35 Ind. L. Rev. 399 Indiana Law Review 2002 Symposium Caught in the Middle: The Role of State Intermediate Appellate Courts *399 THE PRACTICE OF PRECEDENT: ANASTASOFF, NONCITATION RULES, AND THE MEANING OF PRECEDENT IN AN INTERPRETIVE COMMUNITY Lauren Robel [FNa1] Copyright (c) 2002 Indiana Law Review; Lauren Robel Like sailing, gardening, politics, and poetry, law and ethnography are crafts of place: they work by the light of local knowledge. The instant case, Palsgraff or the Charles River Bridge, provides for law not only the ground from which reflection departs but also the object towards which it tends, and for ethnography, the settled practice, potlatch, or couvade, does the same. Whatever else anthropology and jurisprudence may have in common-vagrant erudition and a fantastical air-they have in common the artisan task of seeing broad principles in parochial form. "Wisdom," as an African proverb has it, "comes out of an ant heap." [FN1] Introduction What does it mean to say that a decision by a court is, or is not, precedential? At the most straightforward level, to describe a decision as "precedent" is to say that the decision must be acknowledged, at the least, by the court that issues it; a decision is precedential if courts are bound to follow it or distinguish it, given certain conditions. [FN2] When courts adopt rules, as have most appellate courts, which dictate that certain decisions are not precedent, they adopt legal rules about which we can either agree or disagree on the usual grounds-fairness, [FN3] utility, [FN4] accountability, [FN5] predictability, [FN6] and the like. *400 But, as the quotation from Geertz suggests, rules exist within bodies of practice, discrete cultures, and interpretive communities [FN7] that give them meaning. In a deeply common-law system such as ours, precedent is also a practice that places requirements on both lawyers and judges. Moreover, the practice of precedent does not rely on the above-stated rule of judicial acknowledgment because, at least linguistically, we refer to all decisions that might have persuasive force as "precedent," despite the lack of a legal rule requiring either adherence to them or an attempt to distinguish them. Courts also routinely treat nonbinding authority as precedent, in the sense that they feel an obligation to attempt to distinguish that authority. Although it is common for judges to write that they are bound by precedents with which they disagree, it is rare for a court to dismiss the relevance of a cited case for the sole reason that it is not required, by the narrow legal rule of stare decisis, to follow that case's holding. Our cultural conception of precedent, then, includes more than a sense that opinions have predictive value. It also includes shared understandings of the judicial role, which include burdens of justification. Thus, "precedent" can have multiple meanings, both linguistically and practically. One important interpretive community for judicial opinions, lawyers and judges, has recently renewed the debate about the meaning of precedent, as the result of a short-lived federal appellate decision holding unconstitutional the practice of condemning certain of a court's opinions to nonprecedential status. In this essay, I examine the concrete practices that surround appellate courts' publication rules and the common cultural commitments, as expressed through those practices, that lawyers and judges share about the meaning of judicial *401 decisions. Evidence from the federal appellate system shows that large numbers of participants in that system claim to behave in ways that would make sense only if unpublished (and therefore nonprecedential) opinions [FN8] are, in some sense that I will explore, precedential. Those participants include not only lawyers who practice in the federal appellate courts but also the judges themselves and the publishers who provide for the opinions' pervasive dissemination. [FN9] This evidence suggests that lawyers and judges value these opinions despite the rules limiting citation. This valuation, in turn, suggests a cultural, rather than a rule-bound, conception of stare decisis. I then examine the Eighth Circuit's opinion in Anastasoff v. United States, [FN10] which has served as a focal point for a renewed discussion of citation bans both inside and outside of the judiciary. Anastasoff raises fundamental questions about the meaning of precedent and the judicial role, and has been the wellspring of recent judicial criticism of citation rules in the federal appellate courts. Anastasoff and its progeny confirm that our cultural commitment to precedent includes a normative commitment to justification. Finally, I briefly examine the argument against abandoning the courts' current publication and citation rules-chiefly the argument that such rules are required by caseload-and find that it is less persuasive than commonly thought. More importantly, it is less compelling than the damage to the courts' perceived legitimacy that results from the continuation of the publication and citation rules in their current form. For many participants in the federal appellate system, uncitable opinions are part of the daily diet of cases that they examine and analyze in practice. The ban on their citation strikes at the metaphorical heart of the common-law system. *402 I. The Universe of Precedent and the World of Unpublished Opinions In the federal courts of appeals, widespread citation and publication restrictions date from 1976, when the Commission on Revision of the Federal Court Appellate System recommended that such restrictions be adopted to deal with caseload volume and the "proliferation of precedent." [FN11] In state intermediate appellate courts, the use of citation and publication restrictions is yet more recent. [FN12] Thus, federal appellate courts have about twenty-five years of experience with publication and citation rules, and all have adopted limited publication plans and limitations on citation. [FN13] Those courts now publish decisions in the official reporters in only thirty-four percent of all cases in which appellants had counsel and in only twenty-three percent of cases overall. [FN14] In addition, the rates of publication, as shown in Table 1, [FN15] vary widely from a low of seventeen percent for the Eleventh Circuit, to a high of seventy-one percent for the Seventh Circuit. [FN16] The rationale for publication and citation rules has shifted as a result of experience with those rules. When the rules were first suggested to the Judicial Conference of the United States Courts in 1964, the rationale was the prosaic one of dealing with "the ever increasing practical difficulty and economic cost of establishing and maintaining accessible private and public law library facilities." [FN17] The 1975 Commission on Revision of the Federal Court Appellate *403 System, commonly known as the Hruska Commission, approached its recommendations for limited publication from the perspective of caseload crunch. While repeating the idea that there were library cost savings to be had from limited publication, [FN18] that Commission suggested that judges could accrue significant time savings with limited publication, "for the judges no longer sense the same need to polish the prose and to monitor each phrase as they do with opinions which are intended for general distribution." [FN19] In its final report, the Commission linked its recommendation for adopting limited publication plans to the time opinions take to write and the possibilities for reduced appellate case processing time. [FN20] Recognizing that appellate courts serve both lawmaking and dispute-resolving functions, the plans that courts adopted all