[*PG81]PRECEDENT AND JUDICIAL POWER AFTER THE FOUNDING
Abstract: A recent decision by a panel of the Court of Appeals for the Eighth Circuit enlivened the controversy over court rules that prevent citation to unpublished opinions when it held that the Circuit’s non-citation rule violates Article III of the United States Constitution. This Article affirms the view that judicial power includes a doctrine of precedent, without relying solely upon an originalist interpretation of Article III. This approach identifies a consistent “core idea” of precedent that courts must consider how a similar case was decided in the past, even where there are varying ideas about the binding nature of that precedent. A long-standing tradition has viewed precedent as a necessary starting point for judicial decision. When a court departs from this core idea, it violates the essential function of the judiciary to treat like cases alike or explain the difference.
Like a phoenix rising from the ashes of modern legal theory, respect for precedent is poised for a comeback. The evidence is Anastasoff v. United States1—an opinion from the Eighth Circuit Court of Appeals that created shockwaves when it invalidated that court’s rule against citation of unpublished opinions. Anastasoff announced that the United States Constitution imposes some obligation on federal courts to respect precedent, an obligation that is inconsistent with the current practice of many federal and state courts to consider unpublished opinions to hold no precedential value. This is not a minor issue. In 1999, nearly eighty percent of all federal appellate cases fell into this “not for official publication” category.2 Because in most fed[*PG82]eral circuits an unpublished opinion is not precedent and may not be cited, this means that less than a quarter of all judicial work in the federal circuits in that year became “precedent” for future cases. Anastasoff rejected the idea that, while hiding behind these rules, courts may decide issues of law one way one year and another way the next year, without explaining the difference.
Perhaps, though, the return of the precedent phoenix is not so startling after all. Legal scholars for some time have suggested that restrictions on publication and citation of court opinions are problematic, although no one seems to have argued that Article III poses a constitutional problem.3 If “precedent” is simply the record of how previous cases were decided (so the argument goes), and courts are indeed bound by precedent, then how can a court mandate that what it has decided in the past is not precedent?4
Writing for the three-judge panel in Anastasoff, Judge Richard S. Arnold reasoned that the portion of the rule which declares that “unpublished opinions are not precedent” is an unconstitutional extension of the judicial power described in Article III of the United States Constitution. For the federal courts, according to Anastasoff, the traditional doctrine of precedent is a fundamental component of the notion of judicial power. Because the doctrine of precedent is inherent in the concept of judicial power, avoidance of precedent, at least through non-citation rules, goes beyond the judicial power conferred by Article III.5 So far as I am aware, legal scholars have not pondered the extent to which respect for precedent may have been enshrined in the Framers’ vision of Article III. Rather than engage modern legal scholarship which locates the justification for the authority of prece[*PG83]dent on equitable or prudential grounds,6 Judge Arnold found the doctrine of precedent to be rooted in the Constitution’s conception of judicial power, squarely grounding precedent in the fundamental law authorizing federal courts in the United States. Apparently for the first time, a federal court has found some constitutional obligation to consider precedent.
Anastasoff has already garnered a great deal of attention,7 some of it critical.8 In this article, I argue that Judge Arnold is right about the core meaning of judicial power, even if it is not necessary to rely solely upon an originalist interpretation of Article III to make this point. A contrary result would mock the notion of equal justice under the law. The issue is not about whether courts officially publish their opinions, nor is there any requirement that judicial opinions in routine cases must be longer than a single sentence or paragraph. Instead, Anastasoff is an affirmation of two values at stake in the work of appellate courts: transparency of judicial decisionmaking, and treating like cases alike.
I defend this position by considering two potential criticisms of Anastasoff. The first is directed at the method of constitutional interpretation used in Anastasoff. The court uses an historical method to conclude that Article III was intended to impose some restraint on the federal judiciary with respect to precedent. The second criticism targets the future effects or consequences of the holding. Specifically, it questions whether Judge Arnold’s interpretation of Article III and the theory of precedent as mandatory impose restrictions on what courts may do to deviate from precedent.
Intent is impossible to determine here, one might suggest, because the Framers said nothing specific about this understanding of [*PG84]judicial power. In the absence of solid evidence of specific intent, one might argue, a constitutional interpretation that stands solely on “originalism” is a weak argument for invalidating non-citation rules. This weakness, critics may argue, is demonstrated by historical evidence tending to show a different understanding of judicial power, which begs the question whether an originalist approach provides any better method of interpretation than other possibilities. I will suggest that the same result could be reached by according authoritative weight to long-standing tradition.
Another potential problem with the originalist argument is that some historians have located the beginning of a “strict” doctrine of precedent only in the post-founding period. In English legal history it has been said that the distinction between a line of similar cases as a source of law, and the binding force of a single decision, was not fully formed in England before the nineteenth century.9 For the United States, it has also been claimed that courts in the nineteenth century adhered to stare decisis10 only loosely, and that prior cases were not, “even nominally,” ultimately binding.11 In fact, one scholar has claimed that “we have never known the strict doctrine [of stare decisis] in the United States” and that we have always had a “lax regard for precedent.”12
As described in more detail below, the conclusions of these historians are not inconsistent with the core idea of precedent ascribed to Article III. This core idea is simply that courts must start with their own precedent, even if there are varying ideas about the binding nature of that precedent. That core idea was not only clearly present at the Founding, it has remained consistent over the course of American [*PG85]legal history. Anastasoff makes no greater claim than this bare minimum and does not require the “strict” doctrine discussed by these historians. Furthermore, one can cite earlier instances in which courts strictly adhere to the binding force of a single decision than some historians have previously recognized.
Thus, although I will have a few points to make about historical methodology, this Article does not focus solely upon what the Framers would have assumed about precedent. We cannot know precisely what the Framers had in mind when they used the term “judicial power.” But Anastasoff is at least convincing in its general point. History can be used in other ways to support the result in Anastasoff, in addition to ascribing an intent to the Framers. I suggest there is continuity over time in this basic understanding of how courts will operate, despite cyclical variations in the degree to which prior precedent is considered binding.
The second set of issues that I will address are those directed toward the future effects or consequences of the holding in the case. Thus far, the tendency has been to assume a stricter version of the doctrine of precedent than Anastasoff claims. Assuming Judge Arnold is correct in his interpretation of Article III, and thus assuming that a theory of precedent is mandatory at some level, what are the possibilities for departures from precedent? If there is an Article III restriction on what courts must consider to be precedent, is there not also a corresponding Article III restriction on what courts can do to deviate from precedent (by distinguishing cases, restating holdings, and so forth)?
As described in more detail below, my answer to this last question is no. In Anastasoff, the court stressed that it was not “creating some rigid doctrine of eternal adherence to precedents.”13 The specific result of Anastasoff’s historical inquiry is less ambitious than many fear. According to Anastasoff, courts are free to overrule or distinguish precedent, but they are not free to ignore it entirely. Courts are not strictly bound to follow prior precedent, but at a minimum they must take note of the prior determination and explain any choice to decide the matter differently. This view is consistent with how most legal scholars have thought about precedent at most times. Indeed, it is a fundamental assumption of common-law systems. This core idea of precedent has survived two centuries of wildly changed perceptions of the nature of the American common law. To understand judicial [*PG86]power as the Framers might have intended, it is appropriate to defer to this consistent practice.14 This tradition tends to prove that Article III constitutionalized this core idea of precedent, and it is nothing more than this core idea that Anastasoff has affirmed.
