of their adversarial circumstances, and under this new approach, these efforts would be for the additional purpose of assisting the judge in identifying and measuring future social costs. How would this be accomplished, particularly at the appellate level? Would the standards for appellate practice now take on a new dimension? For example, would a litigant’s attorney who did an adequate job presenting traditional legal arguments nevertheless be vulnerable to a malpractice action if she made a less than stellar effort in presenting “future impact” information or analysis? Under Prof. Kozel’s approach, would judges be entitled to reach outside the record—and witnesses—developed by the parties, perhaps calling experts who will present information of which the judge will take “judicial notice”? Because “social cost” discussions are a regular part of modern law school teaching, would law professors be appropriate expert witnesses in these contexts? Or only some of us? Which ones?

Although this “procedural” problem of judicial assessment of extrajudicial information obviously overlaps with the series of “substantive” points to be made below about the political circumstances in which courts operate, nevertheless if we remain focused here on simply “how” appellate courts will implement Prof. Kozel’s approach, we can then predict a very difficult question in the future: What should an appellate court do if a legislature or administrative agency, knowing that an important existing judicial precedent is being reconsidered by a court, engaged in its own fact-finding exercise concerning the “disruption costs” of a change in judicial direction? In the first instance, I’m not sure how this fact-finding would be presented to the court? An amicus brief? But assuming that these “facts” become part of the appellate record, would the court be bound by these conclusions? Or could the court ignore this information in favor of its own independent analysis? We will consider aspects of these institutional challenges anon.

Overruling vs. narrowing vs. following precedent. If Prof. Kozel is serious about emphasizing future impacts of decisions rather than impacts on past reliance on precedent, then a sequence of difficulties arises. First, why does he limit his approach to judges who intend to abandon an existing precedent completely rather than less dramatically narrowing it? 12Id. This is the point in the article, noted above, supra note 3 , where Prof. Kozel seems to suggest that he may consider this issue in some future article. If social costs imposed by judicial decisions are the key, then a substantial, but not total, change in the law can matter just as significantly as a complete one. But if these less-than-total judicial reconsiderations of precedent are subject to Prof. Kozel’s requirement for future-cost analysis, how is the court (or anyone else) to determine how much of a nontotal change would trigger this responsibility?

Second, it would seem that social costs can be imposed just as readily by a judge adhering to a controversial precedent as overruling it. If that is true, then every judicial decision (particularly, again, at the appellate level) would seem to be subject to Prof. Kozel’s criticism that judges do not do an adequate job recognizing and measuring the social impacts of their work. 13This is the essential message of his analysis of the example cases in Parts IV.B and C of his article.

Perhaps Prof. Kozel’s point here is that he wants to demonstrate that his thesis about future-cost analysis by a court is viable even in situations where past reliance would seem to enjoy its strongest foundation—where legal change is being imposed at the highest level in the most dramatic form. My problem with this possible explanation, however, is that in these high-drama circumstances, past reliance could be for a court the most accurate proxy or estimate of what future costs might be. The need for additional “proof” of costs, then, would not seem to be particularly acute, even for a court that wants to do what Prof. Kozel proposes. Perhaps, again, Prof. Kozel’s ultimate point is that attempting to measure future social costs would be a more “honest” label for the court’s effort, but that honesty requires such radical reconsideration of the judicial role that it would not seem to be institutionally justified.

Consequently, perhaps the tougher but more interesting and challenging situations for Prof. Kozel’s thesis would be less dramatic situations of reconsidering precedent—like narrowing or following—where backward-looking reliance thinking might be at its weakest as a reason for stare decisis. There the legitimacy of future-cost analysis could be tested and assessed independently.

But my more fundamental objection remains the same here: Whether we focus on the most significant instances of legal change or we permit less dramatic circumstances to be relevant, the idea of expecting judges to engage in the analysis Prof. Kozel’s thesis proposes is very troubling even if all we consider are these “procedural” sorts of difficulties. I would argue that it is for these reasons alone that judges simply know they cannot (and will generally refuse to) do what Prof. Kozel wants them to do.