followed similar criteria, attempting to distinguish in advance between opinions that make law and those that merely apply it. [FN21] While there were variations (for example, some courts counseled publication when a case involved an issue of continuing public interest), the central rationale of all the plans was that non-lawmaking opinions need not be published. [FN22] Despite the similarities in criteria, however, the circuits quickly diverged in the percentage of opinions each published and made citable. By the time the Federal Courts Study Committee completed its work in 1990, and despite the huge increase in the federal appellate caseload that in part prompted that Committee's work, [FN23] fifty-eight percent of federal appellate judges believed that they almost always produced published opinions in the cases in which they "should be written." [FN24] Another thirty percent believed that they only "sometimes" had to forgo writing opinions for publication in cases in which they "should be written," presumably because of caseload pressures. [FN25] Given the significant variations in publication rates across circuits both then and now, these figures suggest divergent local norms with respect to what cases meet the criteria *404 for publication of the various appellate courts' plans. [FN26] This divergence exists despite the striking similarities among the plans themselves and the fact that each contained criteria for predicting which cases were likely to have sufficient lasting value as precedent to justify the additional thought and effort required for published decisions. [FN27] Although the publication rates reported in Table 1 suggest some connection between workload and publication rates (for example, the Eleventh Circuit, which has more merits terminations per judge than any other, also has the lowest publication rate), that connection is neither tight nor obvious. For instance, the Fourth Circuit's publication rate of nineteen percent is very low given its also low workload rank, and the Eighth Circuit's publication rate is relatively high despite its high workload rank. Publication limits are coupled in most circuits with citation bans, which vary in severity. [FN28] Citation bans have been controversial from the beginning. The Hruska Commission believed that such bans were necessary to discourage commercial publication (which it believed would undermine cost savings associated with writing less polished opinions) [FN29] and more fundamentally, to deal with the problem of unequal access to these opinions. [FN30] However, it noted that there were members of the bar and bench who "consider[ed] it undesirable and indeed improper for a court to deny a litigant the right to refer to action previously taken by the court." [FN31] Nevertheless, citation bans were adopted almost exclusively because of concerns about unequal access to unpublished opinions. The Hruska Commission feared "that the publication plans would result in a secret body of applicable and pertinent law available only to certain advantaged litigants and the courts before which they routinely appeared." [FN32] II. The Practice of Precedent: What Lawyers and Judges Do Publication plans were intended to make certain opinions disposable by identifying those cases that would add no new information to the canon of substantive law, while limiting citation to those opinions to ensure that lawyers would not spend time reading them and considering their impact or relationship *405 to new cases. But, recent surveys of both litigants and judges in the federal appellate system demonstrate that these rules fail to dissuade substantial numbers of attorneys and judges from these activities. [FN33] The surveys, developed by the Commission on Structural Alternatives for the Federal Courts (also known as the White Commission, in honor of its Chair, Justice White), were sent to all federal appellate and trial judges, and to a random group of attorneys who had cases before the courts of appeals within a specified time frame. [FN34] The surveys yield important information about the behavior of both the consumers and producers of opinions. A. Lawyers The surveys indicate that attorneys do not share the view that there are too many precedential opinions available. When asked to rank a number of possible priorities for courts of appeals, attorneys rank avoiding the "proliferation of published opinions" next to last. [FN35] Further, when asked the most frequent reason for their inability to predict appellate outcomes, as reflected in Table 2, [FN36] fair percentages of lawyers in many circuits said that their difficulty was due to either the lack of circuit precedent on point or a lack of clarity in existing circuit precedent. More important, in agreement with earlier evidence relating to government attorneys, [FN37] the responses indicate that many attorneys monitor unpublished opinions. Though the reasoning behind the citation and publication rules would predict that a significant number of lawyers would regularly read all of the published opinions in their circuits, at least in their fields (Table 2, Column 3), [FN38] they would hardly predict that twenty to twenty-five percent of attorneys in five circuits would do the same with unpublished opinions (Table 2, Column 4). [FN39] Moreover, if the unpublished opinions come up in research in preparation for an actual case, the numbers of lawyers who report that they read the opinions jumps *406 significantly (Table 2, Column 5). [FN40] In no circuit do fewer than twenty-eight percent of responding attorneys read these cases, and in seven circuits over forty percent read them. Indeed, Table 2, Columns 4 and 5, [FN41] when taken together, suggest that significant percentages of lawyers do not feel free to ignore these opinions either generally or with respect to specific cases. If we assume that lawyers are not reading these opinions as a leisure activity, then the most plausible alternative is that they are reading them because they provide useful information in support of their clients' cases. Two lawyers from an eminent law firm make the point succinctly: It is true that citation prohibitions allow lawyers and courts to ignore unpublished opinions with the confidence that they have not overlooked binding precedent. There are two reasons, however, why this concern is minimal. First, for many issues, there are few on-point but uncitable appellate decisions, so the time it takes to review these decisions is short. For the most part, the myth that there exist great batches of redundant unpublished appellate cases is true only in certain discrete areas of law where meritless cases are litigated even to appeal . . . . Second, in practice, citation prohibitions hardly ease the case-review burden on the prudent practitioner. Practitioners often review uncitable cases to mine them for new ideas. A prudent lawyer also reviews unpublished cases, lack of precedential value aside, because they indicate how the appellate court has ruled in the past and thus might rule in the future. Moreover, it behooves counsel to review unpublished opinions because they still may influence a court that reads (or remembers deciding) them itself. [FN42] For these lawyers and others like them, the rule-based conception of precedent embodied in "citation prohibitions" is almost beside the point because their practice is based on a cultural conception of precedent that views all decisions as evidence of reasoning that might be persuasive in the future. True, the citation rules might save the lawyer who ignores unpublished opinions from charges of malpractice (giving them "confidence" that they can be ignored), but it simply is not consistent with "prudent" practice in a common-law environment to ignore these opinions. Moreover, other evidence strongly suggests that, even under a rule-based conception of precedent, lawyers are correct to read these opinions. A recent study found that unpublished opinions are routinely (indeed, promiscuously) cited by the federal courts of appeals and relied upon by the federal district courts. [FN43] The study found that the courts of appeals have cited unpublished *407 dispositions (usually their own) 4460 times, and that 3161 federal district court opinions cite unpublished federal circuit court opinions, relying on them for legal support in 1967 cases. [FN44] In this light, it is rational for lawyers to stay apprised of unpublished dispositions, for they often explicitly influence outcomes. It is impossible to determine how much more often they influence outcomes more subtly. The survey responses of appellate and district court judges confirm that lawyers are not the only ones taking notice of "noncitable" opinions on a regular basis. B. Judges As Table 3, Column 2 [FN45] shows, many district judges regularly read all of the unpublished opinions from their circuit courts. When significant numbers of district judges report regularly reading these opinions, it suggests that they are doing so to predict how their court of appeals would decide an issue. In some instances, district judges may be forced to monitor these opinions because their circuits provide so little published caselaw. In the Fourth Circuit, for instance, where the published corpus represents only nineteen percent of the merits terminations, almost sixty percent of the district judges monitor all or most of the unpublished appellate opinions. In circuits with higher publication rates, like the Eighth Circuit, the relatively high number of district judges reporting that they regularly read the unpublished opinions might suggest problems with the application of the rules. My working hypothesis is that these district judges are not behaving irrationally, so if they are reading the opinions regularly, they must believe that the decisions either predict outcomes or provide direction. In five of the circuits, twenty percent or more of the district judges regularly read unpublished opinions. The survey also asked the district court judges whether there were issues or areas of circuit law that were particularly inconsistent or difficult to know. As shown in Table 3, Column 3, [FN46] affirmative responses to this question varied by circuit, from a low of thirteen percent in the First Circuit to a high of fifty-seven percent in the D.C. Circuit. The judges indicated that the difficulty they faced was largely due to inconsistencies among published opinions, but in five circuits, twenty-four percent to over forty percent of the judges attributed difficulty to inconsistencies between published and unpublished opinions. These numbers *408 suggest that the judges are not only monitoring the opinions, but also are analyzing them in light of existing published opinions in an attempt to predict the development of doctrine within their circuits. Additionally, district judges are as skeptical as lawyers about their ability to find a precedent when they need one. As shown in Column 4, [FN47] in eight circuits, twenty percent or more of the judges attributed the inability to know an area of law in their circuit to the lack of precedent. Appellate judges are also consumers of unpublished appellate opinions. There are three reasons why appellate judges might devote time to reading most "nonprecedential" opinions. First, they may feel obligated to monitor the application of their circuit's publication rule. Second, they may feel a duty to stay abreast of the overall work of their courts. Finally, they may believe that there is important information contained in the opinions that is not available elsewhere. It seems most plausible that appellate judges read unpublished opinions to monitor compliance with publication rules. The publication rates, as noted earlier, suggest that, among the courts of appeals, there are different cultures with respect to publication. As shown in Table 3, columns 1 and 2, [FN48] with one notable exception, the courts with the strongest commitments to publication (the First, Seventh, Eighth, and D.C. Circuits) also have the highest number of judges reporting that they read unpublished opinions. [FN49] These results suggest that those courts continue to be less comfortable with nonpublication and monitor unpublished opinions more closely to ensure compliance with the rules (these figures do not, however, demonstrate that they are successful in achieving compliance; we need to look to the lawyer-consumers of the opinions for that information). Proponents of citation and publication plans hoped, apparently in vain, that the plans would work so well that the unpublished opinions' inherent worthlessness would discourage their use. [FN50] Appellate courts have tried several unsuccessful strategies to discourage the use of unpublished opinions. In the early days of the plans, they attempted to withhold these opinions from commercial publishers, reasoning that lawyers could not cite what they could not find. [FN51] Commercial publishers have, however, responded to demand for the opinions and now, with few exceptions, unpublished opinions are generally available in the searchable electronic databases. [FN52] The opinions are also available *409 on court websites and in specialty reporters. [FN53] Courts have also tried simply not writing opinions at all. [FN54] However, this option proved unattractive to litigants and judges alike, [FN55] because to strip opinions of all rationale is to leave them vulnerable to claims of lack of accountability. And since nonpublication is so routinely coupled with lack of oral argument, neither litigants nor lawyers can discern whether the court either understood or acknowledged their arguments. [FN56] As I will discuss below, the failure to write an opinion at all is inconsistent with deeply-held conceptions of the judicial role. III. Fractures in the Interpretive Community: Anastasoff and the Renewed Debate The evidence from every quarter is that substantial numbers of those involved with the federal appellate courts-judges, litigants, and publishers alike-do not sharply distinguish in their practices between published, citable opinions, and unpublished, noncitable opinions. The one glaring exception is that the latter cannot be cited by lawyers, although they are apparently often cited by courts. Also, because the judges are reading unpublished opinions, they are also presumably making the judgments that typically attend analysis of opinions: that this case is like or unlike another, well-reasoned or not, persuasive or not. When trial judges, lawyers, and litigants find that their best precedent is one that they are not supposed to cite, they face a clash between the cultural conception of precedent evident in their behavior, and the rule-based conception of precedent embodied in citation bans. In point of fact, the view of precedent embodied in the publication rules is itself both cultural and rule based. These rules typically couple a command (i.e., citation is banned or limited) with a set of goals for nonpublication that echo a practice-based view of precedent. Characteristic of these rules, the First Circuit couples a citation ban with a set of guidelines that attempts to identify those opinions that would "serve . . . as a significant guide to future litigants," because they articulate a new rule of law or modify an established rule. [FN57] The Second Circuit describes the goal as the attempt to predict "those cases in which . . . each *410 judge of the panel believes that no jurisprudential purpose would be served [by publication]." [FN58] The Third Circuit looks for those opinions that have "precedential or institutional value." [FN59] The courts look, in other words, to the future value of the opinions. However, unless the courts publish almost nothing, the future value of what they write is not, culturally speaking, determined by the authoring judge alone; rather, it is determined by consumers. Hence the citation rules, which attempt to strip the opinions of value by fiat. The rules ask judges through language that echoes the common-law understanding of the meaning of precedent to predict the future value of their opinions to consumers. That common-law understanding itself depends upon consumers' perception of future value. The rules, therefore, embody the clash between cultural, practice-based views and rule-based views of precedent. It is trivially easy to find examples of cases that, according to the criteria of the publication plans, should have been published, because in hindsight (or perhaps even with foresight) the opinions provide information that consumers would find useful in predicting future outcomes. [FN60] One result of the clash between cultural views of precedent and rule-based views is that intermediate appellate courts' publication practices have been the target of academic criticism for years. [FN61] But the past year must have set a record for such criticism from outside of the academy. The California Assembly's Judiciary Committee, for instance, unanimously approved a bill that would have required that "[a]ll final opinions of the Supreme Court, of the courts of appeal, and of the appellate divisions of the superior courts" be "made available for private publication, in full," and mandated that those opinions "shall constitute precedent under the doctrine of stare decisis the same as opinions published in the official reports." [FN62] The proposed bill, which would have changed the status of over ninety percent of California's appellate decisions, did not pass the California Assembly, but it did provoke commentary from almost every legal organization in the state. Indeed, an entire website exists to challenge California's (and every other state's) nonpublication and no-citation rules. [FN63] Meanwhile, the United States Court of Appeals for the Eighth Circuit held unconstitutional that circuit's rule condemning unpublished opinions to *411 nonprecedential status. [FN64] Though that decision was later vacated, it generated enormous discussion, [FN65] among both commentators and journalists, [FN66] and in other courts. [FN67] The Anastasoff decision ratchets up the clash between practice and rule, adding another piece to the description of the practice of precedent. It also provides additional insight into why the rather obscure rules on citation have garnered political and popular attention. [FN68] A. Anastasoff: Precedent Practice as Normative Commitment Faye Anastasoff sought a $6000 tax refund, but the government contended that her request for that refund had arrived one day late. Anastasoff argued that her request was timely because it had been mailed before the expiration of the refund period; the IRS disagreed, citing an earlier Eighth Circuit case that had held the so-called mailbox rule inapplicable. [FN69] Anastasoff argued that the cited case was unpublished (despite the fact that it had decided an issue of first impression under federal tax law) and, therefore, nonprecedential under the Eighth Circuit's rule. [FN70] The Eighth Circuit disagreed, and in an opinion by Judge Richard Arnold, held its own rule on citation unconstitutional "because it purports to confer on the federal courts a power that goes beyond the 'judicial"' *412 within the meaning of Article III of the Constitution. [FN71] Judge Arnold's constitutional analysis was both historical and structural. In his view, the founding generation saw the obligation to follow precedent as definitional of judicial power and at the core of what distinguished the judiciary from the political branches. [FN72] Moreover, Judge Arnold argued that the obligation to follow precedent served a separation of powers function, by "limit[ing] the judicial power delegated to the courts by Article III," [FN73] presumably by assuring that judges were not free to behave arbitrarily, but were required to justify their actions in the present by resorting to what they had done in the past-a classically conservative approach to policy. Thus, "[t]he duty of courts to follow their prior decisions was understood to derive from the nature of the judicial power itself and to separate it from a dangerous union with the legislative power." [FN74] A departure from the doctrine of precedent, noted Judge Arnold, quoting Justice Joseph Story, "would have been justly deemed [by the Framers] an approach to tyranny and arbitrary power, to the exercise of mere discretion, and to the abandonment of all the just checks upon judicial authority." [FN75] Finally, Judge Arnold argued, the rule against treating decisions as precedential violates a principle of equal treatment. The courts, he said, should reject a doctrine that, in essence, states, "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." [FN76] The Anastasoff decision carefully avoids two potential misconceptions about its scope. First, Judge Arnold does not advance a view of precedent that requires "eternal adherence" to previous decisions. Rather, he argues, treating all decided cases as precedential puts an appropriate burden of justification on the judiciary. [FN77] Second, treating all decisions as precedent does not on depend on publication. The Founder's understanding of precedent depended not on publication, a relatively recent practice, but on the existence of a decision. Historically, precedent could be established "only by memory or by a lawyer's unpublished memorandum." [FN78] Indeed, argued Judge Arnold, "entry on the official court record [was] sufficient to give a decision precedential authority whether or not the decision was subsequently reported." [FN79] *413 During Anastasoff's brief tenure as a precedent (and even into its ironic demise), federal judges cited it to question the citation bans in their courts and in connection with their own concerns about the application of the publication rules. The Fifth Circuit, for instance, had in an unpublished opinion held that the Dallas Area Rapid Transit (DART) was a state governmental entity entitled to sovereign immunity under the Eleventh Amendment. [FN80] A year and a half later, a subsequent panel of the Fifth Circuit (which included one of the judges from the earlier unpublished case) determined that DART was not a state entity and, therefore, not entitled to immunity. The difference, of course, was that the latter decision was released for publication. [FN81] In dissenting from a petition for rehearing en banc, three judges noted that the continuing "justification for refusing to confer precedential status on [unpublished] opinions is . . . tenuous," citing the wide availability of such opinions online. [FN82] In the Ninth Circuit, a dissenting judge accused his court of creating a circuit split on the question of the interpretation of a sentencing guidelines, citing both published and unpublished cases. [FN83] The unpublished cases from the Fourth and Seventh Circuits apparently represented the only authority in those circuits interpreting the sentencing guideline as it applied to a particular crime. [FN84] In the Fifth Circuit, a district court in Texas published a plea to the court of appeals to reconsider its citation ban, noting that the holding and reasoning