[I]n the late eighteenth century, the doctrine of precedent was well-established in legal practice (despite the absence of a reporting system), regarded as an immemorial custom, and valued for its role in past struggles for liberty.
—Anastasoff v. United States (2000)15
Before proceeding further it is useful to provide a working definition for what is often termed the “traditional” doctrines of precedent, and then a brief summary of the Anastasoff decision itself. As usually stated, the doctrine of precedent at its core is that the holding of a case must be followed in similar cases, until overruled.16 “Binding” precedent—from factually similar cases in the common law setting and from prior interpretations of statutes in the area of legislation—must either be followed, distinguished, or overruled. Binding precedent is a presumptive but not an absolute constraint on what courts may do. When the source of law is the common law, as opposed to interpretation of a statute or a constitution, it is often said that a case is squarely on point only if the facts of the case are sufficiently similar and the general principles necessary to the decision correspond to the present case. If not, the prior case may be “distinguished,” and there is then said to be no precedent to follow.17 This is the “case system game” described by Karl Llewellyn, “the game of matching cases.”18 Indeed, it is often said that the classification of a [*PG87]legal system as common law, as opposed to civil law, is because it is based upon a system of precedent.19
In Anastasoff, the Eighth Circuit faced precedent in which, except for the fact that it was an unpublished decision, there was no question of distinguishing or the “game of case matching.” Faye Anastasoff sought relief in the district court from the refusal of the IRS to consider her claim for a tax refund. The IRS received the claim one day after the expiration of the statutory claim period, although Anastasoff had mailed the refund before the close of the three-year period. The IRS, and the federal district court, took the position that the statutory scheme requires the date of receipt, and not the date of mailing, to determine whether the claims period has expired.20
A prior, unpublished decision of the Court of Appeals for the Eighth Circuit, Christie v. United States, held squarely in favor of the IRS’s interpretation of the statute.21 Furthermore, it was the only decision within the Eighth Circuit ever to address the precise statutory question raised by Faye Anastasoff. On appeal to the Eighth Circuit, Ms. Anastasoff made no attempt to distinguish Christie.22 Rather, she contended that the court was not bound by Christie because it was an unpublished decision and thus not precedent under the Eighth Circuit rule that “unpublished opinions are not precedent . . . .”23
[*PG88] The outcome of the dispute in Anastasoff was easy for the panel. Ms. Anastasoff lost the appeal because Christie previously decided the same question against her, and that decision is controlling because a panel cannot, on its own, overrule prior precedent. That must be done, if at all, by the Eighth Circuit sitting en banc.24 Under Judge Arnold’s view, precedent is simply the record of how courts have decided prior cases, whether or not those decisions are published in a memorandum opinion. Whether fully explaining its reasoning, or only tersely so in the context of a per curiam opinion designated as not for publication, “[i]nherent in every judicial decision is a declaration and interpretation of a general principle or rule of law . . . . This declaration of law . . . must be applied in subsequent cases to similarly situated parties.”25 For the panel to ignore Christie simply because it was unpublished would mean, potentially, that like cases are not treated alike. Persons in similar circumstances may get different results with no explanation. As Judge Arnold put it, courts would essentially be telling litigants: “We may have decided this question the opposite way yesterday, but this does not bind us today, and, what’s more, you cannot even tell us what we did yesterday.”26 In other words, there could be an intra-circuit split, but one of the decisions would be “underground” law, the other, “legitimate” law.
The bulk of the opinion in Anastasoff is devoted to a theory for the invalidity of the non-citation rule. According to the court, the portion of the rule which specifies that unpublished opinions are not precedent violates the Constitution because it purports to confer [*PG89]greater power on the federal courts than was contemplated by Article III.27 Specifically, the Court stated “The Framers of the Constitution considered these principles [which form the doctrine of precedent] to derive from the nature of judicial power, and intended that they would limit the judicial power delegated to the courts by Article III of the Constitution.”28 Anastasoff attributes to the Framers an ideal of the constraints under which courts would operate.
In support of this proposition, the opinion provides an interesting look at the doctrine of precedent in English and colonial practice immediately before the ratification of Article III. The basic argument rests upon the proposition that the Framers accepted a particular understanding of judicial power—one that considered the doctrine of precedent to be essential—when they constructed the judicial power in Article III. The court’s review of historical material was central to its originalist interpretation: “The Framers accepted this understanding of judicial power (sometimes referred to as the declaratory theory of adjudication) and the doctrine of precedent implicit in it.”29 That is, the Framers intended to confer only this power on the federal judiciary, a power that “the judge’s duty to follow precedent derives from the nature of the judicial power itself.”30 And though we might point out that judges in the Founding period accepted a wider range of acceptable sources of law than would our own era,31 the evidence suggests, as Judge Arnold noted, that “[t]he Framers thought that, under the Constitution, judicial decisions would become binding precedents in subsequent cases.”32 Thus, the source of the requirement that federal courts consider their prior decisions as precedent comes from Article III, not from a general policy or tradition as usually portrayed.33
[*PG90] I have referred to the court’s underlying method as originalist, meaning the method that uses history to interpret the Constitution in order to gain some understanding of what the Framers’ intended when they chose the words that they did. Under this theory the Framers’ intent is controlling. The text itself is of binding authority. Judicial power, however, is never defined in the Constitution. In cases of unclear text, originalists consider the intentions of the drafters to be binding, although such efforts often reveal the extreme difficulty in practice of either ascertaining original intent or functioning without an explicit statement of it. There are, of course, extreme forms of originalism such as “strict textualism” and “strict intentionalism.”34 Judge Arnold’s opinion appears to follow—by necessity—what Paul Brest has described as “moderate originalism,” because he is more concerned with the adopters’ general understanding than with their specific intentions.35
With an originalist approach, even assuming one has identified the relevant categories of persons whose thought processes are binding, all questions of specific intent are difficult, if not impossible, to reconstruct later. Few will find this surprising given that originalist conclusions claiming to find an intent or an understanding are inherently precarious. Inquiries seeking original understandings necessarily determine which of various competing groups hold the views that matter. Originalists must follow some historical method to determine what these relevant groups thought or believed, often when no direct evidence on the specific point exists.36
We cannot know much more about what the Framers thought about precedent because no one at the Constitutional Convention, or afterward in the ratification effort, seems to have said anything specific about it. Perhaps some features of the judicial power—such as that courts would generally rely upon precedent in deciding cases—were so fundamental that the Framers’ assumptions never became issues of debate. In any event, in Anastasoff the doctrine of precedent is [*PG91]an assumed feature of their belief system, based upon the prevalence of the idea at the time. One wonders, for example, whether an open and notorious refusal to take note of precedent would have been considered an impeachable offense, given that “the constitution, by a positive injunction, prescribes the duty of the judiciary department . . .[by] prescribing limits to its authority, which if violated would be good cause of impeachment, and of removal from office.”37 No one addressed that specific question either, so far as I can tell.
