Copr. © West 1999 No Claim to Orig. U.S. Govt. Works

85 CALR 541

California Law Review

March, 1997

Kirt Shuldberg [FNd]

Copyright © 1997 California Law Review, Inc.; Kirt Shuldberg

Table of Contents

Introduction ........................................................ 542

I.    History ............................................................. 545

      A. Historic Rationale for Selective Publication ..................... 547

        1. Cost ........................................................... 547

        2. Judicial Efficiency ............................................ 548

B. Historic Rationale for Prohibiting Citation to Unpublished

        Opinions .......................................................... 549

II.   The Problem: Behavior of Judges and Litigants Demonstrates That

        Unpublished Does Not Mean Unimportant ............................. 551

III.  A Critique of Historic Rationale in Light of Current Technology ..... 556

      A. The Selective Publication Argument Revisited ..................... 557

        1. Storage Costs .................................................. 558

        2. Research Costs ................................................. 559

        3. Intellectual Costs ............................................. 561

      B. The No-Citation Rationale Revisited .............................. 563

IV.   Fundamental Decisions in Crafting Citation Policy ................... 566

      A. Whether to Allow Electronic Dissemination of Unpublished Opinions  566

        1. Access Inequity ................................................ 566

        2. Judicial Accountability ........................................ 567

        3. Informational Importance ....................................... 567

      B. Circuit Rules on Citation to Unpublished Opinions ................ 569

V.    A Model Rule ........................................................ 570

      A. Electronic Dissemination Should be Allowed ....................... 571

      B. Citation to Unpublished Opinions Should Be Allowed When No

        Published Opinion Would Serve as Well ............................. 571

      C. Unpublished Opinions Should Be Given Persuasive Authority ........ 573

Conclusion .......................................................... 574

*543 For centuries scholars have lamented the rapidly expanding collection of case law, fearing that storing and processing this information would eventually become impossible. In the early 1970s, in response to these concerns, the Federal Circuits developed and implemented a variety of plans limiting publication to relatively few of the thousands of opinions they produced each year, and restricting citation to unpublished opinions. Limiting publication has had several benefits. First, costs are reduced and fairness increased: the fewer opinions published, the lower the cost of legal research and the less significant disparities in resources are in determining legal outcomes. And, judicial efficiency is increased: because judges need not spend time perfecting opinions that will not be published, they have more time to spend on those that will be. However, this Comment argues, time has shown that unpublished does not necessarily mean unimportant; unpublished opinions are frequently used even if citation to such opinions is prohibited. Moreover, with the advent of the computer, the world of legal research has changed dramatically. Accordingly, this Comment critically examines the original justifications for limiting the publication of federal appellate opinions, concluding that many of the limited publication/no-citation plans currently in use are suboptimal in light of modern digital storage and research capabilities. Finally, the Comment proposes a solution in the form of a model rule focusing on setting appropriate guidelines for the availability and citation of unpublished opinions.

Introduction

Each generation, our courts struggle to devise means of coping with an ever- increasing case load. [FN1] For federal appellate courts, one remedy has been to publish only selective judicial opinions, and prohibit citation to unpublished opinions. For more than twenty years, *544 each of the federal circuits has devised and administered its own plan toward that end. Justifications supporting the structure and content of these plans, developed in the early 1970s, were premised on a system of legal information found almost exclusively on the printed page. [FN2] Today, however, legal information can be stored, accessed and retrieved in digital form including on-line subscription services, such as Westlaw and Lexis, CD-ROM, and the Internet. Because information is the "stuff" of law, [FN3] this change in technology has had a profound impact on the law's shape as well as its use. As one scholar has noted:

[I]nformation technology is different [from other technological advances] and presents the law with a very different kind of challenge. It is different because . . . the law runs on information and because much of law is information. Thus, all of law is not affected by the automobile because law is not composed of automobiles. But law is, in almost all its parts, dependent on communication and information. [FN4]

The digital availability of information directly impacts both what it means for an opinion to be "unpublished" and the rationale supporting the content of existing limited publication plans. When the printed volume was the sole source of law, the decision not to publish an opinion meant that the public did not have access to that opinion. Now, however, opinions that are not published in print are nevertheless made available to the public in digital format. Moreover, digital technology allows for more efficient storage and research of information, which impacts the rationale underlying the limited publication plans.

In this Comment, I seek to critically examine these existing limited publication/no-citation plans, which are tied to an older, less powerful and less interactive means of communication, and to challenge the applicability of these plans to the changed circumstances of today's digital world. I first provide a brief history of selective publication, examining the historic rationales for the decision to limit publication and restrict citation. I then revisit these rationales and concerns in light of current technology, focusing on theoretical as well as practical limits of existing plans. From there, I examine several possible solutions. Finally, I propose a model rule.

*545 I History

Complaints about the proliferation of case law are not a new phenomenon. Generation after generation of commentators has feared rapidly expanding case law and made dire predictions of the possible effects. [FN5] More than three and a half centuries ago, Sir Francis Bacon, then Lord Chancellor, recommended to King James I that cases "merely of iteration and repetition" be omitted from the case reports. [FN6] The author of an 1824 article complained:

[T]he multiplication of reports, emanating from the numerous collateral sources of jurisdiction, is becoming an evil alarming and impossible to be born [sic]. . . . Such has been this increase, that very few of the profession can afford to purchase, and none can read all the books which it is thought desirable, if not necessary to possess. By their number and variety they tend to weaken the authority of each other, and to perplex the judgment. [FN7]

A century later, another commentator, astounded at the growth rate of case law, noted that "[s]ixty-five thousand precedents [had been] added to the already existing myriads of precedent in the brief space of five years!" [FN8] This jurist then lamented:

Given that commentators complained that the publication of 65,000 cases in five years was an unmanageable mass, it is unsurprising that similar protests are heard today. A jurist of our generation, noting the limited capacity of judges and lawyers to research and assimilate the mass of judicial opinions, commented that "[t]hose limits are dangerously near at present and in some systems may already be exceeded," and that "[c]ommon law in the United States could be crushed by its own weight if present trends continue unabated." [FN11]