of a recent unpublished decision of that court would have decided the case before it (and relying on that decision nonetheless). [FN85] What makes these decisions troubling? Anastasoff's central insight-albeit an implicit one-is that precedent is not a concept that can successfully be constrained by rule in a deeply, and historically, common-law legal culture such as ours. Judge Arnold approaches this insight from the perspective of producers of opinions, arguing a constitutional duty for a court to acknowledge the authority-in-fact of its own work product. The power of this critique comes from its implicit premise that to do otherwise would be to engage in the arbitrary exercise of authority, because "opinions" are not simply cultural artifacts, but also the actual decisions of the judiciary with respect to the litigants' lives, liberty, and property. Imagine a worst-case, if fanciful, scenario: two appeals involving the validity of the imposition of the death penalty, on the same *414 grounds, one resulting in reversal-unpublished and therefore uncitable-and the second resulting in affirmance. No one-no judge or litigant-would find this an acceptable proposition. But no-citation policies produce equivalent results, although not so morally outrageous, with regularity. Indeed, the no-citation policy denounced in Anastasoff would have, if applied in that case, potentially left Christie, the taxpayer in the unpublished case cited by the government, out-of-pocket an unspecified amount at the same time Anastasoff collected her $6000, on diametrically opposed-but not disavowed-interpretations of the same statute by the same court. Whether one agrees with Judge Arnold's constitutional analysis, his opinion describes the process of common-law decisionmaking in deeply cultural terms. To note the depth of our historical cultural commitment to justification may not be compelling constitutional analysis, but it is fine cultural observation. Common-law understandings of precedent entail more than a practical view that all opinions contain information with predictive value. What makes those opinions predictive is the requirement of justification for deviance from them. The normative commitment to justification is as firmly entrenched in our understandings of precedent as is our belief in the predictive value of case results. The publication and citation plans are controversial because they violate every piece of the cultural view. [FN86] B. The Contemporary Irrelevance of Publication and Citation Plans The empirical assumptions underlying publication plans-the assumption that decisions can be sorted into precedential and nonprecedential stacks before they are written, or that such distinctions are even possible-have been widely discredited. [FN87] The rationale behind the citation bans-fear of unequal access-has almost evaporated in the electronic age, which makes these opinions both accessible and searchable with the laser-like capabilities of modern legal databases. [FN88] Large numbers of participants in the federal appellate system, including judges, use unpublished opinions in ways not contemplated by the publication plans, although completely consistent with common-law understandings of practice surrounding precedent. The sheer numbers of these opinions coupled with the familiarity of judges and lawyers with them have fueled renewed concerns about the legitimacy of both citation bans and courts' *415 claims to exemption from the norm of justifying departures from their earlier decisions. In the face of Anastasoff and similar criticisms leveled by others, several judges and commentators have attempted to rehabilitate publication plans, either by reworking the meaning of precedent, [FN89] or by providing alternative rationales for limitations on precedential effect more closely tied to the special roles of intermediate appellate courts. [FN90] While intriguing and thoughtful, these attempts to reconfigure a set of court docket practices to other purposes are attempts to rehabilitate a system that should be abandoned. [FN91] The only remaining argument for citation and publication rules-the argument from caseload-is ultimately unpersuasive. It is unpersuasive not because the caseload claims of the intermediate appellate courts are overstated; they are not. [FN92] Though we might romanticize an earlier era, when appellate courts applied more extensive processes to case decision, increased caseload without increased decisional capacity now makes that vision unrealistic. However, saying that courts can do none other than what they do is not the same as saying that the rules governing what they do make contemporary sense. To see why the argument from caseload is unpersuasive, take it seriously. Imagine that courts continue deciding cases in exactly the same way they are deciding them now, giving to each case exactly the attention it now gets, and writing exactly what they now write, no more and no less. Next imagine that the only change is to the rules that govern what lawyers can do with those opinions. [FN93] What would be lost in abandoning limitations on citation? Judges make three arguments. First, judges argue that "[t]here is value in keeping [the] body of law cohesive and understandable, and not muddying the water with a needless torrent of published [and therefore citable] opinions." [FN94] *416 But the value of cohesive and understandable law is that it provides predictability. If it does so only because cases that would "muddy" that cohesiveness are exempt from the norm of judicial justification, it is a false cohesiveness, achieved only by ignoring decisions that create the mud. It therefore provides few of the benefits of predictability, for (as the lawyers quoted earlier already suspect), the picture on the surface is partial. Second, judges argue that the cases that receive less attention really have little predictive value; indeed, they have negative value. Judges Kozinski and Reinhardt of the Ninth Circuit ask, "[W]hat does precedent mean? Surely it suggests that the three judges on a panel subscribe not merely to the result but also to the phrasing of the disposition." [FN95] Because caseload pressures make it impossible to craft the phrasing of every opinion, and because in their view it is the language and not the result of an opinion that has predictive value, lawyers should stop "[t]rying to extract from [unpublished opinions] a precedential value that we didn't put into them." [FN96] The assertion that authors, not readers, control the meaning of their writing is theoretically contestable, [FN97] and given the history of the publication rules, highly debatable empirically. If the opinions have negative information, in the sense these judges suggest, then busy lawyers trained in common-law methods will not spend time on them. But in the Ninth Circuit, nearly half of the lawyer respondents read these opinions when they come up in their research. This battle has been lost. Finally, there is the burden on judges of having to read and respond to citations from enlarged sources, and I do not want to discount it. [FN98] However, there is a serious counterweight. The wide variation in publication rates among the circuits, coupled with the local variations in the behavior of judges and *417 lawyers shown through the surveys, demonstrates that we are not living in the world imagined by the publication plans. When sixty percent of the district judges in a circuit that published nineteen percent of its decisions feel required to read those decisions regularly, we can no longer talk as if noncitable opinions are a trivial issue. The serious issue is the legitimacy issue that was identified in Anastasoff. The growing