When one engages in an originalist inquiry, the use of history is unavoidably selective: we select those parts that we believe aid us in answering the question about the original meaning of the language chosen for Article III. The historical evidence marshaled in Anastasoff highlights the best of the secondary evidence of what the Framers and ratifiers would have assumed. Although I do not restate that evidence here, it included the struggle for independence of the courts and supremacy of the common law over rival systems. One could also note, however, other evidence from the ratification debates that supports Anastasoff’s general point. Both Brutus and the Federal Farmer viewed the tendency of common law courts to move toward “rigidity,” presumably from their practice of following precedent. The Federal Farmer warned that “the rigid systems of the law courts naturally become more severe and arbitrary, if not carefully tempered and guarded by the Constitution, and by laws, from time to time.”38 Similarly, Brutus noted the tendency for principles that form court decisions to “become fixed, by a course of decisions . . . .”39 Furthermore, within fifteen years of ratification of the Constitution, William Cranch wrote in the preface to his reports of Supreme Court decisions:
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. Every case decided is a check upon the judge. He can not decide a similar case differently, without strong reasons, which, for his own justification, he will wish to make public. The avenues to [*PG92]corruption are thus obstructed, and the sources of litigation closed.40
On the other hand, skeptics of the determinacy of originalist interpretations might point to contrary evidence on the issue of stare decisis as a mandatory feature of the federal judiciary. One might point out that the debates over Article III were not about how courts would decide cases but were concerned with length and tenure of office, manner of appointment, removability, jurisdiction, how many courts and what kind, and other structural issues.41 For these reasons, one might argue, it is a stretch to add “respect for precedent” to the mix of issues that judicial power was viewed by the Framers to be about. Indeed, the term judicial power appears twice in Article III—not just in Section 1,42 cited by Anastasoff, but also in Section 2: “[t]he judicial Power shall extend to all Cases, in Law and Equity, arising under this Constitution . . . .”43 In this latter occurrence of judicial power, it is quite clear from the proceedings of the Constitutional Convention that judicial power was considered to be interchangeable with the original phrasing of that clause: “The jurisdiction of the Supreme Court shall extend . . . .”44 The “jurisdiction” language dominated what is now Article III, Section 2, throughout the constitutional debate, and likely was changed to “judicial power” near the end of debate merely as a matter of style. In terms of the final structural result, the judicial power in Section 1 vests, while the judicial power in Section 2 extends. The court in Anastasoff chose Section 1 as the source of the understanding that courts must accord at least some precedential value to their prior decisions. But it also could have cited Section 2, because judicial power in Section 2 of Article III seems to have a similar meaning, although more closely tied to court structure rather than function.
I have no illusions about determinacy in originalist interpretations of the Constitution. Yet one could judge favorably Anastasoff’s historical method by a preponderance of the evidence standard. While I recognize there are points on which one might quibble with the historical record presented in Anastasoff, there is ample evidence [*PG93]to prove Judge Arnold’s general point.45 It is reasonable to conclude that informed thinkers in the Founding period expected a doctrine of precedent to work at least some minimal degree of constraint on the methods all courts would use to decide controversies.
I also propose, however, that one might support the constitutional argument in other ways. There are also other methods of constitutional interpretation, including versions of textualism and those that focus upon structure and function of the Constitution as a whole. There have been some suggestions that due process and equal protection may support the notion that courts must consider the precedential value of unpublished opinions, although no court as of yet has agreed.46 In the next section, I survey several aspects of nineteenth-century jurisprudence in order to propose that the understanding of judicial power suggested in Anastasoff may be supported by other interpretive methods as well.47 In particular, I argue that this core notion of beginning with precedent has been consistently followed for over two centuries, most importantly by the post-Founding generations. To be sure, the historical record reveals significant disagreement at various points concerning the binding nature of precedent. But prior to the appearance of the non-citation rules at issue in Anastasoff, no one had suggested that courts should not at least begin their reasoning process with prior decided cases. In section III, I use these historical illustrations and arguments to suggest why we are only now considering the extent to which Article III dictates to courts the use of precedent as a starting point for decision.
Unless the rule of stare decisis is adhered to in the administration of justice under a government of laws, all property must be rendered insecure.
—Jones v. Anderson (1808)48
The following survey of three important aspects of nineteenth-century jurisprudence shows continuity over time in the Framer’s core notion of precedent. I discuss, in turn, vested rights and protection of property, the codification debate, and the debate over general federal common law. Each of these issues affected the development of American law generally and the development of doctrines of precedent specifically. I chose vested rights and protection of property because in the immediate post-Founding period we see courts articulating a rationale for adherence to precedent specifically linked to protection of private property rights. I chose the codification debate because, over a period of several decades, proposals for codification forced debate about the nature of the common law, and implicitly, the doctrine of precedent within it, while in the end the core notion of precedent survives. I chose the third topic, the nineteenth-century question of the existence of a general federal common law, because the issue led to acceptance of the idea of separate common-law jurisdictions within one nation—a blow to the theory of the common law as one system of general principles existing apart from case law.
Taken together, these three discussions, I suggest, show that the core idea of precedent that Anastasoff attributes to the Framers has been fairly consistent over time, despite changed perceptions of the nature of the American common law. These discussions also show that when historians locate the beginnings of a strict, traditional doctrine of precedent only in the nineteenth century, they are not disputing what I term the core idea of precedent that Anastasoff claims. Indeed, it is the effect of these historical episodes on the question of the binding nature of precedent, not precedent as a starting point, that has caused historians to locate different dates for the beginning of the “strict” notion of precedent.
By the early decades of the nineteenth century, the doctrine of precedent took on a particularly urgent role. Courts articulated a [*PG95]strict doctrine of precedent in order to stabilize property rights. Judges frequently spoke of the necessity to follow precedent because of settled expectations about property.49 One of the earliest statements reflecting this view came from Pennsylvania. The 1808 case of Jones v. Anderson, quoted at the outset of this section, specifically held that without the rule of stare decisis all property is rendered insecure.50
Another early view was expressed in an 1823 decision from New York. In Lion v. Burtiss, Chief Judge Spencer wrote: “Stare decisis is a maxim essential to the security of property; the decisions of courts of law become a rule . . . and where that rule has been sanctioned and adopted in our courts, it ought to be adhered to, unless manifestly wrong and unjust.”51 This “maxim,” from the New York decision announcing it, is frequently cited in other courts throughout the nineteenth century.52
Protection of private property ranked high among the fundamental rights. In an argument before the United States Supreme Court in 1849, counsel urged that the Court had only to “adhere to the just rules already laid down, to practise [sic] the great maxim which secures respect and renders certain the rights of property and life, Stare decisis.”53 Among the recognized elements of a case demanding the support of the doctrine of precedent, said the Court, is “especially, and above all, its constituting a rule of property or becoming a muniment of title to it.”54
[*PG96] Judicial rhetoric links the doctrine of precedent with protection of settled property expectations throughout the nineteenth century.55 Indeed, attorneys considered the doctrine of precedent to be a valuable tool for arguing cases. In an Alabama court, an attorney urged that “[t]his decision has become a rule of property, and has been uniformly acted upon as such in this State for nearly thirty years. It is too late to disturb it. The maxim, ‘stare decisis,’ must apply here or be repealed.”56
The notion of a judiciary geared to protect economic interests is generally consistent with the reigning paradigm of interpretation for this period—a paradigm which emphasizes the role of economic thought in shaping judicial decisions.57 Another explanation for the new emphasis on protection of property, apparent by the early decades of the nineteenth century, may be the larger context of the development of constitutional theory in the post-Founding period. One important strand of nineteenth-century American constitutionalism was the doctrine of vested rights, rooted in a higher law tradition.