II. On Precedential “Substance”

Now, however, we need to go deeper. I suppose one could, thus far, defend Prof. Kozel’s thesis against my response by assuming that these issues of proof of future costs and the range of judges expected to consider them are mere matters of administrative detail that could be resolved sufficiently. But they can’t and they won’t be, simply because of more “substantive” problems with his thesis to which I alluded earlier—issues concerning the nature of judicial decisions within our political and legal system.

The politics, rather than costs, of overruling. I believe that Prof. Kozel’s analysis overlooks to its detriment that in studying precedent, we must be mindful of the nature of the “kind” of law, if you will, that is at stake in this appellate context. Although Prof. Kozel limits his article to a high court reconsidering only its own prior decisions, 14This is what Prof. Kozel means by “horizontal” precedent, noted supra note 8 . note that, even at the most superficial level of considering this decision, it can be of three very different sorts: The prior decision may have been only about a common law issue, over which the appellate court would have direct authority; or it could be a case involving constitutional interpretation; or it could be one in which the court must interpret a statute or administrative regulation. For purposes of the following discussion, I will combine the common law and constitutional situations within the “common law” label, simply because the appellate court’s authority is often viewed as roughly the same in both contexts. In the statutory/regulatory circumstance, however, the critical difference is that whatever meaning the court may ascribe to the statute or regulation, the authors of that document in other branches of government have always had the power to respond—to endorse or reject—the court’s work. Thus, in a current case before the court in which a prior interpretation is now being reconsidered, the court’s decision, whatever it may be, is simply a signal to the other branches that the “legal ball” is being kicked back to them. Indeed, more broadly, even an appellate decision involving only the common law will be a legal announcement that the legislature can reject or modify.

Prof. Kozel’s thesis, I believe, never seems to take other branches of government seriously as participants in this stare decisis circumstance. He does note this issue, but only in passing in a footnote, where he summarily dismisses it as being a matter of detail that can readily be brought within the article’s thesis:

This Article draws no distinction between statutory and constitutional precedents. Its focus is the treatment of stakeholder expectations, which arise in both contexts. To the extent one believes that statutory precedents deserve an additional degree of deference based on notions of implied congressional acquiescence or separation-of-powers norms, that deference could be integrated with the framework this Article develops for analyzing disruption costs. 15Id. at 1463 n.14.

I am not satisfied by this move. The “disruption” or “future social cost” judicial calculus that Prof. Kozel proposes is already remarkably murky, and now a judge must add to it some measure of the degree to which people in general or legislators in particular believe the work of other political branches deserve “more” deference (whatever that might mean in and of itself) now makes this calculation impossible to pin down meaningfully. What “amount” would this be, and how would it compare to other “amounts” of disruption? How would one put a “value” on, for example, “separation-of-powers norms”—that is, the fact that the judge will now do what other branches were elected to do? This would seem to me to be pure speculation, and judicially self-serving at that. 16Prof. Kozel does note in his article a version of my concern, but again he does not apparently consider this kind of cost approximation to be particularly serious to his thesis:A court will never be able to express these disruptive impacts with absolute precision. It may, however, conclude that the effects are likely to be minor, moderate, or extreme, and it may strive to understand the myriad ways in which disruption is likely to occur. That type of rough accounting is not perfect, but it is useful in capturing the disruptive costs to be weighed against the benefits of overruling a dubious precedent.Id. at 1497–98. And at the end of his article he again acknowledges the significant lack of detail and guidance concerning this fundamental point of cost assessment:This Article does not presuppose or advocate any particular view of the relative weight to which considerations of disruption should be entitled in the doctrinal calculus. My point is simply this: any judge or commentator who is inclined to give some regard to settled expectations needs a theoretical justification for paying attention to such expectations and an analytical framework for assessing them.Id. at 1506. And my point is simply that, in the absence of meaningful guidance regarding the judicial assessment of future costs of disruption, Prof. Kozel’s article does not meet the standard he otherwise imposes on others.

The key here, I would argue, is not really “judicial deference” or “disruption costs” so much as political responsibility and accountability. It is about the simple political fact that different branches of government were created to do different sorts of things, which now have been collapsed by Prof. Kozel into a one-branch calculation. To the contrary, if a judge decides to treat legislation with an added measure of deference, or decides that a particular legal issue is best left to the legislature to decide, that is not, and should not be, based in any way on a calculation of future “costs.” It is rather simply a statement about relative institutional obligation and (hopefully) competence. The question that Prof. Kozel must answer is how such straightforward, traditional legal analysis would be enhanced by additional speculation about “costs” imposed on our political system.