circle of criticism-beyond the academy and into politics-demonstrates that it is time for the intermediate appellate courts to face up to the plans' infirmities. Conclusion Law in this country is, in important respects-perhaps in the most important respects-an activity, a practice. Both unburdened and unaided by the tools that might mark it as a discipline-a distinctive set of methodologies or an overarching theoretical paradigm-law gets by nonetheless by creatively scavenging true disciplines and adhering to practices, such as ethical and logical norms that, while not distinctive to law, have served the needs of society adequately, and sometimes spectacularly. At the metaphorical heart of legal practice, in an historically common-law system like that of the United States, is a commitment to the idea of precedent. That commitment has both advantages and drawbacks, [FN99] and in many ways it seems quaint: law is practiced in many places besides courts that have no commitment to the hierarchical and analogical reasoning that play such a part in precedent's role in litigation. But its practical centrality to legal practice in American courts is hardly controversial. The publication and citation plans widely adopted twenty-five years ago strike at this metaphorical heart. They say to American lawyers that vast numbers of decisions from the appellate courts have less precedential value than, say, a decision from France, which can be freely cited for whatever persuasive value it might have. The plans are not accepted in practice by either judges or lawyers. They should be abandoned. *418 Table 1. Workload and Publication Rates by Circuit, FY 1997 TABULAR OR GRAPHIC MATERIAL SET FORTH AT THIS POINT IS NOT DISPLAYABLE *419 Table 2. Appellate Attorney Survey: Selected Responses TABULAR OR GRAPHIC MATERIAL SET FORTH AT THIS POINT IS NOT DISPLAYABLE *421 Table 3. Appellate and District Court Surveys: Selected Responses TABULAR OR GRAPHIC MATERIAL SET FORTH AT THIS POINT IS NOT DISPLAYABLE [FNa1]. Associate Dean and Val Nolan Professor of Law, Indiana University School of Law-Bloomington. My thanks to the participants in the symposium, Caught in the Middle: A National Symposium on the Role of State Intermediate Appellate Courts, March 30-31, 2001, hosted by the Indiana University School of Law-Indianapolis, and the participants in the symposium, Managing the Federal Courts: Will the Ninth Circuit be a Model for Change?, March 24, 2000, hosted by the University of California, Davis, School of Law, for helpful comments on various versions of this paper. [FN1]. Clifford Geertz, Local Knowledge: Further Essays in Interpretive Anthropology 167 (1983). [FN2]. For what remains the most thorough explication and critique of the argument from precedent, see Frederick Schauer, Precedent, 39 Stan. L. Rev. 571 (1987). [FN3]. See, e.g., Lauren K. Robel, The Myth of the Disposable Opinion: Unpublished Opinions and Government Litigants in the United States Courts of Appeals, 87 Mich. L. Rev. 940 (1989) (discussing differential access to unpublished opinions between government litigants and others). For an argument that publication rules are not the "rules," because there are no sanctions for violating them and no true policing mechanisms, see Mitu Gulati & C.M.A. McCauliff, On Not Making Law, 61 Law & Contemp. Probs. 157, 164 (1998). [FN4]. See, e.g., Douglas A. Berman & Jeffrey O. Cooper, In Defense of Less Precedential Opinions: A Reply to Chief Judge Martin, 60 Ohio St. L.J. 2025, 2040 (1999) (arguing that nonprecedential decisions could aid in law development); Alex Kozinski & Stephen Reinhardt, Please Don't Cite This!, Cal. Law., June 2000, at 44 (arguing that the primary purpose of noncitation rules is to permit judges to deal expeditiously with caseloads). That caseloads require nonpublication is asserted even by judges who oppose the noncitation rules. See, e.g., Richard S. Arnold, Unpublished Opinions: A Comment, 1 J. App. Prac. & Process 219, 221 (1999). However, the publication practices of the various appellate courts suggest that the rate of publication is dictated more by local norms than by caseload. See discussion infra Part I; see also Danny J. Boggs & Brian P. Brooks, Unpublished Opinions & the Nature of Precedent, 4 Green Bag 2d 17, 19 (2000) (arguing against justifying "on high theory a practice that is in fact justified for simple efficiency reasons"). [FN5]. See, e.g., Thomas E. Baker, Rationing Justice on Appeal: The Problems of the U.S. Courts of Appeals 128-29 (1994); William M. Richman & William L. Reynolds, Elitism, Expediency, and the New Certiorari: Requiem for the Learned Hand Tradition, 81 Cornell L. Rev. 273, 282-83 (1996). [FN6]. See, e.g., Deborah Jones Merritt & James J. Brudney, Stalking Secret Law: What Predicts Publication in the United States Courts of Appeals, 54 Vand. L. Rev. 71, 113-15 (2001) (demonstrating substantial intercircuit variation in the application of rules on the publication of opinions). [FN7]. An interpretive community is a group of readers that share understandings because of common cultural commitments. See generally Stanley Fish, Doing What Comes Naturally: Change, Rhetoric, and the Practice of Theory in Literary and Legal Studies (1989); William S. Blatt, Interpretive Communities: The Missing Element in Statutory Interpretation, 95 Nw. U. L. Rev. 629 (2001) (applying the concept to legal interpretation). [FN8]. In this essay, I will refer to those opinions that are designated nonprecedential opinions by federal appellate court rule as "unpublished." Though I use the term "unpublished opinions," it is important to note that I am referring to those opinions that have been designated uncitable by rule and, therefore, not submitted for publication in an official reporter. [FN9]. These opinions are widely available in searchable databases such as Westlaw and Lexis, see Michael Hannon, Developments and Practice Notes: A Closer Look at Unpublished Opinions in the United States Courts of Appeals, 3 J. App. Prac. & Process. 199, 209 n.48 (2001). "Westlaw estimates there are about 336,000 unpublished federal appellate opinions in its case databases." Id. Others appear in commercial specialty reporters. Id. at 206. See also Boggs & Brooks, supra note 4, at 18 (2000) ("The 'unpublished opinions' debate . . . is badly misnamed. Between Lexis and Westlaw, Internet sites maintained by universities and some of the circuit courts of appeals, and networks of attorneys practicing in particular fields, it is the rare opinion that is not disseminated for mass consumption.") [FN10]. 