The case that best exemplifies the vested rights doctrine is Fletcher v. Peck, probably best known as the case first used by the Supreme Court to extol the sanctity of vested rights in property and thus to secure broad meaning to the contract clause.58 In Fletcher v. Peck, the Supreme Court ruled that the state of Georgia had infringed a valid contract for property.59 For the first time, the Court expressly held state legislation invalid because it conflicted with the federal Constitution.60 According to Chief Justice Marshall, a government must be[*PG97]have under the limitations of a private person.61 If a private person had granted land, there is no remaining right to take it back once it has “vested.”62 A government has no greater right. 63 As Chief Justice Marshall wrote in Fletcher: “It may well be doubted whether the nature of society and of government does not prescribe some limits to the legislative power; and if any be prescribed, where are they to be found, if the property of an individual fairly and honestly acquired, may be seized without compensation.”64
The ascendancy of substantive protection for property is also evident in Gardner v. Village of Newburgh.65 Chancellor Kent considered the prohibition of a government’s taking private property without just compensation to be “a great and sacred principle of private right,” even though the New York state constitution lacked an express takings provision.66 This period is notable for judges’ tendency to define property expansively and to protect it against government intrusion based upon natural-law-type arguments.67 The Supreme Court even suggested that there was a constitutional right to rely upon precedent when private property is at issue. In Gelpcke v. City of Dubuque,68 the Court in effect said that an unconstitutional taking of property occurred when railroad bonds were issued in reliance on precedent and a state court later disavowed that precedent.69
The severity of the Court’s willingness to protect private property from government takings in this period contributed to the greatest crisis in the nineteenth century: The conflict over slavery. Emancipation, it was commonly believed, could not be legally accomplished without government compensation to slave owners.70 The most extreme example of the vested rights doctrine was the suggestion that prohibiting slave owners from taking their slaves into United States [*PG98]territories violated their property rights.71 It is no surprise, then, that the doctrine of precedent is shaped to buttress the protection of private property.
Thus, in at least one general area of law—property rights—courts express an urgency for the doctrine of precedent. This is apparent in the early decades of the nineteenth century. Numerous cases, state and federal,72 emphasize the doctrine of precedent as essential to the security of private property.73 We also know that the Framers worried about property rights and how best to preserve them. The emphasis and rationale for the doctrine of precedent is consistent with this view.
It is true that the compulsion to follow precedent, by the early nineteenth century at least, depended to some extent upon the subject matter involved. Courts noted with some frequency that, when the rule of a prior case involved something other than property, they were less constrained by the doctrine of precedent to follow it.74 Property reliance interests received significant judicial notice, but that fact does not mean that courts were not concerned about precedent in other contexts. These varying emphases fall within Lord Mansfield’s admonition, in 1762, that “the reason and spirit of cases make law; not the letter of particular precedents.”75 Whatever the degree of concern to follow precedent, one had to begin with it.
When the original understanding of intent of the Framers is not clear from debates and contemporaneous writings, then the behavior of the participants, after but sufficiently close in time, should also be [*PG99]considered persuasive.76 The immediate post-Founding generation behaved as though starting with precedent was a natural obligation of courts. To the extent the post-Founding era had forgotten the Framers’ intent, or that it was “lost from memory” as Grant Gilmore might have said,77 they behaved in a way consistent with Anastasoff’s view of judicial power in Article III.
The codification movement, a feature of legal debate in the mid-decades of the nineteenth century, was the first sustained, critical examination of the common law system in the United States.78 It was pervasive. As one scholar has noted recently, “[t]hroughout the nineteenth century Americans regularly debated whether to reduce the common law to a written code.”79 Proponents of codification had various aims: Moderates thought that the common law, or at least some portions of it, might be restated and clarified through reducing it to a code of general principles.80 Radical codifiers, on the other hand, advocated a replacement of many of the common law’s provisions with new provisions contained in a code, no doubt influenced by the Benthamite utilitarian ideas propounded earlier.81 At least some proponents of codification explicitly wanted to adopt a civil law system like that in Europe, and in particular something similar to the Napoleonic Code.82 It is probably true that the decades of the codification movement are the closest the United States ever came to adopting a legal system different from that of its mother country.83 [*PG100]The Field Code of Civil Procedure, precursor to the modern Federal Rules of Civil Procedure, is the one great success story of nineteenth-century codification zeal.84
Despite the anti-English rhetoric in some circles during the early years of the confederacy and the Republic, there was never any serious doubt that the states and the federal courts would receive the common law of England, meaning its method more particularly than its substantive rules.85 The method, of course, was that courts would generate a body of law through case-by-case adjudication. The moderate codifiers wished to preserve the basic common-law approach, but they hoped to clarify it by codifying certain elements. Indeed, the limited success that the codification movement can claim in California, whose private law is more codified than some nations, nonetheless is a fundamental affirmation of common-law method.86 Similarly, in other states which codified large portions of their common law, the method that courts used to interpret these new codes did not change.87 That is, courts continued to view prior cases as the primary source of law in traditional common-law subjects. The new codes were merely evidence of what the common law was, but the courts themselves were free to continue developing this body of law on a case-by-case basis.88
It is probably true that the debates about codification—beginning as early as Bentham’s criticisms that the doctrine of precedent does not constrain judges and that no one can know in advance what [*PG101]the law is—coincide with the transformation of the view that prior cases are themselves sources of law, not merely evidence of the “true” law. At the time of the Framers, prior cases, though considered precedent, were authority because they were evidence of the “general principles” of the common law.89 The challenge of the codifiers brought into debate the conservative response that judicial opinions, case law, were themselves sources of law. A single decision was, in a sense, a mini-statute because of stare decisis. Those who wished to preserve the common-law system from a complete codification helped solidify the theory of a case as a source of law, a step better than consulting precedent merely because it is evidence (which may be wrong) of the true law.
In Georgia, for example, a legislative act in 1858 prohibited the overruling of any prior decision of the Georgia Supreme Court.90 Such decisions were to be considered the law of the state “where they have not been changed by the legislative enactment.”91 Three years later, a revised statute permitted the overruling of prior decisions, but only by “the full bench.”92 Commenting upon the authority of prior case law in 1866, the Georgia Supreme Court stated:
When a question has once been decided by this Court, we desire it to be distinctly understood that such a decision is, with us, authority. If counsel can furnish us a decision, of this Court, in point, he need go no further in his investigations. It is unnecessary to consume his own time and ours, in arguing that such a decision is the law. With us such decision is conclusive of what the law is, until changed by the law making power.93
The statute prohibiting overruling prior cases, as well as the 1866 admonition by the Georgia Supreme Court affirming its view of precedent, are contemporaneous with the first successful codification of common law in Georgia in 1860.94
Thus, the codification debates resulted in an affirmation by all but the most radical elements that the American legal system should [*PG102]remain distinctively common law in its methodology. Legislatures considered and ultimately rejected the idea of replacing entirely the common law’s theory of case-by-case development of law and its doctrine of precedent with codes. The debates forced the conclusion of the positivist view that prior cases were themselves sources of law. They also focused the inherent tension of the common-law system: The greatness of the common law was often said to be its ability to adapt and change with changing conditions in society. On the other hand, legal positivism brought increasing awareness of the notion that judges make law rather than discover it. Supporters of codification often argued that capturing the common law in code form would enable citizens to better know what the law is. Furthermore, in theory at least, courts cannot overrule portions of the code with which it may come to disagree.