And now the “procedural” point made earlier 17See supra p. 40. about “fact-finding” by nonjudicial branches of government concerning “disruption costs” comes into play: Are courts bound by these “facts,” or are judges entitled to conduct an independent analysis? If so, what would be the basis for this judicial authority? We will return to this point yet again at the end of this Response where we note Prof. Kozel’s reference to the Constitution’s phrase “the judicial Power.” 18See infra p. 50.

But even if we allow Prof. Kozel’s proposal to be limited only to the most ordinary common law situations, and we therefore focus entirely on the closed loop of judges reconsidering their brethren’s previous work product, significant difficulties with the future-cost analysis remain.

What is precedent? One of my most basic concerns with Prof. Kozel’s proposal is that the key concept of stare decisis is itself underanalyzed. What is it, exactly, that a future court must follow or reject? What if such a court disagreed with and disavowed the reasoning in a prior case but noted that the result reached in that decision for those parties in that case was nevertheless correct—the “reversed on other grounds” idea? To be appropriately justified, that determination would seem to demand the future social cost analysis. But I am not certain whether that is true. If rejections of reasoning but not results do trigger Prof. Kozel’s proposal, then how much of a prior case’s reasoning needs to be rejected or endorsed for future social costs to be implicated?

A jurisprudential avenue into this issue reveals a remarkable amount of potential detail that Prof. Kozel’s analysis simply asserts is beyond the scope of his article, 19Seesupra note 3 . but I believe is vital to its viability. One possibly useful way of imagining the problem here is in the early work of Prof. Ronald Dworkin where he developed his well-known theories of common law adjudication. He argued that any judicial decision has two forms of “force” associated with its precedential value. One is the decision’s narrow “enactment” force, 20Ronald Dworkin, Taking Rights Seriously 111 (1977). which is simply the direct impact of the court’s legal conclusions on the parties in that case—who won, how much, why, and so on. The other is the decision’s “gravitational” force 21Id. —the persuasiveness of, and respect accorded, its reasoning that might be applied in similar later situations. 22Indeed, the concept of gravitational force is obviously relevant as well to what one means by a “similar” case. A particularly challenging aspect of Prof. Kozel’s proposal to have appellate courts explicitly consider future social costs is therefore that the foundational legal idea (and ideal) of “treating like cases alike” may take on a whole new, and far more controversial, meaning. In other words, could a case that involves very similar future social-cost implications be cited as authority in another case where the two cases have nothing legally in common? This dichotomy in turn identified for Prof. Dworkin two ways in which a precedent that is now regretted—a legal “mistake”—could be given no authority in a later decision. First, a really bad decision could have both forms of force denied—the result and the reasoning are rejected, and neither has any future relevance. Examples would often be mundane, of course, like the Supreme Court reversing a lower court. Or they could be more like what Prof. Kozel has in mind of a big shift in the law, where the Supreme Court says that if these same parties in the prior Supreme Court case came before the Court today, the result as well as the law for them would be the opposite. 23See, e.g., Brown v. Bd. of Educ., 347 U.S. 483, 494‒95 (1954) (“Whatever may have been the extent of psychological knowledge at the time of Plessy v. Ferguson, [the harm of segregation on African American children] is amply supported by modern authority. Any language in Plessy v. Ferguson contrary to this finding is rejected.”). Either of these situations Dworkin would label “corrigible” 24Dworkin, supra note 20 , at 121. legal mistakes, those that can be fully expunged. Or, second, a now regretted judicial “mistake” might be considered an “embedded” 25Id. mistake, where its enactment force remains intact, but courts chip away at its gravitational force—limiting or recasting its reasoning, scope, and so on. 26The judicial angst over Roe v. Wade, 410 U.S. 113 (1973), quickly comes to mind as an example. See, for example, the remarkable effort by the Court in Planned Parenthood v. Casey, 505 U.S. 833 (1992), to rethink Roe’s precedential scope and force, a project that seems never ending. For an analysis of the changes that Roe v. Wade has undergone over the years, see Victoria Baranetsky, Aborting Dignity: The Abortion Doctrine After Gonzales v. Carhart, 36 Harv. J.L. & Gender 123 (2013). Would both of these judicial responses to “mistaken” precedent be subject to Prof. Kozel’s requirement for future social cost assessment?

