Restructuring the Federal Circuit

By Jeremy W. Bock*

A pdf version of this article may be downloaded here.

Introduction

Patent-related issues are becoming ever more salient in the national economy, as reflected in the intense interest in patent policy exhibited by all three branches of government in recent years.[1] No wonder, then, that the Chief Judge of the United States Court of Appeals for the Federal Circuit—which has exclusive jurisdiction over patent appeals[2]—recently declared “We are the most important court in the United States.”[3] While not everyone may agree with that sentiment,[4] the Federal Circuit has, nevertheless, been one of the more closely-scrutinized appellate courts by virtue of its specialized jurisdiction[5] and its mandate to create national uniformity in the adjudication of patent disputes.[6]

Of particular concern to interested observers of the Federal Circuit are certain persistent problems identified in the academic literature, such as: that its outcomes are strongly panel-dependent on certain issues;[7] that it has formalist tendencies;[8] that its case law leads to indeterminate results;[9] and that it enforces certain doctrines too strictly,[10] while being too lax on others.[11] Criticisms of the Federal Circuit are contained in a vast body of literature, and are also aired at numerous patent law-related conferences, often in the presence of one or more Federal Circuit judges in attendance. And yet, many of the complaints seemingly endure (e.g., the Federal Circuit’s conflicting claim construction methodologies lead to panel-dependent outcomes[12]), and change is often substantially delayed (e.g., certain problematic rules associated with willfulness determinations were overruled after twenty years[13]), if not denied (e.g., the Federal Circuit decided to reconsider the de novo standard of review for claim construction[14]—after fifteen years of complaints from not only commentators,[15] but also district judges,[16] and its own members[17]—only to reaffirm it[18]).

In light of these criticisms, one question that arises is whether the Federal Circuit has difficulty reconsidering and correcting the suboptimalities in its case law in a timely manner, whether by clarifying, limiting, reconciling, or overruling precedents that have become problematic.[19] Because it is effectively the court of last resort in patent cases, given the rarity of Supreme Court review,[20] self-correction by the Federal Circuit is critical. This, coupled with the extreme improbability of timely Congressional action,[21] creates a situation where the primary responsibility “for assuring that gaps are filled, uncertainties resolved, and stupidities corrected”[22] in patent law falls squarely on the Federal Circuit. Otherwise, without periodic self-correction, Federal Circuit precedents may ossify in a suboptimal state, which, as John Golden has observed, may be the real danger posed by a centralized appellate scheme for patent cases.[23]

Compared to the regional circuits, the Federal Circuit likely requires a faster, more robust mechanism for making corrections to a body of precedent, for two reasons. First, patent law must be able to keep up with advances in technology.[24] Second, as the exclusive appellate venue for patent cases, the Federal Circuit does not experience the type of corrective case law “percolation” that occurs among the regional circuits.[25] It is possible then, that suboptimal case law might endure more readily at the Federal Circuit, when, at the same time, the need to make timely adjustments to precedent may be more pressing in light of the issues raised by new technologies.

While a variety of solutions have been proposed in the literature that seek to compensate for or otherwise mitigate the Federal Circuit’s shortcomings,[26] scant attention has been paid to the cognitive and situational elements that define the internal adjudicatory environment within that court. As such, this Article undertakes an exploration of the behavioral factors that may underlie the various complaints about the Federal Circuit, with a specific focus on the cognitive and situational elements that allow suboptimal precedents to be generated and maintained. In doing so, this Article draws lessons not only from the literature on judicial behavior, but also from cognitive psychology and management science—in particular, organizational behavior[27]—in evaluating the difficulty of self-correction. As described in greater detail infra,[28] this analytical path suggests that a potential solution may lie in changing the organizational structure of the Federal Circuit to mitigate the behavioral elements that may negatively affect the court’s ability to take timely actions to refine or otherwise repair suboptimal case law.[29]

When the complaints about the Federal Circuit are viewed through a cognitive and situational lens, two possibilities emerge. First, with certain precedents, some judges may simply fail to recognize the need to take corrective action. As discussed in greater detail infra,[30] a Federal Circuit judge’s inability to recognize the existence of a problem could be an artifact of his or her expertise. Specifically, a phenomenon known in the cognitive psychology literature as the “curse of expertise”[31] may prevent experts (i.e., Federal Circuit judges)[32] from properly weighing criticism from others, and may also impair their ability to accurately assess the difficulties encountered by non-experts (e.g., district judges).[33] Second, with other precedents, some Federal Circuit judges may be fully aware that problems do exist for which correction is necessary, but the court, as an institution, is unable to undertake timely remedial action. This second problem, as explored infra,[34] is known in the management science literature as the “knowing-doing gap,”[35] which may arise from the situational dynamics among the Federal Circuit judges.