It seems reasonable to conclude that the moderate codifiers, wary of the commitment of the judiciary to stare decisis, hoped to commit the Framers’ vision of precedent to statutory form. While codification in the nineteenth century never succeeded on the scale that its original proponents hoped (though it did have some success, more particularly in the early decades of the twentieth century), their goal continued in other forms.95 Langdell’s scientific approach to law, an attempt to reduce cases to black-letter principles, was a late nineteenth-century precursor to the Restatements of Law. Both Langdell’s quest for scientific principles, and the Restatements, are essentially an alternative method of codification—stating the law in a clear fashion so that courts might more easily follow it and less easily strike off on a different path.
Once again, although the participants in these codification debates argued about stare decisis,96 or the binding nature of prior precedent, there was little quarrel with the basic proposition that, as a matter of method, courts must approach their decisionmaking process by at least consulting prior cases.
There are several ways in which our contemporary vision of the common law and its method differs from the past. One of the most [*PG103]interesting differences between the nineteenth and twentieth centuries is the vision of common law rejected by Erie Railroad Co. v. Tompkins97—the case that ended the Swift v. Tyson line of jurisprudence recognizing a federal general common law.98 The specific holding of Erie was that federal courts must apply the substantive common law of the state in which they sit, in the absence of federal law to the contrary.99 For almost one hundred years, federal courts had claimed authority to disregard state common-law decisions for areas of law deemed to be governed by “general” law, a doctrine most closely associated with Swift.
The debate about the common-law jurisdiction of the federal courts was a long one. One aspect of the debate, however, received early resolution. In 1812, within a generation of the Framers, the Supreme Court held that federal courts had no authority to create or recognize common-law crimes.100 The reason had to do with the peculiarities of the limited federal government in a federal system. Federal courts lack power to find sources of law in a body of customary law collected in prior court decisions, absent specific statutory authority to do so.101 It took state judges significantly longer to yield to legislatures their power to define crimes.102 The question returned in another guise in the years leading up to Swift. In 1834, the Court stated in the case of Wheaton v. Peters: “It is clear, there can be no common law of the United States.”103
For my purposes, the well-traveled area of Swift, Erie, and the continuing presence of pockets of federal common law are important, not for the federalism issues that they raise,104 but for the particular view of common law over which Swift held sway for portions of the [*PG104]nineteenth century. As understood by Swift, the common law was a body of general principles for a court to interpret. Swift represents the idea of a “general” common law, a law that exists apart from individual, past decided cases, which may themselves be “wrong” interpretations of a customary law. Swift may have represented competing notions of governing law, but it is consistent with the idea that precedent remained important within each system.
In Swift, Justice Story envisioned a general commercial law—a customary body of rules common to many jurisdictions. One scholar has noted that “[a]t the time Swift was decided, the states did not clearly conceive of ‘general commercial law’ as state law.”105 Indeed, in the early part of the nineteenth century, both state and federal courts decided cases under a general, unwritten system of common law, guided by former decisions which did not themselves have force of law.106 Over time, and until the decision in Erie, federal courts claimed more areas in which to apply this “general common law.”
In the decades following Swift, however, more and more jurists began to view the common law as a creation of sovereign will.107 It became common-place to view the common law in a legal positivist sense—it varied from state to state because the law consisted solely of past court decisions, not some “true” or universal common law existing apart from what courts decide in individual cases. Unwritten law, which the common law was widely said to be, had become “written” law at least in the sense that jurists recognized that courts were, in effect, issuing individual pieces of legislation. Although piecemeal and gap filling, not to mention wholly subservient to the legislature, courts were recognized as making law, and as they did so within their respective sovereign spheres there were bound to be differences.
The Swift phenomenon adds yet another complication to our attempt to understand precisely what past generations meant when they considered that courts would follow precedent. And in a sense, it is a manifestation of the very contradictions addressed by the codifiers— the larger quest to articulate predictability, simplicity, and transparency for the common law as a legal system. On the other hand, if one views prior case law merely as “evidence” of what the common law is, prior cases can be “wrong” more easily than if prior cases are viewed [*PG105]in the strict positivist sense to be themselves sources of law.108 That is, the justification for departing from prior precedent changes quite dramatically depending upon one’s view of the prior cases as themselves a source of law.
But it is precisely this feature of the debate—whether cases themselves are considered to be binding sources of law—that has caused historians to select various historical dates for the beginning of the so-called “traditional” doctrine of precedent. What the review of history has shown is that today we tend to conflate the ideas of precedent and stare decisis. The two are not entirely the same, however. Stare decisis is commonly used to refer to a strict sense of the degree to which courts are to be bound by prior precedent. Stare decisis, meaning a strict adherence to the doctrine of precedent, is a feature of the legal formalist period predominant in later parts of the nineteenth century.109 With the possible exception of this formalist period, in the United States there has never been any clear point of agreement concerning stare decisis.110 Yet whatever view one takes about the desirability of stare decisis, the core notion of precedent as a beginning point has held true.111 However courts viewed the common law—as evidence of a “true” law or in a more positivist sense—no one suggested that individual court decisions were not precedent, and no one suggested that courts did not at least begin the decision process with a consideration of what courts had done in the past.112
[*PG106] At no point during this debate over whether cases are a binding source of law was there a widespread attempt to justify precedent as carrying on the traditions of the Framers. It never would have occurred to the next generations to ask this question because the basic methods of the common-law system were never challenged. They probably either assumed that they were carrying on the traditions, or were simply not interested in the issue. The point is, these jurists behaved exactly as if Anastasoff’s understanding of judicial power was their own. There were no rules or practices by which courts would have designated any of their prior decisions not to be worthy of consideration, at least as a starting point, for future cases.
The preceding survey suggests that, at least in these snapshots of the nineteenth century, the doctrine of precedent served various needs and functions since the Founding period. Thus, it is of little concern to me that the nineteenth century is sometimes characterized as one of loose adherence to precedent.113 Even if this characterization is generally true (and I suggest the property cases, at least, cast doubt on this assertion), this fact does not undermine Judge Arnold’s characterization of the importance of precedent around the time of the ratification of Article III.
While it is not necessary to rely solely upon an originalist argument to reach the same result, I do not suggest that originalism was the wrong approach in Anastasoff. It is, however, instructive to consider alternative methodologies. These methodologies could also be used to argue that judicial power in Article III assumes a bare minimum of consideration of prior precedent. The behavior of the post-Founding generations is consistent with this view of judicial power, and this fact should be persuasive because it reflects the participants’ understanding of what judicial power required.
Furthermore, beyond the post-Founding generations, at no point in American legal history has this core idea been seriously questioned. Federal courts, and all state courts (with the possible exception of Louisiana), have never viewed themselves to have the power to change the fundamental nature of the American legal system. Prior to the advent of the non-citation rules for unpublished opinions, no court has ever claimed that it could disregard entirely a prior judicial decision in a similar case. The core idea of common-law court systems is that what courts have done in the past, to some extent and to some [*PG107]degree, must at least be considered when a similar case comes along.114
There is now wide agreement that a judge can and should participate creatively in the development of the common law.