Despite criticisms like these, however, federal appellate courts continued to authorize publication of virtually every opinion well into the 1960s and 70s. [FN12] Even one-word Memorandum Decisions were published. [FN13] In 1964, the Judicial Conference of the United States expressed its concern about the "ever increasing practical difficulty and economic cost of maintaining accessible private and public law library facilities . . . ." [FN14] Accordingly, the Judicial Conference recommended that the United States Courts of Appeals authorize publication of "only those opinions which are of general precedential value . . . ." [FN15] In 1972, endorsing the detailed recommendation made earlier that year by the Board of the Federal Judicial Center, the Judicial Conference directed the circuits to develop plans to limit publication of judicial opinions. [FN16] Each circuit responded by devising its own publication plan. According to the Judicial Conference, this diversity in plans constituted "11 legal laboratories accumulating experience and amending their publication *547 plans on the basis of that experience." [FN17] Because the Judicial Conference saw the possible rewards of such experimentation as "rich," it agreed to delay prescribing further guidance. [FN18] Today, after more than twenty years of such "rich" experimentation, diverse circuit plans still exist, and the several circuits continue to administer their individually devised publication and citation plans.

These plans involve two key issues: (1) whether and how to limit publication; and (2) whether to allow future parties to cite to unpublished opinions. Following this structure, I will first examine the concerns underlying the decision to publish selectively. I will then address the related issue of prohibiting citation to those decisions that have not been published.

A. Historic Rationale for Selective Publication

As noted, the decision to publish selectively was primarily driven by a generalized fear of the exponential growth in case law. Underlying this generalized fear are a number of more specific concerns. I have identified two categories under which to discuss these issues: (1) cost concerns and the underlying issue of fairness; and (2) concerns of judicial efficiency. I will briefly outline each of these concerns below and will then revisit them in Part III, discussing their relevance in light of current technology.

1. Cost

Perhaps the dominant concern underlying limited publication plans is the cost associated with expanded publication. The term "publication" has historically referred to the printed reproduction of a case in a bound volume. Purchasing these bound volumes, from either a private or a government-sponsored publisher, is expensive. [FN19] Moreover, it costs money to maintain a collection of these volumes, for the more volumes that are published and purchased, the larger the physical facility required in which to store them. This is why critics complain about miles of bookshelves stretching away "into the dim distance." [FN20]

In addition to its effect on the cost of buying and storing law reporters, expanded publication increases research costs. More time is required to sort through more cases. The legal researcher must spend *548 additional time searching to find the relevant precedent and to distinguish the irrelevant. Because time is money in legal practice, the cost of research increases with the increase in case law.

Driving this cost issue is an underlying concern for fairness. This concern is not simply about judges and lawyers incurring greater costs to update their libraries. Rather, the concern is that these higher costs will ultimately be passed on to the consumer. If it becomes more expensive to research the law, those with greater resources will be increasingly favored and the access-to- justice inequities that currently exist between the rich and the poor will be further magnified. [FN21]

2. Judicial Efficiency

The federal court system faces an enormous caseload. [FN22] One means of helping to alleviate the backlog of cases is to limit publication of appellate opinions. [FN23] Appellate judges spend a great deal of time writing opinions. [FN24] The judicial efficiency argument urges that if opinions are selectively published, judges can spend less time writing opinions and more time deciding a greater number of cases.

The judicial efficiency argument is premised on the belief that judges can predetermine which cases are so unimportant that publication is unnecessary. To understand which opinions are worthy of publication and, therefore, efficient to publish, it is necessary to first examine the primary functions served by appellate opinions.

Courts, through their opinions, serve two paramount social functions: resolving disputes and enriching the supply of legal rules. [FN25] These two functions have been termed the dispute-settling and the law-*549 making functions of courts. [FN26] The dispute-settling function of a judicial decision is to decide the case, putting an end to the immediate dispute. [FN27] The opinion explains to the parties the result of the case. The law- making function of a judicial decision is "to establish the law itself, to determine what the content of the law shall be. This is the function of common law precedent, and of the rule of stare decisis." [FN28] Pursuant to this function, courts not only establish the law, but also criticize legislation and legal doctrine, and comment upon social and legal problems. [FN29]

Selective publication of opinions is justified as follows: The public at large receives benefits only from those opinions that serve the law-making function. [FN30] Those appellate opinions that serve only the dispute-settling function benefit only the trial court and the parties to the case, and have no value to the general population. Accordingly, the ends of judicial efficiency are served by limiting opinion publication to those opinions which establish law or have precedential value. These are the only opinions that are worthy to be preserved for future generations on the printed page. [FN31] Consequently, it is only upon these opinions that judges should spend time polishing their prose and checking for proper citation. [FN32]

B. Historic Rationale for Prohibiting Citation to Unpublished Opinions

Once the decision has been made to limit publication, a second question arises: Should citation to unpublished opinions be allowed? The historical justification for prohibiting citation can be subdivided into two related arguments: allowing citation would (1) frustrate the purposes of selective publication and (2) promote unfairness. [FN33]

The first argument against citation to unpublished opinions is that if it is allowed, the purposes of selective publication--reduced cost and increased judicial efficiency--will be frustrated. It is argued that if citation to unpublished decisions is permitted, a market for these opinions *550 will develop. [FN34] Legal researchers will feel compelled to search these opinions for relevant case law. Practitioners and law libraries will then purchase and index these opinions in response to the developing market. Accordingly, substantial cost savings will be lost. [FN35]

Judicial efficiency may also suffer since, if citation is allowed, judges may feel obliged to spend more time writing these opinions. For example, an opinion written only for the parties to the case can exclude background and factual information, with which it can be presumed the parties are already familiar. If, on the other hand, a case may be cited by future litigants, this familiarity may not be taken for granted. Therefore, judicial energy may be spent reciting such background information.

The second argument against allowing citation to unpublished opinions is that to do so would be unfair. Unpublished opinions are more difficult to access because they are not published in the bound volumes and are not located in the traditional digests available to the legal researcher. The combination of this restricted access with freely allowed citation would benefit attorneys with greater time and resources.