Taken together, the curse of expertise and the knowing-doing gap may impair the Federal Circuit’s ability to address the suboptimalities in its case law in a timely manner because the former may prevent the court from realizing that a problem exists, while the latter may hinder the court from taking action on the problems that it is aware of.

This Article contributes to the literature on the institutional analysis of the Federal Circuit in two ways: First, it combines lessons from the literature on judicial behavior, cognitive psychology, and management science to evaluate possible behavioral explanations for how and why the Federal Circuit generates or maintains suboptimal precedent. Second, based on this behavioral analysis, a proposal is introduced that may improve the Federal Circuit’s ability to identify and self-correct suboptimal precedents: staffing the court with district judges who serve staggered terms of limited duration.

In essence, by changing the organizational structure of the Federal Circuit, it may be possible to compensate for, or at least mitigate, some of the cognitive biases and situational influences[36] within the court’s deliberative environment that may inhibit timely self-correction. This proposal focuses on a beneficial effect of regular changes in personnel that has not been fully appreciated in the literature relating to judicial term limits and rotations, which tends to concentrate on the Supreme Court[37] and issues relating to democratic accountability, the politicization of nominations, and mental decrepitude,[38] rather than the ability of an intermediate appellate tribunal to self-correct.

This Article proceeds in four parts. Part I describes the Federal Circuit’s internal mechanisms for self-correction, and also summarizes several existing proposals in the literature that endeavor to remedy or compensate for the Federal Circuit’s deficiencies. Part II identifies potential cognitive and situational barriers to timely self-correction at the Federal Circuit—namely, the curse of expertise and the knowing-doing gap. Part III analyzes how the cognitive and situational barriers identified in Part II may be overcome by staffing the Federal Circuit with district judges who serve staggered terms of limited duration. Part IV addresses various objections and concerns about the proposal, and is followed by a brief conclusion.

I. The Current State

A. Internal Mechanisms for Quality Control

As an institution, the Federal Circuit has several internal “quality control” mechanisms for precedential opinions. They generally fall into two categories: before issuance and after issuance.

1. Before Issuance: Review of Precedential Opinions

Shortly after the Federal Circuit was formed, its first Chief Judge, Howard Markey,[39] and Judge Giles Rich, one of the drafters of the 1952 Patent Act,[40] touted the adoption of two procedures designed to help the newly-formed appeals court fulfill its mandate of bringing uniformity to patent law. First, every draft precedential opinion would be circulated to the entire court for several working days in order to allow the non-panel judges to weigh in and provide comments prior to issuance.[41] (Presently, the review period is ten working days.[42]) Second, the Federal Circuit’s internal administrative departments would include an office of the “Senior Technical Assistant” (STA),[43] whose staff members are charged with analyzing draft precedential opinions during the review period and circulating memos to the entire court that highlight potential conflicts with existing case law.[44]

While these two mechanisms for pre-issuance correction of precedential opinions have endured throughout the Federal Circuit’s existence, the persistent complaints about the court suggest that Judges Markey and Rich might have overestimated their effectiveness. The utility of circulating draft precedential opinions to the entire court is dependent on the willingness of the non-panel judges to closely monitor them, provide substantive comments, and, if necessary, issue “hold sheets” that prevent their issuance until substantive concerns are addressed.[45] At the same time, the receptiveness of the authoring judge and the other panel members to suggestions from the rest of the court may also be a significant factor. However, persistent concerns about panel-dependent outcomes[46] and the existence of divergent lines of precedent[47] suggest that the internal circulation of draft opinions may be a weak mechanism for self-correction.