—Justice Roger Traynor115
Anastasoff characterizes how the average jurist understood the doctrine of precedent in the late eighteenth century. We seem to understand it differently today, if the prevalence of court rules concerning the precedential value of unpublished opinions is any indication. Most of the rules have come into being in the last several decades116 in response to dramatically rising caseloads.117 Many of these rules specify that unpublished opinions are not precedent.118 In the federal courts, nearly eighty percent of all appellate decisions in 1999 were designated not for publication.119 In some state courts, well over half of the court’s decisions fall into this category.120
The review of history in the preceding section and in what follows supports two propositions. First, the new non-citation rules that consider some judicial decisions not to be precedent are an aberra[*PG108]tion in our historical practice. Second, jurists have always understood the doctrine of precedent to require at least some initial reference to a prior-decided, similar case, but the American doctrine has always been flexible. Moreover, the binding force of precedent—at its extreme, the doctrine of stare decisis—has never meant precisely the same thing in any given era. Regardless, jurists have continued to follow what I refer to as the core doctrine of precedent. What a court has done before has always been considered at least relevant. Anastasoff does not require that courts subscribe to any one particular theory of the binding nature of precedent.
Anastasoff also stresses that it is not attacking the practice of withholding some opinions from official publication.121 These rules came into being with several purposes, many of which are legitimate concerns quite apart from the question whether unpublished opinions hold precedential value.122 But courts may not ignore the fact that a case was decided a particular way in the past. We now must consider the argument that, at least for the federal courts, the rules are beyond the courts’ judicial power to the extent they deny the precedential value of prior decisions. Somewhere between the Founding and the cavalier attitude toward the doctrine of precedent exemplified by non-citation rules, we seem to have lost our way.
We are tempted to lay the blame for the disjuncture between the Framers’ supposed beliefs about precedent and our own squarely at the feet of the Legal Realists. Karl Llewellyn understood the dynamic nature of precedent,123 and other realists viewed precedent to be [*PG109]infinitely malleable and denied that law could be derived from it.124 Because “we are all legal realists now,”125 we, too, tend to be skeptical about the constraining force of precedent.126 Clever lawyers, we know, can always distinguish cases or otherwise convince a court that the principle derived from a prior case was not necessary to its decision. Although Legal Realism may be the most recent influence on our thinking about precedent, many of these issues had been raised before by Jeremy Bentham127 and others. As I have noted, there were earlier points at which jurists and lawyers began to think differently about the nature and methods of the common-law system. These episodes occurred significantly closer in time to that of the Founders and perhaps cast some light on the views of that earlier generation.
The point I wish to make here is that Anastasoff, and its claim about the requirements of judicial power, is not about the binding nature of precedent. Instead, judicial power means merely that there is a bare minimum—a core idea of precedent. Courts in common-law systems traditionally have understood their decisionmaking process to embody this core idea. The decisionmaking process at least begins from prior precedent, whether the court then considers itself “bound” by precedent, able to “overrule” precedent, or whether the court chooses one of Karl Llewellyn’s sixty-four possible treatments of precedent.128
The second line of attack on Anastasoff’s interpretation of Article III, I suggest, will be directed at the future consequences if other courts follow the lead of Anastasoff. One such criticism is likely to be that it is impossible to define any doctrine of precedent inherent in the judicial power of Article III, because to do so must mean that courts are now constitutionally required to follow prior cases. And if this is the significance of Anastasoff, have we not long ago discarded the idea that precedent is ever binding on a court if the court wants to [*PG110]decide a case in a particular way? In other words, we might ask what is the fuss about whether these cases are to be considered precedent, if binding decisions are not really binding—if courts can distinguish between “narrow” holdings,129 “central” holdings,130 “essential” holdings,131 not to mention cases that are “not controlling in the strictest sense”?132 Some academics have long suggested the principle of stare decisis is ignored or at least in serious decline.133 Others argue that the doctrine of precedent is, in fact, the “workhorse” of the system for the majority of “unexceptionable” cases.134 If anything, in the age of statutes it has become easier for courts to adhere to precedent. When interpreting statutes, courts rarely must reason by analogy in light of prior cases and their principles. In Anastasoff, for example, there was no possibility to distinguish Christie. A panel of the Eighth Circuit had previously interpreted the statutory scheme to require a “receipt” rule rather than the “mailbox” rule, and subsequent panels must follow that interpretation until overruled by the court sitting en banc.135
Another variation of the future consequences criticism of Anastasoff will be that these rules are necessary for the court to manage its workload. Some have suggested that the work of courts will grind to a halt if the nearly eighty percent of unpublished decisions now rendered by courts must be considered more carefully.136 Other solutions to the workload problem (appointment of more judges, or a certiorari-type jurisdiction) are not within the courts’ control.
[*PG111] I suggest all of these prospective dire consequences of Anastasoff are a misreading of Judge Arnold’s opinion, which holds that courts must seriously consider how they have decided a like controversy in the past before rendering a new decision in a similar case. Courts are not free to determine that some of their judicial decisions are not precedent. To the extent the result is that courts must now take more time with the kinds of cases previously designated not for publication, the remedy must be legislative, not judicial.137 It is certainly a serious problem courts face. But other values outweigh the need for courts to adopt the cavalier approach to decisionmaking reflected in the non-citation rules. These include the value of transparency in judicial decisionmaking, avoidance of an “underground” body of law that suggests some litigants get lesser justice than others, and the effort, at least, to decide like cases alike. In fact, the only way a court can know that it is treating like cases alike is through the discipline of a doctrine of precedent.
One justification for non-publication rules has been as a matter of convenience for courts. Many cases are routine, so the argument goes, and thus lengthy, reasoned opinions are unnecessary to dispense justice. These cases do not alter precedent—they merely affirm it. If this is universally true, then Anastasoff will have no effect on the continuation of this practice. Nothing in Anastasoff’s reading of Article III prevents courts from designating cases not for publication and deciding them in a short, per curiam fashion. What they may not do, however, is pretend that those cases never happened.
But the situation in Anastasoff itself belies the argument that all of these unpublished decisions are, in fact, unimportant from the perspective of determining what is the court’s precedent on a prior topic. [*PG112]The issue would never have arisen in Anastasoff if Christie fit this description of a per curiam opinion merely affirming, in summary form, a long line of already established precedent. Christie was the only Eighth Circuit decision to interpret this particular statutory problem, and thus arguably it should never have been designated not for publication. At least one other circuit, in fact, has disagreed with Christie’s interpretation.138 To reverse Christie would require a majority vote of the en banc court. If Faye Anastasoff’s position were followed, the panel could hold the direct opposite of Christie and have no obligation to justify the difference. The ability to cite unpublished opinions may be the only way to uncover conflicts in the circuit.139
Anastasoff itself thus provides compelling evidence that we have little assurance that unpublished opinions do not create the dangers we fear: That courts will interpret statutes, for example, one way one day and a different way the next, with no accountability. Worse, perhaps, is the fact that we can have no assurance that the Christies and the Anastasoffs of the world are treated to the same set of laws. If they are not, why not simply have a lottery system to determine the outcomes of these disputes?