Another fairness concern is that those who regularly practice in a particular court will benefit. [FN36] Recall that opinions are not published because they are considered to have value only to the particular litigants in the case and not to the public at large. [FN37] Because unpublished opinions are distributed to the parties to the case, the concern is that repeat litigants, such as government attorneys, may maintain a file of these opinions that would be unavailable to future opponents who may litigate similar issues. [FN38] Citation should therefore be prohibited so that these *551 attorneys do not benefit at the expense of smaller or less well-funded litigants. [FN39]

These historic rationales for the limited publication/no-citation plans warrant re-examination in light of current technology. Increased access to both published and unpublished legal opinions through the computer brings to the forefront new concerns while relegating some old concerns to the past. Further, as technology alters the available body of law, it exacerbates some of the practical problems with current limited publication/no-citation plans. I will revisit the historic rationale in light of these changes below.

II The Problem: Behavior of Judges and Litigants Demonstrates That Unpublished

Does Not Mean Unimportant

Selective publication is intended to serve as a sorting device, separating the wheat from the chaff. Opinions that have general precedential value or other public significance are separated from those that do not. If the plans function without error, then examining technology's effect on the justifications for selective publication would be superfluous. If, on the other hand, the selection mechanism works imperfectly, then the plans are open to question. Therefore, before embarking on an examination of the effects technological advancements have had on the rationale for limiting publication, I first seek to demonstrate that a problem does, in fact, exist: many unpublished opinions do contain legal analyses that are important to future litigants and to the public at large.

Limited publication plans generally seek to publish only cases of general precedential value. [FN40] Accordingly, it may be widely assumed that unpublished decisions are of little value because they are largely frivolous cases or little more than a mechanical application of existing law to a set of facts. [FN41] In many cases this assumption is undoubtedly true. [FN42] In many other instances, however, this is not the case. As Judge *552 Posner has noted: "[M]any appeals that formerly would have been decided with a full opinion . . . are now decided with an unpublished opinion. These are not frivolous appeals; one-line treatment ["affirmed," for example] would be inappropriate. They call for an opinion and they get it, but it is not published." [FN43] If what Judge Posner says is true, then it cannot fairly be assumed that unpublished opinions are, by definition, unimportant. [FN44]

We need not simply take Judge Posner at his word. A survey of court behavior readily refutes the assumption that unpublished opinions are without value to future litigants. One illustration of this fact is that unpublished opinions are sometimes accompanied by a dissent. [FN45] If two judges, both ruling from the same trial court record, disagree about the correct application of the law, it would seem quite doubtful that the opinion was such a mechanical application of law that it is of no value to future litigants. [FN46] Recognizing this contradiction, three circuits have specifically provided that an opinion accompanied by a dissent should be published in some circumstances. [FN47] Certainly, parties disputing an issue similar to that previously decided would benefit from the divided reasoning of the prior judicial panel. The court in the subsequent case would presumably also benefit by knowing the reasons for the split of opinion in the previous case.

A perverse twist on the mechanics of the publication decision further illustrates that some unpublished opinions are indeed important. In the Federal Circuit, opinions accompanied by a dissent are required to be published only if the dissenting judge so insists. Judge Philip Nichols, Jr. of the Federal Circuit candidly admitted that, if dissenting, he would never insist on publication. His rationale: Better to have the opinion banished from existence than be bound by what he considers *553 bad precedent in the future. [FN48] Judge Nichols also notes that in circuits which require the publication of opinions accompanied by a dissent, [FN49] an even more disingenuous practice likely occurs. In these circuits, he believes, "tying up the question of dissenting with publication may work adversely on the dissenter, constraining him not to dissent . . . ." [FN50] Apparently this practice is more than mere speculation. A 1991 article attributed one scholar as saying that law clerks had told her "that judges sometimes would agree not to dissent if an opinion remained unpublished." [FN51] It can fairly be assumed that these agreements also were made to limit disagreeable precedent. Paradoxically, this practice demonstrates the importance of rendering a decision "unimportant" (or, more correctly stated, the importance of rendering a decision non-precedential). As I will later argue, suppressing precedent in this fashion is directly at odds with the fundamental nature of an evolving common law. [FN52]

California's experience with unpublished opinions from the federal courts provides another example of the future value these opinions sometimes serve. In California, state courts of appeal have disagreed over the meaning of Rule 977, which prohibits citation to "[a]n opinion that is not ordered published." [FN53] The rule does not specify whether it applies only to unpublished opinions of state appellate courts, or whether unpublished opinions of federal courts are also within its intended scope. [FN54] Because of this ambiguity, some state appellate courts have cited and relied on unpublished federal opinions while others have held these opinions beyond the cognizance of the state courts. [FN55] The *554 disagreement has generated enough controversy to motivate the California Supreme Court to recently request public comment on proposed amendments to Rule 977. [FN56] Regardless of the ultimate resolution of this dispute, the fact that unpublished opinions from the federal courts have generated such controversy demonstrates the law-making function some of these opinions serve.

Additional evidence that unpublished opinions can be important to the resolution of future cases can be found in the confused practice of the Fifth Circuit. The Fifth Circuit limits case publication by requiring that publishable judicial opinions meet a defined set of criteria. These Fifth Circuit criteria are designed to publish "every opinion believed to have precedential value." [FN57]

However, the judges sometimes miss the mark. In Hodges v. Delta Airlines, Inc., [FN58] the court stated that, contrary to its belief that the federal Airline Deregulation Act of 1978 should not have preemptive effect, it was bound by an earlier, unpublished case to hold that the federal act preempted a tort suit for personal injuries caused by unsafe in-flight conditions. The court was bound in this case because in the Fifth Circuit, "[u]npublished opinions . . . are precedent." [FN59]

Similarly, in Peters v. United States, [FN60] the court again relied on an unpublished opinion as precedent. However, not only did the court rely on an unpublished opinion in Peters, but it also originally intended that *555 Peters itself be an unpublished opinion. [FN61] Had the court not acquiesced to the publication request of the Assistant United States Attorney on the case, the court may have fashioned an "underground stream of precedent-- unpublished opinions relying on other unpublished opinions." [FN62]

Although these experiences are anecdotal, they do refute a fundamental premise upon which limited publication plans are built. If these opinions do indeed contain analysis important to the resolution of future cases, then they are too important to allow judicial efficiency concerns to take precedence over the value of common law precedent. To quote one commentator: "[with non- publication] [s]tare decisis is twice diminished. First, the decision itself is freed from the responsibility to reason within full view. Second, an increment of precedent is rendered unusable." [FN63]

These examples do not represent isolated experiences. A number of scholars have demonstrated that in many cases unpublished decisions are in fact precedential or do have future importance to legal decision makers. These studies range from the anecdotal to the empirical. [FN64] For instance, many scholars have examined sample groups of unpublished opinions and have noted numerous instances of unpublished opinions that in fact did make law. [FN65] These studies and other anecdotal evidence readily demonstrate that many unpublished opinions do in fact turn out to contain analyses that are important to the resolution of future cases.