As for the STA’s memos analyzing draft opinions for conflicting precedent, they are merely advisory, and, for that reason, it is possible that some Federal Circuit judges disregard them.[48] Notably, the STA’s conflicts check procedure has actually been scaled back in recent years: when the STA’s office was first established at the Federal Circuit, it reviewed every draft precedential opinion for potential conflicts prior to issuance;[49] however, on July 7, 2010, the Federal Circuit’s Internal Operating Procedures (IOPs) were changed so that the STA reviews a draft precedential opinion for conflicts only if requested.[50] The Federal Circuit’s decision to scale back the original role of the STA’s office is puzzling, if not troubling, especially when divergent lines of precedent still persist—and, in some instances, have emerged after July 7, 2010, on issues such as patentable subject matter[51]—and decisional disagreement at the court is becoming more severe.[52]

2. After Issuance: The En Banc Process

Once issued, precedential panel decisions may be overruled only by an en banc court.[53] However, the en banc process is seldom invoked: according to an empirical study by Christopher Cotropia, the Federal Circuit’s en banc rate is relatively low.[54] This is unsurprising as the en banc process is viewed by judges as a time-consuming, labor-intensive endeavor[55] whose uncertain outcomes might not be worth the cost of disturbing the collegial atmosphere that courts strive to maintain.[56] As a result, an extended period of time may elapse before the “right” case appears for which a majority of judges would agree is worth the hassle of en banc consideration. In some instances, a change in the composition of the court may be necessary in order for an issue to be ever considered en banc.[57] And when en banc review finally does occur, there is no guarantee that the outcome will necessarily improve the situation: the en banc court might simply reaffirm the status quo,[58] take the precedent in a more problematic direction,[59] or create further uncertainty by issuing a highly fractured decision with no majority opinion.[60]

B. Existing Proposals in the Literature

The literature offers a variety of proposals for remedying or compensating for the deficiencies in the Federal Circuit’s case law and its decision-making process.

One popular proposal is to develop patent law expertise at the trial level[61] such as by establishing specialized trial courts[62] or by changing the venue rules so as to concentrate the filing of patent cases to certain districts.[63] This proposal has largely come to fruition in the form of the “Patent Pilot Program,”[64] which helps district judges cultivate patent law expertise through the reassignment of patent cases from judges who seek to avoid them to judges who are interested in hearing more of them.[65] However, empirical support for the expected benefits[66] of enhancing trial judge expertise is, at best, mixed.[67] Furthermore, specialization at both the trial and appellate levels could exacerbate the Federal Circuit’s “exceptionalist” approach to patent law.[68]

Rather than focusing on the trial courts, some commentators have questioned the unitary appellate regime for patent cases. Craig Nard and John Duffy propose expanding the number of circuit courts that hear patent appeals, on the theory that “a polycentric, competitive appellate structure” may facilitate doctrinal development through incremental innovation and experimentation, as well as provide clearer signals to the Supreme Court to intervene.[69] Judge Diane Wood of the U.S. Court of Appeals for the Seventh Circuit has expressed support for a similar proposal.[70] As some commentators have observed, however, it is unclear whether inter-circuit percolation is likely to provide benefits adequate to offset the loss of uniformity,[71] especially in the event the other circuit courts find it expedient to defer to the expertise of the Federal Circuit by adopting its case law.[72] In addition, with the enactment of the Leahy-Smith America Invents Act,[73] Congress has reaffirmed its commitment to a unitary appellate regime in patent cases by abrogating Holmes Group, Inc. v. Vornado Air Circulation Systems, Inc.[74] and channeling all patent-related appeals to the Federal Circuit, including those appeals raising a patent issue that was introduced only through a counterclaim.[75]

Others have considered whether the scope of the Federal Circuit’s jurisdiction might be a source of its dysfunction. Paul Gugliuzza, for example, proposes modifying the Federal Circuit’s jurisdiction so that it has a mix of cases that is closer to that of the regional circuits,[76] under the theory that Federal Circuit judges who are exposed to a more generalized docket would be “more policy conscious, less formalist, and, ideally, more responsive to the different innovation dynamics present in different industries.”[77] This proposal presupposes that the adjudicatory style of individual Federal Circuit judges may change if their knowledge of generalist legal principles is enhanced. As will be discussed later, a substantial part of the problem with the Federal Circuit may not only be a lack of certain knowledge, but also the lack of the will to act on that knowledge.[78]