Another justification for barring citation to unpublished decisions has been accessibility. Opinions not designated for official publication were, at one time, available only from the clerk’s office. A litigant with easy, and nearly exclusive, access to unpublished opinions was thought to have an unfair advantage. Now, however, the opposite is true: The so-called “unpublished” opinions are widely available on electronic databases, including the internet.140 Therefore, this “unfair advantage” justification for declaring that unpublished opinions have no precedential value has dropped out.
I want to propose one further, related question: Assuming we agree that an appropriate interpretation of judicial power in Article III includes some respect for precedent, how is it that we are only now constitutionalizing this link? Although courts have considered the term “judicial power” in Article III to impose constraints in other ar[*PG113]eas,141 no one has suggested before that the notion of judicial power includes an obligation to respect precedent, however undefined this obligation may yet be. It is thus a new theory of the meaning of judicial power. Assuming this is a plausible construction, why, then, has it taken two centuries for the doctrine of precedent to be linked to judicial power in Article III? If the original understanding of judicial power ascribed to the Framers in Anastasoff is correct, why do so many find this notion surprising today? In short, how did we come to forget what was seemingly so clear to the Framers?
To think about these questions, I have suggested that it is helpful to consider the post-Founding role of precedent in American legal thought. From that consideration, straightforward reasons emerge for this seemingly late realization about precedent’s central position in judicial power. It would never have occurred to a jurist of a prior generation to ask whether the basic operation of judicial power included a requirement that it begin its consideration of a current case with a view of what it decided in a past similar case. The reason is that courts have never before considered that any of their prior decisions could be irrelevant to future cases. The spate of court rules declaring non-published opinions not to be precedent are a radical departure from any court practice of the past. This is true even in periods in which courts have not felt themselves strictly bound by past precedent.
If it is apparent that the doctrine of precedent has been cyclical in the degree to which it might be considered “foundational,” it is also apparent that views of what I call the “departure” power—that is, the circumstances under which courts may refuse to follow precedent—have changed in response to various perceived needs. In other words, the circumstances under which courts have believed it to be appropriate to overturn prior case law have also changed over the course of time. Often this cyclical approach is associated with predominance of one or the other of the competing views about the nature of the common law: Is it merely a collection of precedents or a method of analysis?
[*PG114] Justice Traynor, in the quotation set out at the beginning of this section, expresses a view consistent with the claim that courts in the United States are not now particularly disposed to follow a strict doctrine of precedent. On the other hand, Traynor also wrote that in contrast to the legislative process, “a judge invariably takes precedent as his starting-point; . . . [s]tare decisis signifies the basic characteristic of the judicial process that differentiates it from the legislative process.”142 A few decades earlier, Benjamin Cardozo stated “that the rule of adherence to precedent, though it ought not to be abandoned, ought to be in some degree relaxed.”143 Judge Arnold in Anastasoff wrote that he was not advocating “some rigid doctrine of eternal adherence to precedents.”144 The model for such a rigid doctrine might be the British House of Lords, which before 1966 took the position that in its judicial capacity it could not overrule prior precedents.145
The rigid model that the English system became has been rare, if non-existent, in court practices in the United States. Instead, in some periods courts have celebrated the adaptability of the American common law. There are many theories about how and why common law in the United States has changed. Judge Richard Posner is a proponent of an “efficiency” thesis, positing that the common law tends to “evolve” in ways that favor the economically efficient outcome, whether or not consciously intended by judges.146 Justice Holmes was also interested in the process of change and how principles of the ancient common law evolved into modern law.147 Holmes reminds us that precedent is not rigidly binding on future courts and for many [*PG115]reasons should not be. Our comfort level with that idea depends very much on our conception of what policies are appropriate for judges to pursue.
For several reasons, however, I suggest that even this lax regard for precedent is nonetheless consistent with Judge Arnold’s view of judicial power in Article III. In each of the historical discussions in the prior section, the constraining features of precedent underwent some change. The result is that we should not expect any consistency over time in the doctrine of precedent as understood by the Framers, but only in the core notion that a court must at least consider how it has addressed a similar problem in the past.
Returning to the questions posed at the outset: If Article III restricts the extent to which judges can disregard precedent, then must it also have a compulsory side and not merely a negative one? If we say that Article III requires courts to consider as precedent unpublished opinions—that is, they may not declare any of their prior decisions not to be precedent—then does Article III also define the extent to which courts are required to follow precedent? That is, must we have a uniform, normative vision of the circumstances in which a court may depart from precedent?
To some extent the court in Anastasoff avoided the issue because it could not overturn the prior precedent at issue in that case. As Judge Arnold noted, a three-judge panel of the court could not overturn prior precedent—that is an option only available to the court sitting en banc. Thus the panel is bound by the unpublished decision that its own rule would classify as not precedent. Nonetheless, Anastasoff cautions that it is not “creating some rigid doctrine of eternal adherence to precedents.”148 Sometimes, said the court, cases should be overruled: “If the reasoning of a case is exposed as faulty, or if other exigent circumstances justify it, precedents can be changed.”149 When a court determines that a precedent should be changed, however, there is a “burden of justification” in which the reasons for rejecting it “should be made convincingly clear.”150 There is little guidance, then, from Anastasoff about how rigid the constitutional doctrine of precedent may be. For instance, is it beyond the scope of judicial power to overturn a prior line of cases without fulfilling this “burden of justification”?
[*PG116] One way to consider this question might be to attempt to reconcile the various rationales articulated by courts for overturning precedent. The most wide-ranging recent summary is provided by the plurality opinion in Planned Parenthood of Southeastern Pennsylvania v. Casey.151 Justices O’Connor, Kennedy and Souter stated that “[t]he obligation to follow precedent begins with necessity, and a contrary necessity marks its outer limit . . . . Indeed, the very concept of the rule of law underlying our own Constitution requires such continuity over time that a respect for precedent is, by definition, indispensable.”152 The doctrine of precedent, the joint opinion suggested, is animated by “principles of institutional integrity.”153
On the other hand, “the rule of stare decisis is not an inexorable command.”154 Casey suggested a balancing test to determine the appropriateness of overruling prior case law:
[W]hen this Court reexamines a prior holding, its judgment is customarily informed by a series of prudential and pragmatic considerations designed to test the consistency of overruling a prior decision with the ideal of the rule of law. . . . Thus, for example, we may ask whether the rule has proven to be intolerable simply in defying practical workability; whether the rule is subject to a kind of reliance that would lend a special hardship to the consequences of overruling and add inequity to the cost of repudiation; whether related principles of law have so far developed as to have left the old rule no more than a remnant of abandoned doctrine; or whether facts have so changed, or come to be seen so differently, as to have robbed the old rule of significant application or justification.155
Casey suggests that these circumstances under which a court may overrule prior case law would apply whether the issue to be decided is one of common law, statutory interpretation, or constitutional adjudication, although the Supreme Court has suggested in the past that precedent carries less weight in constitutional adjudication than in [*PG117]statutory interpretation.156 Ironically, stare decisis carries even less weight when the issue involves a mere “procedural rule.”157 The traditional view, expressed in both Anastasoff and Casey, is that the decision to overrule is a matter of propriety for a court—what it should not do absent compelling circumstances. Thus, “a decision to overrule should rest on some special reason over and above the belief that a prior case was wrongly decided.”158
Presumably the factors summarized by the joint opinion in Casey will provide the other part of the equation, that is, when a court may decline to follow precedent that it has found to be relevant. In other words, the decision to depart from precedent will remain a prudential one, not one of constitutional proportion. We should not be surprised, however, if jurists disagree with the joint opinion’s view of the binding nature of precedent. Moreover, even if the Supreme Court were to announce that no precedent is ever purely binding even in cases of statutory construction, Judge Arnold’s point in Anastasoff remains sound. At a minimum, the doctrine of precedent requires that courts consider prior decisions, but it does not require that they follow them.