*556 III A Critique of Historic Rationale in Light of Current Technology

In 1964, when limited publication was first officially proposed by the Judicial Conference of the United States, [FN66] computer-assisted legal research was little more than theory. [FN67] A few years later, when the federal circuits devised and implemented their limited publication plans in the early 1970s, [FN68] computer-assisted legal research was an idea whose time would come, but as a commercially-available product it was only in its infancy. [FN69] Lexis, the first commercially-available computer-assisted legal research service, was introduced to the market in April 1973. [FN70] By the fall of 1973, a handful of law firms had begun using Lexis to perform legal research. [FN71] In 1975, Lexis' first competitor entered the market when a primitive Westlaw product was introduced. [FN72] Computer-assisted legal research was thus only beginning to proliferate when the publication plans were adopted.

Accordingly, the arguments in favor of the limited publication plans were necessarily premised on legal storage and research as it then existed--on the printed page. Questions of cost, fairness, access and efficiency were all fundamentally linked to this paper-based publishing regime. Reevaluation of the premises for selective publication is thus relevant in view of the widespread availability of digital information today.

Advancements in information communication have revolutionized legal research and access to legal information. Legal information is no longer available exclusively in print and the researcher is now no longer constrained by the limitations of the printed page. Instead, this information can be stored, indexed and retrieved digitally. Because it is an institution built on the creation, storage, processing, and communication of information, "law cannot remain unaffected by large-scale changes in the communication of information." [FN73] Given this premise, I seek to critically examine the limited publication rules--rules that are tied to an older, less powerful and less interactive means of communication--and *557 to challenge the applicability of these rules to the changed circumstances of today's digital world.

A. The Selective Publication Argument Revisited

Digitally available case law has affected the rationale for limiting opinion publication in several ways. Most dramatically affected have been the direct costs of performing legal research. The impact of advanced communication technology on the price of legal information, however, is complex [FN74] and I do not endeavor here to sort out the effect technology has had generally on price. [FN75] Instead, I will focus on the *558 effects of digitally- available case law upon the rationale for the no-publication/no-citation plans.

1. Storage Costs

The most obvious effect technological advancements have had on cost involves the cost of physically storing information. Today, the practitioner's library need not have space for miles of bookshelves. One CD-ROM disc, which is wafer thin and a mere five inches in diameter, can store as much information as that contained in 30 to 60 printed volumes. [FN76] The information contained in thousands of volumes can now be stored in a file cabinet. The cost savings from the resulting need for less physical space can be dramatic--especially in the high-rent office space in which many law firms are located. [FN77] Moreover, the current cost of case law on CD-ROM is far less than that of the printed case reporters. [FN78] The physical storage cost savings can be realized while simultaneously saving money by purchasing the information in digital, rather than printed, form. [FN79]

While public law libraries will likely continue to purchase the books rather than the compact discs, this need not be the case. One can imagine a library of the future where patrons sit at computer terminals searching through compact discs or on-line information rather than browsing through bookshelves. [FN80] The cost savings resulting from storing legal information in digital, rather than print, format would be substantial. However, even if we assume that the library of the future will continue to look like the library from the past and retain books, unpublished *559 opinions will not significantly add to the book collection because, for the most part, they are only published electronically. [FN81] Current technology allows an opinion to be ordered "not published" but nevertheless to be widely disseminated in digital format. Thus, forests do not have to be destroyed [FN82] and library bookshelf space does not have to increase in order for parties with digital capabilities to have access to, and use of, unpublished opinions.

2. Research Costs

Although more case law seems inevitably to lead to increased costs, technology has probably mitigated this effect. The decision to selectively publish was primarily driven by a generalized fear of the exponential growth in case law. [FN83] This concern was particularly keen given the inherent limitations of the paper-based publishing regime that dominated at the time the publication plans were introduced. Who can doubt that the prospect of manually negotiating millions of cases in order to find the relevant few was indeed overwhelming? [FN84]

As the availability of computer-assisted legal research has become more widespread, the use of manual research has declined. A researcher today can sift instantly through the vast sea of case law, retrieving or Shepardizing a relevant case in a matter of seconds. Accordingly, the need to search through a greater number of cases is at least partially offset by the aid of more efficient computer-assisted research techniques. Computerization is not a panacea, however. For all of its power and speed, it brings with it new costs and unique problems. [FN85]

One such problem, particularly relevant to the electronic dissemination of unpublished opinions, is that as databases get larger the ability to efficiently locate relevant documents diminishes. [FN86] This phenomenon occurs due to the inexact nature of free-text searching. With free-text searching, the researcher searches for words that are deemed relevant to *560 the problem at hand. [FN87] The problem with this approach is that there is an imperfect correspondence between words and ideas. [FN88]

[M]uch of the meaning of natural language texts is contained in complex relationships between words rather than in the individual words themselves. Simple word recognition of the kind done by LEXIS and WESTLAW cannot discern the meaning of complex expressions because the relationships between words in natural text are much more subtle than the connectors that may be used to join terms in [computer-assisted] searches. [FN89]

Due to this low correlation between words and ideas, a free-text search retrieves both relevant and irrelevant documents. In small databases, this is not a problem because the researcher can easily scan the documents to weed out the irrelevant. However, in very large databases winnowing the irrelevant can become impractical. [FN90] In order to increase the precision of the search (retrieve fewer irrelevant documents), it thus becomes necessary to add more detail to the search request. This tactic does succeed in retrieving fewer documents; the problem, however, is that the search will also exclude relevant material. [FN91] Accordingly, the larger the collection of documents in a database, the poorer the performance of the computer-assisted search. [FN92] If unpublished opinions are added to digital databases, their inclusion will exacerbate this problem.