Prof. Kozel takes this issue of the “scope” of a precedent off his analytic table, 27Kozel, supra note 1 , at 1470. but I do not believe he can so easily avoid this issue. Prof. Dworkin’s analysis of a court’s treatment of its own prior decisions, for example, is directly connected to key aspects of his theory of “law” itself. Any decision, he argued, will constitute within our legal system a “rule” on which future courts can depend in reaching decisions. 28Dworkin, supra note 20 , at 22‒28, 71‒80. But that rule will simultaneously be only a particular, narrow application of a larger, deeper, normative “principle” 29Id. (or a set of principles) that justifies the use of that rule in the future case. One would then surmise that Prof. Kozel’s proposal for judicial measuring of future social costs would be most relevant in any case in which a judge might be considering rejecting a prior decision’s use or understanding of such important legal principles. That, however, can occur in far more situations than just those in which a court intends to completely eliminate a corrigible mistake. In turn, if that’s true, how would a judge identify the embedded mistakes that are important enough to require future social cost assessment?

The point of this brief tour through a few aspects of Prof. Dworkin’s thinking is that Prof. Kozel’s dramatic restructuring of the appellate judicial role requires, to be convincing, a similarly deep consideration of what “future social cost” assessment would mean to our sense of “law” in the first place. I don’t see that in this current article by Prof. Kozel, but perhaps that is on his agenda for later work.

Reviewing this jurisprudential nuance demonstrates, I think, that any judge’s consideration of precedent is already remarkably complicated simply on the basis of the dazzling mosaic of the law itself—colliding rules, overlapping principles, and the like—which judges are required to comprehend and make coherent. On top of that challenge, Prof. Kozel now argues for more judicial responsibility regarding topics outside the law itself that may not have been addressed by the parties at all. It is no wonder to me, then, that Prof. Kozel’s proposition has not been suggested before. And he must demonstrate to us why it should gain any traction among judges now.

Justifying Stare Decisis. All the preceding steps lead to my trouble with the central tenet of Prof. Kozel’s article: that the fundamental basis for the doctrine of stare decisis, which he links directly (and solely) to the vindication of reasonable “reliance interests,” is inadequately grounded. 30Kozel, supra note 1 , at 1462‒69. He sees an “apparent tension” between precedent being, on the one hand, “entitled to judicial respect,” yet on the other, simultaneously “mutable.” 31Id. at 1460. Precedent is somehow both certain and flexible. This sort of conundrum ought, then, one would think, require a fairly serious explanation and justification, which Prof. Kozel believes is absent. To fill that vacuum, he contends that judges should more honestly, directly, and openly do what they seem to be doing inconsistently and implicitly: when considering whether to overrule a prior decision, judges should assess the future social costs of a change in the law. 32Id. at 1462.

I believe Prof. Kozel is wrong on both counts: The existing judicial attitude toward precedent, including the “tension” he identifies, does have an adequate foundation, and his effort to improve the doctrine of stare decisis is a cure worse than any existing legal disease.

Concerning foundations, Prof. Kozel expends considerable energy in the third section of his article critically assessing a number of possible reasons why reliance protection has remained the dominant (if not sole) theory of stare decisis.33Id. at 1470‒85. Depending on how you count the sections and subsections in his discussion, this range (and I count eight possible foundations), as he puts it, “from fairness implications to investment incentives” 34Id. at 1471. still leads him to the conclusion that “protecting precedential reliance is too uncertain and underdeveloped to be entirely persuasive.” 35Id. The point of these separate, focused examinations is, as I understand it, that each of them might explain a piece of what lies beneath stare decisis, but none of them provides the kind of satisfying resolution to the doctrinal “tension” that he seeks. 36Id.