Judge Arnold does not suggest that we are equally bound by what the Framers’ may have thought about this “departure power.” He does not attempt to demonstrate that courts are today bound to follow their prior precedent to the same extent that the Framers, or anyone else, would have thought appropriate. The constitutional violation is not that courts do not follow the precedent set in their unpublished decisions. A court does not have to follow its prior precedent. Rather, the constitutional violation is when a court asserts that its decisions are not precedent and cannot be cited at all. It is a distinction with a difference.159 According to Anastasoff, courts are free to overrule precedent, but they are not free to ignore it. The Framers no doubt at least understood the common law’s ability to grow. Common law that fails to adapt to changing social conditions is quickly replaced by statute. The common law is forever weighing the old tensions seen here, between stability and change, growth and stagnation. To the extent this [*PG118]reality was unclear at the period of the Founding, I argue it was made clear in the course of events of the nineteenth century.
Our expectations of judicial power do not vary greatly from those of any prior period. We expect courts to treat everyone alike, and we expect them generally to keep the law stable. Neither of these expectations can be accomplished if courts create large bodies of law which they explicitly disavow. We expect courts to have some beginning working point to ensure some stability and some relative equality of treatment of persons. That beginning point, at least in our common-law legal system, is precedent—the record of how courts have decided issues submitted for adjudication. Courts look first to how they have decided cases in the past, not only because it saves time in the judicial reasoning process when a similar question arises again, but also to ensure that litigants close in time receive roughly the same interpretation of the law.160 Plaintiff Anastasoff should be treated like plaintiff Christie. If not, the court must at least provide a good reason. Judge Arnold’s point is nothing more than this: Non-citation rules that permit judges to decide cases one way one year, and another the next, without explaining the difference, have no part in a system that claims to operate within these basic, historical notions of judicial power. Even Justice Holmes, who was generally revolted at the notion that a court must follow precedent simply because a court had decided a question before,161 would have been revolted at the opposite notion—that a court need not even consider what it decided in a prior similar case.
The question at one level is whether one must rely on an originalist interpretation alone to reach this understanding of judicial power in Article III. I have argued that other interpretive methods support this result as well. In one sense, Anastasoff is akin to Marbury v. Madison in its assumptions about the nature of the American judici[*PG119]ary.162 Marbury was the vehicle used by Chief Justice Marshall to establish the power of judicial review—a power not clearly defined in the United States Constitution but arguably assumed there. Marbury affirmed the power of judicial review without any express provision of the Constitution. Marshall centered the power of judicial review chiefly upon the perceived purpose of a written constitution as fundamental law, and that “[i]t is emphatically the province and duty of the judicial department to say what the law is.”163 Marshall justified the exercise of the power on something very close to a fundamental understanding of judicial power, linked more closely to the existence of the written Constitution itself. But in another sense, Marbury itself supports Anastasoff’s specific result in that it establishes also the preference for written judicial opinions, so that the Court itself might know how it has interpreted the Constitution in the past.
When I suggested at the outset that respect for precedent is making something of a comeback, I did not mean to discount the intense debates about respect for precedent in constitutional adjudication,164 particularly those involving Roe v. Wade. But to focus upon the intense debates surrounding precedent in constitutional law would be to miss the implications of precedent in every sort of case that a court encounters—common law, administrative regulation, statutes, or constitutional interpretation. One might question, for example, whether the Supreme Court carried its burden of justification in the series of cases following National League of Cities v. Usery.165 In only a nine-year period, precedent was created, exceptions carved, and finally the original holding was overruled in Garcia v. San Antonio Metropolitan Transit Authority.166 By contrast, more than one-half century passed before Plessy v. Ferguson was overruled.167 And in other circumstances, the Supreme Court has also said it will affirm a line of precedent even when it considers that precedent to be wrong.168 We have been con[*PG120]cerned with this issue of stare decisis versus non-binding precedent for centuries. The new element, and a radical departure from our prior practice, has been the relatively recent court rules declaring certain court decisions not to be precedent at all.
While I have largely discussed the doctrine of precedent in the context of common-law adjudication, Anastasoff was a case of statutory interpretation. The doctrine of precedent is not restricted to areas of judge-made common law but is also a feature in cases involving statutory interpretation. Precedent is one of the primary ways in which common-law systems are distinct from civil-law systems.169 We know that federal courts are said not to have a general common law, but that does not mean they are not common-law courts in their method. These core elements of starting from precedent have been the same whether the question is common law, or statutory or constitutional interpretation. These core elements, I have argued, have been more or less constant throughout our history and provide an independent rationale for affirming the result in Anastasoff.
This discussion about core understandings of judicial power begs a very important question which, by necessity, I can only touch upon here. What about state courts? Is it unseemly to say that precedent is an essential component of judicial power only for the federal courts, and not also for state courts? Although Anastasoff is limited to federal courts, some state constitutions are worded similarly to Article III in that they extend judicial power to the courts of their state.170 States have developed constitutional norms of judicial power, and judicial power issues often arise in separation-of-powers concepts or in standing or advisory opinion contexts. State reception provisions may lend support here.171 One could argue that when states “received” the common law, they received its methods as well, so that statutes or constitutional provisions directing state courts to apply the common law of England were also directing them to apply its doctrine of prece[*PG121]dent. The fundamental assumption of the non-citation rules in both state and federal courts is that judges have the authority to determine which of their decisions will be precedent for future cases and which will not. Anastasoff flatly disagrees with this view.
There may be no compelling distinction between the general point in Anastasoff and the practices of state courts. Whether state courts might abrogate non-citation rules on matters of prudence and policy, or on state constitutional grounds akin to Anastasoff, the issue cannot be ignored. In California, in fact, the state legislature in its 2000 session considered a bill that would have required that all state appeal opinions be valid as precedents.172 Also in California, a lawsuit against the State’s Supreme Court challenged the court’s practices of non-published opinions.173 And finally, an American Bar Association resolution urged all appeals courts to provide “at a minimum, reasoned explanations for their decisions.”174
The inquiry outlined here suggests that not only has the perceived need for a theory of precedent changed over time, so have the justifications for departing from precedent and overturning prior rules. The experience of jurists only a few generations removed from the Framers raises questions that are surely pertinent to how we now view the Framers’ understanding of judicial power. I have suggested that these experiences provide another way to affirm the Anastasoff court’s view of Article III. In the end, Anastasoff’s greatest contribution is to link the debate over non-citation rules with the debate about the nature of precedent. The opinion teaches that precedent matters to one’s theory of the law, of the judicial function, and of the compelling force of history.