This is not a fatal flaw, however. As one prominent law librarian points out, failure to retrieve all the relevant cases in one workable search does not make the search useless. [FN93] In reality, the researcher need only retrieve a few relevant cases. The researcher is then tapped into the larger legal discussion of the issue at hand. [FN94] By finding even "one good case," [FN95] other relevant cases and authorities can then quickly be found by locating the authorities cited in the cases retrieved. The computer's power facilitates this process by allowing a researcher, with only a few keystrokes, to instantly locate the relevant text of a cited authority. [FN96] The overall effect of the digital revolution--even when consider ing *561 the unique problems it brings--has been to make negotiating lots of law less daunting. The general fear about too much case law, though real, is surely mitigated by the power of advanced communication technology. Thus, increased research costs provide less impetus for limiting publication than they did before computerized research was a reality.

3. Intellectual Costs

A final cost of too much published case law that has been discussed in the academic literature is not financial but intellectual. Some commentators argue that if there are too many decisions, there will be too much diversity of opinion. [FN97] Accordingly, it will not be possible to distill coherent legal principles from this amorphous mass of case law and confusion and inconsistency will result. [FN98] As one author stated:

When the number of printed cases becomes like the number of grains of sand on the beach, a precedent-based case-law system does not work and cannot be made to work. . . .

. . . The theory of precedent depends, for its ideal operation, on the existence of a comfortable number of precedents, but not too many. [FN99]

However, the fact that a greater number of opinions is published on a discrete topic of law does not necessarily mean that confusion will result. A string of decisions might ultimately reach the same result, thus solidifying, rather than weakening, the understanding of that topic of law. Indeed, Judge Posner has noted that "[d]espite the vast number of published opinions, most federal circuit judges will confess that a surprising fraction of federal appeals are difficult to decide, not because there are too many precedents but because there are too few on point." [FN100]

That many decisions may reinforce prior holdings does not mean that the more recent holdings should not be published. There is value to a litigant in knowing the relevant strength of a legal doctrine. The litigant will benefit by knowing whether a legal principle has been consistently upheld in varied factual circumstances or whether the doctrine has only been challenged a few times. Similarly, the litigant may benefit by knowing whether the legal doctrine has been dormant for a number of years or whether it has withstood challenge more recently. If more recent opinions have not been published because the law was *562 merely applied to new facts, the litigant will be denied the benefit of knowing both the strength and the vintage of the doctrine.

Moreover, if the publication of more cases actually does result in inconsistent rulings, then appellate courts will be forced to clarify the law. Indeed, the implications of the argument that more cases will lead to confusion are striking. Unstated in this hypothesis is the admission that there will indeed be inconsistent rulings but that, to avoid confusion, some of these should simply go unpublished. This notion goes directly against a key argument for limiting publication in the first place--that only those opinions that make law have enough future value to deserve publication. Decisions that are inconsistent with the weight of precedent are, by definition, law-making. These opinions therefore should be published even in the face of fears about an incoherent mass of case law.

Some legal theorists have suggested a modern rationale supporting the public availability of inconsistent legal opinions. Citing the power of current communication technology, Jean Stefancic and Richard Delgado suggest that today's communication technology could pave the way toward faster law reform [FN101]--arguably a "good" thing. A primary tenet of critical legal theory is that "legal reasoning rarely, if ever, has just 'one right answer.' Instead, there will be multiple interpretations of the available precedent from which judges may choose on the way to resolving the case before them." [FN102] Advanced technology provides legal researchers with the capacity to perform such tasks as free-text searching. This ability frees the researcher from the constraints of the West paradigm--a system whereby "the information in each of the cases was parsed into a preexisting framework that inevitably tended to suppress subtle changes and to enforce judicial and professional conformity and conservatism." [FN103] It is thus much easier for the researcher to locate different legal interpretations on emerging points of law. [FN104] "[T]he stability of printed texts [has] been challenged by computerization. Precedent will become more manipulable; arguments and lines of cases for opposing viewpoints will be readily available, arguably to the *563 advantage of law reform." [FN105] Since unpublished opinions frequently contain one of the multiple interpretations from which judges may choose, their widespread availability would seem an important ingredient to those pushing for legal reform.

The fear of common law's being crushed by its own weight appears to be voiced in frustration with the overwhelming prospect of searching so much law. At the time these fears were voiced, legal research was done manually and the task of locating relevant law was very time-consuming. In the current era of the computerized search, however, this prospect is much less daunting. Given the power of computer-generated search engines, reports of common law's impending demise appear to have been greatly exaggerated.

B. The No-Citation Rationale Revisited

Because it is believed that allowing citation to unpublished opinions frustrates the objectives of selective publication (reduced cost and increased judicial efficiency) and promotes unfairness, some circuits forbid citation to unpublished opinions by parties unrelated to the case. [FN106] This prohibition relies on the assumption that, by forbidding citation, the courts will preclude the use of unpublished opinions altogether.

The weight of evidence, however, points to the opposite conclusion. Barring citation does not prevent the use of unpublished opinions; it merely alters the character of that use.

Despite the official ban on citation, unpublished opinions are regularly used by litigants in making fundamental litigation decisions. For instance, one survey of the federal government, a frequent litigant, found that federal agencies have used such opinions in making litigation or settlement decisions and in determining whether to appeal or oppose an appeal. [FN107]

Attorneys and others may also use unpublished opinions as authority in settlement negotiations. For example, in Prudential-LMI Commercial Insurance Co. v. Colleton Enterprises Inc., [FN108] a 1992 unpublished decision, the Fourth Circuit court held that a hotel damaged by Hurricane Hugo could not recover "windfall" losses from its insurer *564 for being unable to accommodate the influx of repair persons and construction workers who would have booked rooms after the disaster if the hotel had not been damaged. [FN109] According to plaintiffs' attorneys, insurance adjusters later used an insurance trade journal article about the decision to deny similar claims after Hurricane Andrew. [FN110] Thus, although the Fourth Circuit had banned this opinion's citation, it did not thereby render the decision "unusable."