My problem with this conclusion, however, is that I do not think that any of these separate explanations and/or justifications for reliance protection ever claimed to be the only basis upon which to understand our respect for precedent, ambivalent and inconsistent though it may be. Rather than looking at each of these possible foundations separately, they ought instead be appreciated as whole: None of them alone may be overly impressive as a justification for limiting judges to assessing backward-looking reliance, but as a group they are quite powerful. This amalgamated approach is much like the analysis of a doctrine within a particular area of law with which I happen to be familiar (because I teach it): the foundation for the so-called no-contact rule in legal ethics, codified in in Rule 4.2 of the ABA’s Model Rules of Professional Conduct. The details of that Rule of course do not matter here, only that this ethical restriction is much debated among legal ethics scholars. 37A useful summary discussion of the range of potential professional values implicated by the no-contact rule appears in the casebook by Stephen Gillers, Regulation of Lawyers: Problems of Law and Ethics 100 (9th ed. 2012). While none of the possible justifications for that remarkable doctrine fully captures our—again like stare decisis—ambivalent and sometimes inconsistent respect for it, the full list certainly does. It is as if, for example, each of Prof. Kozel’s listed potential reasons for our current reliance focus captured maybe as little as 15% of the explanation/justification for our attitude—not all that stunning alone, but as a group, more than enough to support the traditional understanding of stare decisis.

And this absence of a singular justification for stare decisis does not strike me as a fault of that doctrine itself or our understanding of it. It is instead a function of the fact that we have always expected, and indeed demanded, that judicial precedent be both durable and malleable. We want and need that kind of flexibility because of the nature of what we want “law” to be most fundamentally. I would want Prof. Kozel to challenge me at that level of analysis. In its absence, my conclusion is that his search for some additional foundation for stare decisis—either as a supplement to or a replacement for the existing array of partially successful explanations—isn’t really necessary in the first place.

But more importantly, even if there were a gap in this doctrine’s foundation, I am not at all convinced that Prof. Kozel’s method of filling it is a good idea. There are very good reasons why judges do not perform the sort of legislative, future-oriented function his thesis recommends, and they are reflected in the “procedural” difficulties I discussed earlier. A matter comes before a judge if and only if there is a lawsuit that generates it. No matter how fundamental and far-reaching the results of the litigation may be, you still need a plaintiff and a defendant to get things started. This necessarily reduces any court’s field of vision. Discovery in the case is not intended to develop all possible social implications that the case’s decision may entail. Certainly some well-known and dramatic cases have suggested the contrary, where arguments and information seemed to make all of modern life relevant to the case. But—and this is the key point here—those cases are quite rare, even at the level of the Supreme Court.

And they should be rare. Sweeping social pronouncements by judges that are not anchored firmly and clearly in what we understand (admittedly in a seat-of-the-pants sort of way) to be “the law” are generally unwelcome and viewed with significant social suspicion. 38Prof. Kozel does at least acknowledge to some degree the institutional, political differences between courts and legislatures. See, e.g., Kozel, supra note 1 , at 1503 (discussing the judiciary’s “substantial insulation from popular accountability for their decisions”). He simply does not apparently find such distinctions nearly as compelling as I do. Even without sophisticated jurisprudential training, people know that there is some important distinction between legal doctrine and social engineering—the kind of separation that Prof. Dworkin, once again, labeled as “legal principles,” 39Dworkin, supra note 20 , at 90. on the one hand, and “social policies,” 40Id. on the other. The former are supposed to be about “rights”—about the deontological foundations of our legal system, while the latter are about teleological consequences that are the proper province of legislatures. This is why arguing something as basic as “economic efficiency” to a court as the potential foundation for a decision is uncomfortable. 41This debate has been raging for many years, of course. The most famous initial salvos were launched in the debates among Professors Dworkin, Guido Calabresi, and Richard Posner, among others. See Guido Calabresi, About Law and Economics: A Letter to Ronald Dworkin, 8 Hofstra L. Rev. 553 (1980); Ronald Dworkin, Why Efficiency?: A Response to Professors Calabresi and Posner, 8 Hofstra L. Rev. 563 (1980); Richard A. Posner, The Ethical and Political Basis of the Efficiency Norm in Common Law Adjudication, 8 Hofstra L. Rev. 487 (1980). But the disagreements continue. See, e.g., Anita Bernstein, Whatever Happened to Law and Economics, 64 Md. L. Rev. 303 (2005); Stephen O’Hanlon, Equality, Entitlement, and Efficiency: Dworkin, Nozick, Posner, and Implications for Legal Theory, 8 Cardozo Pub. L. Pol’y & Ethics J. 31 (2009). The question for Prof. Kozel then must be, if we cannot settle this issue of doing a standard economic analysis in a judicial context, how can we hope to resolve the challenge of a assessing a judge’s effort to calculate what seem to be far more abstract future “disruption costs”? That kind of concern is more appropriately addressed to political branches with the resources and more general perspective and responsibility to make (hopefully) competent decisions for which they will then be (at least theoretically) politically accountable.