A common limitation in the aforementioned proposals is that they may not be adequate to counteract the primary pathology underlying the complaints about the Federal Circuit, which, as identified by John Golden, is the ossification of precedent[79] resulting in “suboptimal legal equilibria.”[80] Golden’s solution to the ossification problem requires the Supreme Court to assume the role of “prime percolator,” which periodically grants review to disturb those suboptimal Federal Circuit precedents that have “frozen legal doctrine either too quickly or for too long.”[81] In addition, Golden suggests that the Federal Circuit itself can further promote percolation by not writing or reading panel opinions unnecessarily broadly, and granting en banc review when precedent has reached a “suboptimal doctrinal equilibrium.”[82]

There are several limitations to Golden’s proposals. First, given the overall rarity of certiorari grants,[83] and the limitations of Federal Circuit dissents as signaling devices,[84] the Supreme Court’s execution of its percolation function may not be timely and frequent enough to materially improve the performance of the Federal Circuit beyond its current state—especially where the Federal Circuit views itself as the “expert,” and, as a result, may be unreceptive to suggestions from non-experts, even those located in a superior position in the judicial hierarchy.[85] Second, it is unclear how the tendency of some Federal Circuit judges to write (or interpret) broadly may be effectively restrained. A “maximalist”-style[86] of opinion-writing that makes broad, sweeping pronouncements could be viewed by some Federal Circuit judges as wholly appropriate, if not obligatory, for a court whose primary mission is to provide uniform guidance on patent law.[87] Furthermore, by invoking judicial efficiency concerns, the Federal Circuit has, on occasion, decided issues beyond those needed to dispose of the specific case at hand, such as construing claim terms that are not necessary to the judgment.[88] Finally, as discussed previously, going en banc is a labor-intensive and unpredictable process,[89] such that case law may be stuck in a suboptimal legal equilibrium until the “right” case comes along or if the composition of the court changes.[90]

At a high level, the above-mentioned proposals highlight different approaches to improving the operation of the Federal Circuit. What is missing, however, is an approach that focuses on the behavioral elements that may inhibit timely reconsideration of suboptimal precedents. To this end, the next section explores some of the behavioral elements that may influence how Federal Circuit case law is produced.

II. The Cognitive and Situational Barriers to Self-Correction

This section introduces two theories directed to the behavioral elements that may hinder self-correction at the Federal Circuit: (1) the “curse of expertise,” which may impair the ability of Federal Circuit judges to recognize potential problems in existing case law; and (2) the “knowing-doing gap,” which may contribute to the court’s failure to take corrective action despite some judges recognizing the existence of a problem.

A. The Curse of Expertise

In general, the patent law expertise of Federal Circuit judges is viewed as a positive trait.[91] Although expertise may enhance one’s ability to analyze problems involving complex subject matter—such as patent law—experts are susceptible to certain systematic errors and biases.[92] As described in the cognitive psychology literature, the “curse of expertise” is a term that captures the cognitive pitfalls to which experts are particularly susceptible,[93] such as underestimating the difficulties of non-experts[94] and resisting correction.[95]

For the purpose of analyzing the curse of expertise in the context of the Federal Circuit, the “experts” are the Federal Circuit judges, and the “non-experts” are the generalist district judges. The practitioners, who are also active consumers of the Federal Circuit’s decisional output, include a wide range of individuals, from experts (e.g., specialists who have practiced patent law exclusively for decades) to non-experts (e.g., general litigators who have recently become involved in a patent case). However, the discussion of the curse of expertise in this Article will focus primarily on the expert/non-expert comparison between Federal Circuit judges and district judges because both are engaged in the same function—namely, the adjudication of patent disputes.

1. Underestimation of Difficulties of Non-Experts

According to experiments conducted by Pamela Hinds, experts may be prone to underestimating the difficulties encountered by non-experts who attempt to perform tasks within the expert’s field.[96] At the Federal Circuit, the curse of expertise is at work when it skews the Federal Circuit judges’ perceptions regarding the clarity or the soundness of the court’s precedents, such that they may fail to recognize when existing case law may warrant reconsideration. That is, the expertise of the Federal Circuit judges could potentially interfere with their ability to recognize when an error in the judgment below may be primarily attributable to vague, conflicting, or unworkable case law, as opposed to the district judge’s failure to apply otherwise sound precedent.