There are other ways litigants could use unpublished opinions to their advantage. For example, litigants aware of an unpublished opinion's existence can make unacknowledged use of that opinion in their appellate briefs by borrowing successful arguments or by lifting exact language and hypotheticals. Attorneys can also know "which of the court's published opinions were considered relevant by its citations. . . . [And] if inconsistency exists between a court's published and unpublished work, it would be of great practical value to know what the court does as well as what it says." [FN111]

The court itself may also use unpublished opinions as de facto authority. This may already be happening at the state level. For instance, one member of a Wisconsin state bar committee formed to study the issue of citation to unpublished opinions claims to have heard "stories of lawyers being interrupted when arguing a point in chambers by a judge who blurts out, 'Well, the appellate court ruled the other way in this unpublished opinion I have in my desk drawer."' [FN112] Such a scenario also seems quite possible in the federal circuits, since copies of unpublished opinions are distributed to circuit and district court judges. [FN113] Indeed, during the 1974 hearings before the Commission on Revision of the Federal Court Appellate System, Judge Sprecher of the Seventh Circuit advocated the creation of an indexing system for unpublished opinions in order to avoid intra-court conflict: "[W]e are going to have to have some kind of an intra-court index of unpublished opinions, indexed according to the subject matter and so forth. These matters are going to have to be available for the court, even though they cannot be cited by the court or to the court." [FN114] The evidence thus suggests *565 that unpublished opinions are used by litigants and by the courts even where citation has been forbidden.

This use of unpublished opinions in practice calls into question the purposes the no-citation rule is designed to serve. Recall that the first justification for forbidding citation was that permitting citation to unpublished opinions would frustrate the purposes served by limiting publication--reduced cost and increased judicial efficiency. While allowing citation to unpublished opinions may not undermine judicial efficiency, [FN115] it clearly does undermine the goal of reduced costs. [FN116]

Where litigants and courts continue to use unpublished decisions creatively, their inability to cite to them does little to alleviate, and in fact may compound, the cost concerns. As previously discussed, the evidence suggests that many attorneys and judges regularly use unpublished opinions. [FN117] Assuming this to be true, court participants spend actual time and resources searching these decisions, and the cost savings intended by not publishing, and by prohibiting citation, are not realized. Costs may actually increase where unpublished opinions are not widely distributed and economies of scale in such tasks as collecting and indexing cannot be achieved. Accordingly, because prohibiting citation does not preclude the use of unpublished opinions in practice, the rationale of reduced cost alone does not justify forbidding citation.

The second major rationale for prohibiting citation to unpublished opinions is that if citation is allowed, well-funded or repeat litigants will have unequal access to these decisions. [FN118] This is unfair to the "little guy": If he can't find them, he can't cite them and he is thus disadvantaged. However, the fact that litigants nevertheless use these decisions despite prohibited citation raises another problem of fairness: Unacknowledged use of unpublished opinions occurs; therefore well-heeled and institutional litigants benefit from disparate access. [*566 FN119] Thus, the core of the fairness argument is access: Unfairness results from unequal access.

The fairness argument must proceed with today's changed communications capabilities in mind. When the limited publication plans were adopted, unpublished opinions were essentially banished from public existence since the bound volumes were the sole source of case law. Technological advances, however, have changed this reality. Today, in jurisdictions that allow for public distribution of unpublished opinions, on-line research services make these decisions widely available. Despite an opinion's unpublished status, public access is possible. Modern courts must therefore account for the electronic availability of unpublished opinions when fashioning their citation rules.

IV Fundamental Decisions in Crafting Citation Policy

In crafting citation policy, courts must make two fundamental decisions: (1) whether to allow the electronic dissemination of all unpublished opinions, and (2) whether to allow citation to those opinions. For the reasons discussed below, courts should allow both electronic dissemination of unpublished opinions and citation to these opinions for their persuasive value.

A. Whether to Allow Electronic Dissemination of Unpublished Opinions

Three federal circuits currently prohibit the electronic dissemination of unpublished opinions, [FN120] attempting thereby to deny access to these opinions to all but the parties to the case. Denying public access to judicial opinions in this fashion may succeed in limiting their use by some litigants. Prohibiting electronic dissemination, however, ignores other important policy considerations.

1. Access Inequity

One goal in limiting distribution of unpublished opinions is to deny access to all litigants, thereby leveling the playing field. If neither party to a dispute has access to unpublished opinions, then neither party will be at an advantage. However, as discussed above, repeat or institutional*567 litigants do have access to unpublished opinions in cases to which they were a party. [FN121] Indeed, one survey of governmental litigants found that a number of federal government litigation offices circulate copies of unpublished opinions and at least one such office admitted to indexing them. [FN122] When public access to these opinions is denied, institutional litigants, who have access by virtue of having been a party to numerous cases, are at a distinct advantage. Prohibiting electronic dissemination may thus actually preserve inequity because litigants who more frequently practice in particular courts will have disparate access to unpublished opinions.

2. Judicial Accountability

When a circuit court completely denies public access to judicial opinions, it removes an important check on judicial activity from the legal system. Public availability of judicial opinions helps to hold judges accountable to society for the decisions they reach. [FN123] Public scrutiny helps to maintain the integrity of the judicial system and assure that individual cases are fairly decided.

Forcing judicial decisionmaking into the light of day helps not only to assure fairness in fact, but, perhaps as importantly, to promote the appearance of fairness. When a judge's reasoning in a particular case is open to public scrutiny, litigants may be less likely to believe that the decision was arbitrary or unfair. Therefore, the public availability of unpublished opinions promotes judicial accountability, which may also enhance public confidence in the legal system.