223 F.3d 898 (8th Cir. 2000), vacated, 235 F.3d 1054 (8th Cir. 2000) (en banc). In its subsequent rehearing en banc, the Eighth Circuit stated, "The constitutionality of that portion of Rule 28A(i) which says that unpublished opinions have no precedential effect remains an open question in this Circuit." Anastasoff, 235 F.3d at 1056. [FN11]. Publication and citation rules in the federal appellate courts had their genesis earlier, in a 1964 Judicial Conference resolution permitting federal courts to limit publication, but most publication regimes date from the report of the so-called Hruska Commission. See Comm'n on the Revision of the Fed. Ct. App. Sys., Structure and Internal Procedures: Recommendations for Change-A Preliminary Report (1975) [hereinafter Comm'n Preliminary Report]; Robel, supra note 3, at 945; Hannon, supra note 9, at 207-08 (discussing timeline for federal court publication rules); id. at 253-85 (compiling state and federal court citation bans). [FN12]. "As of 1981, only sixteen states operated under a nonpublication regime." David M. Gunn, "Unpublished Opinions Shall Not Be Cited as Authority": The Emerging Contours of Texas Rule of Appellate Procedure 90(i), 24 St. Mary's L.J. 115, 124 (1992). In 1968, thirty states still published opinions in all of their cases, as opposed to eighteen in 1974 and only six by 1989. Thomas B. Marvell, State Appellate Court Responses to Caseload Growth, 72 Judicature 282, 288 (1989). [FN13]. See Hannon, supra note 9, at 253-57. Nine federal circuits ban citation of "unpublished" dispositions as precedent, except in related cases to establish a defense of res judicata or for similar reasons. Three circuits allow citation to these opinions as persuasive authority, and two, though disfavoring citation, permit it if no published opinion would serve as well. [FN14]. See Comm'n on Structural Alternatives for the Fed. Cts. of Appeals, Working Papers 112 (1998) [hereinafter Working Papers]. [FN15]. See tbl.1, infra. [FN16]. I will return to this table in the discussion of circuit publication cultures infra. [FN17]. Reports of the Proceedings of the Judicial Conference of the United States, Annual Report of the Director of the Administrative Office of the United States Courts 11 (1964) (reporting, in light of these concerns, a Judicial Conference resolution "[t]hat the judges of the courts of appeals and the district courts authorize the publication of only those opinions which are of general precedential value and that opinions authorized to be published be succinct."). [FN18]. Comm'n Preliminary Report, supra note 11, at 72 ("When large numbers of such opinions [with limited precedential value] find their way into the reports, they create logistical problems in terms of sheer space and library maintenance expenditures, and the burden of fruitless research is compounded."). [FN19]. Id. [FN20]. Comm'n on Revision of the Fed. Ct. App. Sys., Structure and Internal Procedures: Recommendations for Change-Final Report 50 (1975) [hereinafter Comm'n Final Report]. [FN21]. Robel, supra note 3, at 941. [FN22]. Id. [FN23]. The Committee itself described caseload as a "crisis." Report of the Federal Courts Study Committee 4 (1990). [FN24]. 2 Fed Cts. Study Comm., Working Papers and Subcommittee Reports 87 (1990). [FN25]. Id. [FN26]. Lauren K. Robel, Caseload and Judging: Judicial Adaptations to Caseload, 1990 BYU L. Rev. 3, 55 (noting a changing "collective sense of what constitutes 'appropriate' attention to a case . . ."). [FN27]. Id. at 50. [FN28]. Melissa M. Serfass & Jessie L. Cranford, Federal and State Court Rules Governing Publication and Citation of Opinions, 3 J. App. Prac. & Process 251, 253-85 (2001). [FN29]. Comm'n Final Report, supra note 20, at 51 ("To allow litigants to cite opinions which the court has designated as 'not for publication' invites publication by private publishers, thus defeating the basic purposes of the program."). [FN30]. Id. Despite citation bans, the Commission's concerns with unequal access were well-founded. See Robel, supra note 3 (reporting on a study of government litigants' use of unpublished dispositions). [FN31]. Id. [FN32]. Robel, supra note 3, at 946 (recounting testimony before Hruska Commission). [FN33]. I have previously argued that lawyers are not dissuaded from using unpublished opinions by rules against citation, because the nonpublication rules select cases for publication on the basis of mistaken assumptions about the nature of precedent. I identified three assumptions that I believed were mistaken: that only lawmaking and not dispute-resolving opinions give lawyers important information; that the only opinions that are important are those that create, rather than apply, rules; and that the plans would operate neutrally with respect to the subject-matter of opinions. See Robel, supra note 3, at 947. [FN34]. The Commission published both its Final Report and Working Papers in 1998. The survey results are reported in the Working Papers. The response rate for each of the surveys was high: eighty-six percent for appellate judges; eighty-one percent for district judges; and fifty-one percent for appellate counsel. Working Papers, supra note 14, at 3. [FN35]. Id. at 81-82. It was narrowly edged out of last place by attorneys' lack of interest in "mediation or pre-argument conferencing." Id. [FN36]. See tbl.2, infra. [FN37]. See Robel, supra note 3, at 957-59. [FN38]. See tbl.2, infra. [FN39]. See id. [FN40]. See id. [FN41]. See id. [FN42]. Salem M. Katsh & Alex V. Chachkes, The Constitutionality of "No Citation" Rules, 3 J. App. Prac. & Process, 287, 301-02 (2001). The authors, attorneys at Shearman & Sterling, argue that citation bans violate the First Amendment. [FN43]. See Hannon, supra note 9, at 235 tbl.6; see also Johns v. State, 35 P.3d 53, 65 (Alaska Ct. App. 2001) (Mannheimer, J., concurring) (noting that Alaska's lower court judges routinely rely on unpublished Alaska opinions). [FN44]. Hannon, supra note 9, at 235. Hannon believes the last figure to be a conservative estimate because he counted only those cases where the court "explicitly cited the unpublished case for legal authority." Id. For these courts, Anastasoff was something of a relief: several cite it for the proposition that their discussion of unpublished opinions is not improper. See, e.g., McGuinness v. Pepe, 150 F. Supp. 2d 227, 235 (D. Mass. 2001). [FN45]. See tbl.3, infra. [FN46]. See id. [FN47]. See id. [FN48]. See id. [FN49]. The one exception is the Fourth Circuit, which has almost the lowest rate of publication and where half of the appellate judges regularly read the unpublished opinions. [FN50]. See Robel, supra note 3, at 944. [FN51]. See id. at 944-45. [FN52]. See Hannon, supra note 9, at 210-13. The Third, Fifth, and Eleventh Circuits "have banned electronic dissemination of unpublished opinions, and these cases are neither added to Westlaw or LEXIS nor available from the courts' websites." Id. at 211. Hannon found, however, that a large number of unpublished opinions from these circuits are in fact available in Westlaw's CTA database. See id. In order to understand how "banning" the dissemination of documents of public record could be possible, one must comprehend the dynamics of the relationship between the courts and legal publishers. [FN53]. See id. at 206. [FN54]. Both the Fifth Circuit and the Third Circuit have tried and abandoned this approach. See Philip Shuchman & Alan Gelfand, The Use of Local Rule 21 in the Fifth Circuit: Can Judges Select Cases of "No Precedential Value"?, 29 Emory L.J. 195 (1980). [FN55]. See Robel, supra note 3, at 943 (discussing reasons why courts abandoned summary dispositions). [FN56]. See Working Papers, supra note 14, at 110 tbl.8 (noting that only one circuit publishes as many as ten percent of its decisions made without oral argument). [FN57]. 1st Cir. R. 36(b). [FN58]. 2d Cir. R. 0.23. [FN59]. 