By way of illustration, consider the following pair of claim construction canons: the claims must be read in light of the specification;[97] however, limitations may not be imported therefrom.[98] District judges have struggled with these canons, which govern the use of the specification in claim construction.[99] Although the Federal Circuit has acknowledged that a fine line separates these canons,[100] it nevertheless believes that they may be reliably applied[101]—even though a schism exists in the court’s claim construction methodology, which places different emphases on the specification vis-‡-vis the claims.[102] This schism, which has endured despite the en banc restatement of claim construction principles in Phillips,[103] has made it difficult for district courts to correctly apply these complementary canons because the “correct” methodology is panel-dependent. Either the canons need to be updated or the schism needs to be resolved, but with the exception of Judge Kimberly Moore (and possibly Chief Judge Randall Rader),[104] the Federal Circuit does not appear to view the current state of the law on claim construction methodology to be problematic enough to warrant en banc consideration.

2. Resistance to Correction

Hinds’s experiments also suggest that experts, when compared to non-experts, may be unusually resistant to correcting or changing their positions even when presented with “debiasing” information that could help increase the accuracy of their analysis.[105] This comports with the general understanding that experts are “often wrong but rarely in doubt.”[106] It is possible then, that Federal Circuit judges, by virtue of their expertise in patent law, may be prone to approaching their work with a level of overconfidence that may render them relatively unreceptive to reconsidering their analysis in the face of debiasing information.[107]

One potential indication of the Federal Circuit’s resistance to debiasing may be its decision to scale back the generation of the STA’s reports of potential conflicts in draft precedential opinions.[108] Another indication may be the court’s perceived reluctance to give substantive consideration to relevant scholarship.[109] But perhaps the most profound indication that the Federal Circuit may be unreceptive to debiasing is the seemingly weak corrective influence of the Supreme Court.[110] For example, some Federal Circuit judges appear to be resisting the Supreme Court’s directive to apply the patentable subject matter requirement more rigorously.[111] Indeed, the rapid succession of patentable subject matter cases heard by the Supreme Court in recent years[112] suggests that it might be engaged in a “battle of wills” with the Federal Circuit. It is possible that some Federal Circuit judges may be reluctant to accord much weight to the guidance provided on issues within their realm of expertise by a non-expert, yet hierarchically-superior, tribunal.[113] This, along with the sparse and episodic nature of Supreme Court review, likely creates a situation where the Federal Circuit is potentially operating without any meaningful moderating influence.

3. Aggravating Factors

Research on cognitive heuristics suggests that the key factors contributing to the curse of expertise include anchoring effects[114]—i.e., the heavy reliance on initial information or impressions[115]—as well as the “availability heuristic”[116]—i.e., the tendency to rely on information that readily comes to mind.[117] This suggests that the adverse effects of the curse of expertise may be aggravated by the selection effects introduced by the appeals process: the mix of patent cases and issues that reach the Federal Circuit may not be representative of what the district judges encounter. If the Federal Circuit judges over- or under-estimate the severity of the problems associated with certain doctrines, they may consequently over- or under-correct them. This skew in perception may be the most severe for those issues that come before the Federal Circuit infrequently but arise often at the district court level (and vice versa), such as discovery issues and evidentiary rulings, for which seasonable appellate review might be available only through discretionary means.[118] Deficient or suboptimal Federal Circuit precedents affecting issues that arise frequently at the district court level, but which rarely make it to the appellate level, not only will have a major adverse impact, but also will endure because the opportunities to develop and refine those precedents will be rare. For example, it took over twenty years for the “adverse inference” and “affirmative duty” rules in Underwater Devices, Inc. v. Morrison-Knudsen Co.,[119] which presented thorny privilege and waiver issues for accused infringers, to be overruled in Knorr-Bremse Systeme Fuer Nutzfahrzeuge GmbH v. Dana Corp.[120] and In re Seagate Tech., LLC,[121] respectively.

* * *

In weighing the relative benefits of expertise against the drawbacks discussed above, it is worth noting that the Federal Circuit, by virtue of its expertise, may be quite comfortable with making broad pronouncements on a variety of issues—which, as mentioned earlier, is something that the court should refrain from doing to avoid lock-in of suboptimal precedents.[122] At the same time, if a given pronouncement proves to be overly broad or otherwise problematic, the cognitive pitfalls associated with expertise could make it difficult for the court to realize that there is a problem. The tradeoff then, is whether an expert court that might be difficult to correct is preferable to a non-expert court that might be easier to correct. As will be discussed in later sections, there are benefits associated with staffing the Federal Circuit with non-experts that could potentially tip the balance toward the latter option.