3. Informational Importance

The decision to prohibit dissemination of unpublished opinions also ignores their informational importance to legal consumers and scholars. Litigants may make more efficient litigation decisions if they know the court's full record in each body of law. For instance, lawyers routinely use judicial decisions in making fundamental choices such as whether to litigate or whether to appeal: "What kinds of information could litigants get from unpublished opinions that are not foreseen by the publication plans? One important piece of information is the shape of the universe of decisions . . . in a particular area of law. The information might be as simple as knowing the odds." [FN124] An attorney may also wish to track decisions in an area of law to discern trends or to detect difficulties in the administration of a statute.

*568 Moreover, certain types of decisions are systematically consigned to non-publication, impairing the decisionmaking of litigants in these areas of law. [FN125] One scholar has asserted that unpublished opinions routinely result in much litigation in which the government is a party. These cases include "review of agency determinations in immigration and social security cases, Federal Tort Claims Act cases, criminal and habeas appeals, civil rights actions, and employment discrimination complaints against the federal government." [FN126] In jurisdictions that circumscribe the public availability of unpublished opinions, litigants in these and other fields of law may be unable to obtain large quantities of relevant and useful information.

The unavailability of unpublished opinions also undermines the legitimacy of the work done by legal scholars and social scientists. A majority of federal appellate court cases are unpublished. [FN127] Numerous studies have demonstrated that a significant number of these unpublished decisions involve cases that are non-routine, law-making, or politically significant. [FN128] In the aggregate, unpublished opinions thus contain a great deal of valuable information about actual court behavior. The exclusion of this information may severely undermine the validity of many important scholarly studies. [FN129] As a result, theoretical and practical insights drawn from knowledge of actual court behavior may be compromised to the public detriment.

In our legal system, which seeks the most well-reasoned resolution to legal disputes, judges are required to provide reasons for the decisions they reach. Well-reasoned analysis requires an informed decision maker. Accordingly, our rules of court should not limit the information available, but should encourage attorneys and judges to develop and perpetuate the most insightful arguments. If a court has articulated a well-reasoned or creative analysis, other decision makers can benefit from it only if they have knowledge of its existence. A system that opposes the use of unpublished opinions prevents lawyers from making their own best judgments about the relevant merits of judicial opinion, constraining the natural development of law. [FN130]

*569 B. Circuit Rules on Citation to Unpublished Opinions

In forming their limited publication/no-citation plans, most circuits chose to allow public access to unpublished opinions but to forbid their citation to the court. Recently, however, allowing citation to unpublished opinions has gained popularity. Six circuits currently allow citation, up from only two circuits in 1994. [FN131]

Within this trend toward allowing the citation of unpublished opinions, the circuits differ in their approach. Four of the six circuits emphasize that citing unpublished opinions, while allowed, is disfavored. For example, the Tenth Circuit rule on citation states in part: "Citation of unpublished orders and judgments is not favored. Nevertheless, an unpublished decision may be cited if it has persuasive value with respect to a material issue that has not been addressed in a published opinion and would assist the court in its disposition." [FN132] The Fourth, Sixth and Eighth Circuits similarly disfavor the citation of unpublished opinions but qualify their citation less restrictively than does the Tenth Circuit. Rather than limiting citation of unpublished opinions to issues not addressed in a published opinion, these circuits state that an unpublished opinion may only be used where "there is no published opinion that would serve as well." [FN133] By contrast, the rules of the Fifth and Eleventh Circuits contain no qualifying language and appear to freely allow the citation of unpublished opinions. [FN134]

*570 The six circuits allowing citation of unpublished opinions also differ in the level of authority they give to unpublished opinions. Two circuits appear to allow unpublished opinions to be cited as precedent, [FN135] while the other four circuits provide that unpublished opinions may be cited only for their persuasive value. [FN136] Important policy objectives can be served by assigning an optimal authoritative weight to unpublished opinions. As will be discussed below, courts should allow citation to unpublished opinions for their persuasive value.

V A Model Rule

Important policy considerations dictate that all judicial opinions should be subject to public critique. Accordingly, the publication plans of the various circuits should explicitly provide that all appellate opinions be made available for electronic dissemination. Further, in setting a policy on the citation of unpublished opinions, the federal circuits should allow attorneys and judges to cite to unpublished opinions when they are relevant to a material issue in the case, and give these opinions persuasive, but not binding, authority. A good example of a rule that has accomplished these goals is that fashioned by the Eighth Circuit. This rule states:

Unpublished 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, parties may cite any 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. A party who cites an unpublished opinion in a document must attach a copy of the opinion to the document. . . . When citing an unpublished opinion, a party must indicate the opinion's unpublished status. [FN137]

This rule best resolves the two most important questions regarding citation to unpublished opinions: In what circumstances should a party be allowed to cite an unpublished opinion and what authority should that opinion be given?

*571 A. Electronic Dissemination Should Be Allowed

Making unpublished opinions available preserves the integrity of the judicial system by allowing public scrutiny of all judicial opinions. Our system provides for an independent judiciary. [FN138] Federal judges are not elected and cannot be voted out of office. Judicial discretion is not checked by popular vote. Rather, two primary checks on appellate discretion exist. [FN139] The first is Supreme Court review. [FN140] Given the enormous caseload of the federal appellate courts, however, the chances for review of any given appellate opinion are very slight. [FN141] For example, in 1993, the Supreme Court decided 163 cases--ninety-six by written opinion and sixty-seven by per curiam or Memorandum Decision. [FN142] By comparison, the federal appellate courts in that same year disposed of more than 25,000 cases. [FN143] We must primarily rely, therefore, upon the second check on judicial discretion.

This second check on appellate discretion is the public critique of judicial decisions by legal consumers. [FN144] As previously discussed, the public availability of opinions helps to hold judges accountable for their decisions because they know that their reasoning and analysis will be exposed to the light of day. [FN145] For this check on judicial discretion to be effective, however, all judicial opinions must be subject to public scrutiny. This check disappears in a world where difficult decisions or faulty reasoning can be swept under the rug by not publishing an opinion, thus preventing public access. Accordingly, the courts should take advantage of current technological capabilities and make all unpublished opinions available to the public through electronic dissemination.