None of these institutional concerns, however, are within Prof. Kozel’s analytic agenda. His approach is not about the “micro” limitations of the judicial context, but the larger “macro” circumstances in which the legal system operates. Here, there is no real difference, he seems to believe, between courts and legislatures, for all these actors would have as their responsibility, as he puts it at the beginning of his article, “the benefit of society as a whole.” 42Kozel, supra note 1 , at 1461. This perspective means that while Prof. Kozel does indeed acknowledge counterarguments to his thesis based in concerns about the “Rule of Law” 43Id. at 1480‒81. and “Judicial Identity,” 44Id. at 1481‒83. he does not find them convincing. Instead, he views the future-cost exercise as well within the judicial function:

Judicial assessments of the net effects of deviating from precedent may be informed by the same types of evidence of disruptive impact that Congress would utilize in making its own determination. Yet the ultimate comparison of the value of settlement versus the value of adjudicative change is better left to the courts as a component of “[t]he judicial Power” that they are charged with exercising. 45Id. at 1499 (alteration in original).

I fundamentally disagree. The supposed benefits of conflating judicial and legislative roles have not been demonstrated to my satisfaction. As I noted earlier, I worry about the relative authority and competencies of judicial versus nonjudicial branches to assess large-scale future “disruption costs.” Here I am emphasizing more fundamentally that speculating about future social costs is not at all the equivalent of analyzing the “rights” of the parties before a court. Judges currently have more than enough to do—and more than enough “Power”—without making the basic doctrine of stare decisis even more controversial.

Footnotes

Professor of Law, Emory University School of Law.

1Randy J. Kozel, Precedent and Reliance, 62 Emory L.J. 1459 (2013).

2Prof. Kozel cites three other articles he has published on the topic of judicial precedent. Seeid. at 1469 n.50, 1489 n.136, 1505 n.234.

3In the section of his article in which he discusses the rather narrow range of his thesis in the present article, Prof. Kozel explains his limited field of view with the comment, “Alas, there is only so much one can do within the confines of a single article.” Id. at 1470. This suggests to me that he has more to say, and perhaps that will now include responding to my response.

4Id. at 1461 (“[T]he question is less about whether past reliance should be protected and more about how departures from precedent are likely to prove disruptive going forward.”).

5Id.

6Prof. Kozel contends that his alternative approach will put the doctrine of stare decisis on “firmer ground.” Id. at 1461.

7As Prof. Kozel notes, the final portions of his article that address the practical issues of implementing his approach provide only a “sketch[] [of an] . . . outline of an answer.” Id. at 1496.

8Id. at 1469.

9Id.

10Prof. Kozel later acknowledges that his future-cost thesis applies to all government branches, not just the judiciary. Id. at 1502. If so, then shouldn’t it apply to the entire judiciary?

11Id. at 1470.

12Id. This is the point in the article, noted above, supra note 3 , where Prof. Kozel seems to suggest that he may consider this issue in some future article.

13This is the essential message of his analysis of the example cases in Parts IV.B and C of his article.

14This is what Prof. Kozel means by “horizontal” precedent, noted supra note 8 .

15Id. at 1463 n.14.

16Prof. Kozel does note in his article a version of my concern, but again he does not apparently consider this kind of cost approximation to be particularly serious to his thesis:A court will never be able to express these disruptive impacts with absolute precision. It may, however, conclude that the effects are likely to be minor, moderate, or extreme, and it may strive to understand the myriad ways in which disruption is likely to occur. That type of rough accounting is not perfect, but it is useful in capturing the disruptive costs to be weighed against the benefits of overruling a dubious precedent.Id. at 1497–98. And at the end of his article he again acknowledges the significant lack of detail and guidance concerning this fundamental point of cost assessment:This Article does not presuppose or advocate any particular view of the relative weight to which considerations of disruption should be entitled in the doctrinal calculus. My point is simply this: any judge or commentator who is inclined to give some regard to settled expectations needs a theoretical justification for paying attention to such expectations and an analytical framework for assessing them.Id. at 1506. And my point is simply that, in the absence of meaningful guidance regarding the judicial assessment of future costs of disruption, Prof. Kozel’s article does not meet the standard he otherwise imposes on others.