3d Cir. I.O.P. app. 5.2. [FN60]. This would be true even if the plans worked perfectly to exclude from publication only non-lawmaking opinions, because applications of rules are as important to practitioners as their creation. See Robel, supra note 3, at 941-42; see also Boggs & Brooks, supra note 4, at 19-20. [FN61]. See generally Thomas E. Baker, Federal Court Practice and Procedure: A Third Branch Bibliography, 30 Tex. Tech L. Rev. 909, 1092-96 (1999) (citing sources of such criticism). [FN62]. Richard H. Cooper & David R. Fine, What's Past is Prologue, 43 Orange County Law. 27 (2001) (discussing California Assembly Bill 2404). Cooper and Fine quote a California Legislative Counsel Report, AB 2404 at 4, which states that "93 percent of California Court of Appeal opinions are unpublished and uncitable." Id. [FN63]. See Committee for the Rule of Law, at http://www.nonpublication.com (last updated Nov. 3, 2001) (citing articles and collecting information about cases challenging citation bans). [FN64]. See Anastasoff v. United States, 223 F.3d 898 (8th Cir. 2000), vacated, 235 F.3d 1054 (8th Cir. 2000) (en banc). [FN65]. A search through the electronic databases for commentary on the Eighth Circuit's opinion reveals a very large amount of interest in this topic from all segments of the practicing bar, from periodicals targeted to corporate counsel to lawyers for the public interest. For compilations of new articles, see Committee for the Rule of Law, supra note 63. [FN66]. See, e.g., Better Not Cite Those Unpublished Opinions Just Yet, 16 Fed. Litigator 41 (2001); Jerome I. Braun, Eighth Circuit Decision Intensifies Debate over Publication and Citation of Appellate Opinions, 84 Judicature 90 (2000); David R. Fine, Keeping Mum Kills Precedents, Nat'l L.J., Feb. 19, 2001, at A21. [FN67]. See discussion infra. [FN68]. Anastasoff was not the first case to have considered the constitutionality of citation bans, although it is the first to have issued a direct ruling on a constitutional ground. See Lance A. Wade, Honda Meets Anastasoff: The Procedural Due Process Argument Against Rules Prohibiting Citation to Unpublished Judicial Decisions, 42 B.C. L. Rev. 695, 712-15 (2001) (detailing previous constitutional challenges and arguing that citation bans violate procedural due process). [FN69]. See Christie v. United States, No. 91-2375MN, 1992 U.S. App. LEXIS 38446, at *5-8 (8th Cir. Mar. 20, 1992) (per curiam). [FN70]. See 8th Cir. R. 28(a)(i). The rule provides that [u]npublished opinions are not precedent and parties generally should not cite them. When relevant to establishing the doctrines of res judicata, collateral estoppel, or the law of the case, however, the parties may cite an unpublished opinion. Parties may also cite an unpublished opinion of this court if the opinion has persuasive value on a material issue and no published opinion of this or another court would serve as well. Id. [FN71]. Anastasoff v. United States, 223 F.3d 898, 899 (8th Cir.), vacated, 235 F.3d 1054 (8th Cir. 2000) (en banc). [FN72]. See id. at 900 ("In sum, the doctrine of precedent was not merely well established; it was the historic method of judicial decision-making, and well regarded as a bulwark of judicial independence in past struggles for liberty."). [FN73]. Id. [FN74]. Id. at 903. [FN75]. Id. at 904 (quoting Joseph Story, Commentaries on the Constitution of the United States
377-78 (1833)). [FN76]. Id. [FN77]. Id. [FN78]. Id. at 903. [FN79]. Id. at 903 n.14. [FN80]. See Anderson v. DART, 180 F.3d 265 (5th Cir. 1999) (mem.). [FN81]. See Williams v. DART, 242 F.3d 315 (5th Cir. 2001). [FN82]. See Williams v. DART, 256 F.3d 260, 261 (5th Cir. 2001) (Smith, J., dissenting). [FN83]. United States v. Lopez-Pastrana, 244 F. 3d 1025 (9th Cir. 2001). [FN84]. See also United States v. Goldman, 228 F.3d 942 (8th Cir. 2000) (noting only authority on a sentencing guidelines point in the Eighth Circuit was an unpublished case, but stating that Anastasoff required that the case be followed). For a thoughtful discussion in the context of the state system, see Johns v. State, 35 P.3d 53, 63-67 (Alaska Ct. App. 2001) (Mannheimer, J., concurring). [FN85]. See Encore Video, Inc. v. City of San Antonio, No. Civ.A. SA-97- CA1139FB, 2000 WL 33348240 (W.D. Tex. Oct. 2, 2000); see also Cmty. Visual Communications, Inc. v. City of San Antonio, 148 F. Supp. 2d 764, 774 (W.D. Tex. 2000). [FN86]. This is the simple answer to the question posed by Boggs and Brooks: "One of the great puzzles of the unpublished opinion debate is why so many commentators believe [the argument that citation rules are necessitated by volume] is not good enough." Boggs & Brooks, supra note 4, at 19. [FN87]. See, e.g., id. (arguing that the publication criteria are unrealistic and unfixable); Robel, supra note 3, at 941-44 (making a similar argument). [FN88]. See supra note 9 and accompanying text. Even strong supporters of citation bans admit that the distinction between published and unpublished opinions "has become a fine, almost meaningless, distinction in a world of electronic legal research." Boyce F. Martin, Jr., In Defense of Unpublished Opinions, 60 Ohio St. L.J. 177, 186 (1999). [FN89]. See, e.g., Kozinski & Reinhardt, supra note 4 (arguing that predictive power is in the language and not the fact of a decision, and that judges should be able to control the meaning of precedent by constraining the concept to those opinions that use particular, authoritative kinds of language). [FN90]. See, e.g., Berman & Cooper, supra note 4; Douglas A. Berman & Jeffrey O. Cooper, Passive Virtues and Casual Vices in the Federal Courts of Appeals, 66 Brook. L. Rev. 685 (2001) [hereinafter Berman & Cooper, Passive Virtues]; Boggs & Brooks, supra note 4 (the authors are a judge of the Sixth Circuit and an attorney); Kozinski & Reinhardt, supra note 4. [FN91]. Cooper & Berman suggest there may be changes to appellate courts' internal rules that would ease the perceived burden caused by abandoning citation rules. One possibility would be to ease or reconfigure the federal appellate rules regarding the precedential strength of the decision of the first appellate panel to consider an issue. See Berman & Cooper, Passive Virtues, supra note 90. [FN92]. See id. [FN93]. This thought exercise assumes that judges both are serious about their claims that they cannot do more than they are doing now and self-disciplined enough to continue doing it, knowing that all opinions are similarly open to scrutiny. [FN94]. Martin, supra note 88, at 192. By using the word "torrent," Judge Martin may be making a different argument, one closer to the earlier arguments about the problems of searching through reams of decisions articulated by the earliest proponents of the publication plans. Given the views expressed in the White Commission surveys and the actual practices of those respondents, coupled with the efficiency of electronic searches, such an argument has little modern force. [FN95]. Kozinski & Reinhardt, supra note 4, at 44. [FN96]. Id. at 81. At bottom, these judges argue that the law-development role of the intermediate appellate courts, coupled with the press of caseload, requires that they be able to control the meaning of precedent issued by their court. "Not worrying about making law in 3800 [unpublished opinions each year] frees us to concentrate on those dispositions that affect others besides the parties to the appeal-the published opinions." Id. at 44. Indeed, the claim is even stronger: by focusing on the language rather than the fact of decisions, they claim that "judges-like legislators-have the power to enact prospective legal rules through opinion drafting. Put another way, a judicial decision has only as much precedent as the writing judges intend to give it." Boggs & Brooks, supra note 4, at 22 (discussing Kozinski & Reinhardt article). [FN97]. See, e.g., Blatt, supra note 7, at 629 ("A text acquires meaning only by reference to its readers. The shared understanding of such readers constitutes the 'interpretive community' for the text."); Fish, supra note 7. [FN98]. Although I suspect the burden is overstated: judges do not currently respond to every case a lawyer cites, and if a previously uncitable case is one that requires a serious response, it is presumably because it makes a serious point. [FN99]. Where to start? At a minimum, precedent offsets optimal justice in individual cases for other values, such as predictability, stability, or cross-case fairness. For what remains the most succinct explication and critique of the argument from precedent, see Frederick Schauer, Precedent, 39 Stan. L. Rev. 571 (1987). END OF DOCUMENT