B. Citation to Unpublished Opinions Should Be Allowed When No Published Opinion Would Serve As Well

The behavior of legal participants and the resulting fairness considerations lead to the conclusion that unpublished opinions must be citable, at least in some circumstances. If unpublished opinions become generally available, then it makes little sense to prohibit all citation. As *572 already mentioned, [FN146] citation bans were intended to effectuate the selective publication plans. If, however, the premise underlying restrictive publication is faulty, then citation bans should be lifted.

Both evidence and common sense indicate that where potentially relevant legal information is available, attorneys will sift through that information for valuable nuggets. Even in the face of a rule barring citation, litigants can still make unacknowledged use of unpublished opinions. [FN147] Given these and other potential uses of unpublished opinions, litigants may thus find it worthwhile to spend time researching these opinions in order to gain an advantage. If we assume, as the evidence indicates, that such behavior will occur, then the cost savings imagined by the no-citation rules will be significantly diminished. Allowing citation does not, therefore, appear to add significantly to the cost burden the rule is designed to protect.

Allowing unpublished opinions to be cited also allows stare decisis to develop naturally. For instance, while the goal of publication plans is to limit publication to opinions with precedential value, this goal is not always realized. At the time a decision is handed down, judges cannot always be absolutely certain that the decision has no future value. Judges are not superhuman and cannot see the future. Accordingly, opinions that, when written, were thought not to contain novel facts or analysis may turn out to prove beneficial to the resolution of future cases. [FN148] When this is the case, the rules of court should not stifle the information available but rather should allow attorneys to access relevant information on the theory that a better-informed decision maker will produce better-reasoned decisions. Such relevant information is not limited to the actual language and reasoning of an unpublished opinion but also includes the credibility provided by citation--the fact that the analysis was given judicial recognition in a previous case.

This is not to say that unpublished opinions should be citable indiscriminately. The no-citation rules of the federal circuit courts seek to address the very valid concerns of cost and efficiency. Accordingly, it is appropriate for the courts to discourage the citation of unpublished opinions that are redundant or add little value to the resolution of the issue at hand. But where an unpublished opinion is relevant in a way that published opinions are not, citation should be allowed.

The Tenth Circuit limits the citation of unpublished opinions to instances where the opinion relates to a material issue that has not been *573 addressed in a published opinion. [FN149] This rule is too restrictive. For instance, a published opinion may have addressed the issue at hand but with a result contrary to that achieved in the unpublished opinion. Unless the published opinion has expressly overruled the unpublished opinion on that issue, there would appear to be an intra-circuit conflict that the court should address. Courts should not restrict citation in these circumstances.

Although a model rule should force attorneys to use careful judgment when choosing to cite to a given unpublished opinion, the rule should not be too restrictive. The better rule limits citation of unpublished opinions to instances where "no published opinion would serve as well." This rule is flexible enough to allow citation to an unpublished opinion in circumstances where a published opinion has addressed the same issue but with a different result. Allowing such citation would have a positive effect on the development of law within the circuit. It would allow opposing counsel to advance competing arguments as to which is the better legal analysis. Ultimately, the judge would be forced to decide between the competing analyses and articulate an opinion of the court.

C. Unpublished Opinions Should Be Given Persuasive Authority

The federal courts of appeals should allow unpublished opinions to be cited for their persuasive value only. Recall that a major reason for limiting publication is to further the goal of judicial efficiency. [FN150] A rule that allowed unpublished opinions to be used for persuasive value would promote this goal. Where an opinion is not believed to have future precedential or analytic value, the opinion is not published. Accordingly, judges spend less time crafting these opinions than those destined for publication. Judges may provide only a cursory review of the facts and may use less careful language than they otherwise would. The hopeful result of this condensed process is that judges are more efficient and are able to handle more cases. Furthermore, attorneys are unlikely to devote as much time to researching unpublished opinions. Because these opinions are less carefully crafted, the facts may be sparse and attorneys may not rely as heavily upon them, the model rule should provide for a middle ground, allowing citation to an unpublished opinion only for its persuasive value.

*574 Conclusion

The questions of whether and how publication of judicial opinions should be limited are complex. The answers to these questions involve the weighing of exceedingly important and contradictory policy considerations. Among these issues are cost and fairness concerns, judicial efficiency, judicial accountability and the role of stare decisis. In the face of these policy considerations is the stark truth that our courts are beleaguered by a colossal caseload. In the absence of more fundamental reforms designed to ameliorate this crush of cases, limited publication seems an inevitable palliative. Given the very overcrowded dockets of the federal court system, the goals served by limited publication are worth pursuing. But at what cost?

A rule of limited publication that allows for both electronic dissemination of, and citation to, all unpublished opinions for their persuasive value strikes the best balance. Such a system allows legal consumers to monitor judicial behavior, thereby promoting judicial accountability. Further, parties may cite an unpublished opinion for its persuasive value when no published opinion would serve as well. Allowing citation in these circumstances recognizes that judges cannot always foresee which opinions may have future value. At the same time, the fact that these opinions do not serve as binding precedent mitigates concerns about the less careful drafting of unpublished opinions as well as the concern that better-funded litigants will benefit because they can afford more research.

Ultimately, the analysis becomes a "balance of harms" test. Recent technology has lessened the "harms" to fairness in that the increased costs associated with more case law are less severe today than when cases were stored in the printed volume and research performed in the library. Further, allowing wide access to, and citation of, unpublished opinions advances important policy considerations. While the costs of digitally retrieving information may not yet be low enough to completely overcome fears about unfair advantage to wealthier litigants, we are moving in that direction. Case law can now be accessed over the Internet. [FN151] Wider case law availability should lead to lower costs of access. [FN152] Lower costs and wider access would go a long way toward easing the fairness concerns that have caused the courts to retain their *575 limited publication/no-citation plans. Since many unpublished opinions do indeed contain useful information and their public availability does serve important interests, courts should re-examine their selective publication and no-citation rules in light of technology's influence on legal information.

Copyright Copr. 1996 California Law Review

FN152. It should also be noted that proposals in a number of states call for moving towards a vendor-neutral citation format. See supra note 74. Such a move may work to increase competition in the digital publication market. If increased competition led to lower access costs, the fairness concerns would be further eased.