17See supra p. 40.

18See infra p. 50.

19Seesupra note 3 .

20Ronald Dworkin, Taking Rights Seriously 111 (1977).

21Id.

22Indeed, the concept of gravitational force is obviously relevant as well to what one means by a “similar” case. A particularly challenging aspect of Prof. Kozel’s proposal to have appellate courts explicitly consider future social costs is therefore that the foundational legal idea (and ideal) of “treating like cases alike” may take on a whole new, and far more controversial, meaning. In other words, could a case that involves very similar future social-cost implications be cited as authority in another case where the two cases have nothing legally in common?

23See, e.g., Brown v. Bd. of Educ., 347 U.S. 483, 494‒95 (1954) (“Whatever may have been the extent of psychological knowledge at the time of Plessy v. Ferguson, [the harm of segregation on African American children] is amply supported by modern authority. Any language in Plessy v. Ferguson contrary to this finding is rejected.”).

24Dworkin, supra note 20 , at 121.

25Id.

26The judicial angst over Roe v. Wade, 410 U.S. 113 (1973), quickly comes to mind as an example. See, for example, the remarkable effort by the Court in Planned Parenthood v. Casey, 505 U.S. 833 (1992), to rethink Roe’s precedential scope and force, a project that seems never ending. For an analysis of the changes that Roe v. Wade has undergone over the years, see Victoria Baranetsky, Aborting Dignity: The Abortion Doctrine After Gonzales v. Carhart, 36 Harv. J.L. & Gender 123 (2013).

27Kozel, supra note 1 , at 1470.

28Dworkin, supra note 20 , at 22‒28, 71‒80.

29Id.

30Kozel, supra note 1 , at 1462‒69.

31Id. at 1460.

32Id. at 1462.

33Id. at 1470‒85.

34Id. at 1471.

35Id.

36Id.

37A useful summary discussion of the range of potential professional values implicated by the no-contact rule appears in the casebook by Stephen Gillers, Regulation of Lawyers: Problems of Law and Ethics 100 (9th ed. 2012).

38Prof. Kozel does at least acknowledge to some degree the institutional, political differences between courts and legislatures. See, e.g., Kozel, supra note 1 , at 1503 (discussing the judiciary’s “substantial insulation from popular accountability for their decisions”). He simply does not apparently find such distinctions nearly as compelling as I do.

39Dworkin, supra note 20 , at 90.

40Id.

41This debate has been raging for many years, of course. The most famous initial salvos were launched in the debates among Professors Dworkin, Guido Calabresi, and Richard Posner, among others. See Guido Calabresi, About Law and Economics: A Letter to Ronald Dworkin, 8 Hofstra L. Rev. 553 (1980); Ronald Dworkin, Why Efficiency?: A Response to Professors Calabresi and Posner, 8 Hofstra L. Rev. 563 (1980); Richard A. Posner, The Ethical and Political Basis of the Efficiency Norm in Common Law Adjudication, 8 Hofstra L. Rev. 487 (1980). But the disagreements continue. See, e.g., Anita Bernstein, Whatever Happened to Law and Economics, 64 Md. L. Rev. 303 (2005); Stephen O’Hanlon, Equality, Entitlement, and Efficiency: Dworkin, Nozick, Posner, and Implications for Legal Theory, 8 Cardozo Pub. L. Pol’y & Ethics J. 31 (2009). The question for Prof. Kozel then must be, if we cannot settle this issue of doing a standard economic analysis in a judicial context, how can we hope to resolve the challenge of a assessing a judge’s effort to calculate what seem to be far more abstract future “disruption costs”?

42Kozel, supra note 1 , at 1461.

43Id. at 1480‒81.

44Id. at 1481‒83.

45Id. at 1499 (alteration in original).