B. The Knowing-Doing Gap

The cognitive elements discussed in the previous section are not the only considerations relevant to evaluating the ability of a group of individuals to self-correct. The situational dynamics of the organization, as a whole, may be just as salient. Organizational inaction despite recognition of the existence of a problem is known in the management science literature as the “knowing-doing gap.”[123] A canonical example is the disturbingly low rate of hand-washing by healthcare professionals who are fully aware that hand hygiene is essential for minimizing hospital-acquired infections.[124] Another common example is a corporation hiring expensive management consultants (sometimes on multiple occasions) to provide recommendations on improving operations or solving a problem, but failing to implement their recommendations.[125]

The concept of the “knowing-doing gap” may be useful in the judicial decision-making context as well, as a means for evaluating how suboptimal precedents may be generated and ultimately persist. Although persistence can arise naturally out of stare decisis and by operation of the Federal Circuit’s IOPs, this Article focuses on the behavioral elements beyond those mechanisms. That is, why does the Federal Circuit have difficulty using the tools available to it to prevent the issuance of, or to timely repair, case law that has been identified as problematic not only by observers outside the court, but also by its own judges?[126] Such institutional inertia is the “knowing-doing gap.” Jeffrey Pfeffer and Robert Sutton suggest that the existence of a “knowing-doing gap” is dependent primarily on organizational considerations, rather than the personal characteristics of individual actors.[127] As such, a solution for overcoming the “knowing-doing gap” could potentially lie in modifying the institutional structure of the Federal Circuit so that it becomes less susceptible to inertia—regardless of who is serving on the court at any given time. It may be helpful, then, to evaluate the internal dynamics of the appellate decision-making environment to determine how the Federal Circuit’s structure could be modified to inhibit institutional inertia from setting in when it becomes advisable to take remedial action.

In this regard, there are several situational considerations within the appellate decision-making environment that may be worth exploring. For example, a judge may decide to stand firm on an issue because he wants to act consistently with his prior positions. Another reason for standing firm is that the judge has a deep-seated conviction regarding how the case should be decided. At the same time, a judge may desire to maintain a collegial working environment (or détente). And finally, a judge may endeavor to defer laborious or otherwise costly tasks if there is no urgency. At first blush, some of these considerations might appear mutually exclusive, in particular, consistency and conviction versus collegiality. However, they may exist as complementary considerations among different judges on different issues as enabled by the third consideration—the lack of urgency.[128] As explained in greater detail below, these situational considerations may be classified as having a structural origin because they appear to be byproducts of the particular manner in which the Federal Circuit is staffed—namely, lifetime appointments.

1. Consistency and Conviction

Suboptimal precedents could emerge or be maintained when judges refuse to reconsider their prior positions. This refusal may arise from a desire to maintain consistency, a deep-seated conviction on an issue, or a combination of both.

In the environment of an appellate court, the desire to maintain personal consistency[129] could contribute to the perpetuation of problematic case law, particularly when correcting it may require one or more judges to take positions inconsistent with their prior votes or opinions, which could expose them to the risk of losing face.[130] If the membership of a court has split into opposing camps, the resolve of individual judges to seek vindication and to stay consistent may be heightened. At the same time, it may take less work for a judge to default to his or her prior position, because crafting an opinion that justifies a change of position on principled grounds may be labor-intensive.[131] As such, for judges who wish to maximize the “leisure” aspect of the judicial utility function,[132] staying consistent may be an attractive option. So long as there is no urgency,[133] a judge may decide to defer the investment of the necessary mental and emotional energy required to properly reconsider a prior ruling.

For other judges, the refusal to reconsider may not be driven by the need to appear consistent or to avoid extra work, but rather by a deep-seated conviction on a specific issue. In these instances, the judge is highly unlikely to yield or be convinced to deviate from his or her position, and he or she may be more than willing to undertake the additional work required to write a dissent, if necessary. It is possible that such principled persistence could be more problematic, and at the same time more severe, at the Federal Circuit than in the regional circuits. This is because some Federal Circuit judges might view the court as having an implicit policy-oriented “mission,” i.e., to protect and encourage innovation in the United States,[134] as opposed to simply deciding cases. As such, some judges may develop strong beliefs as to what the law ought to be on certain issues in order to further the court’s “mission.” In addition, as Cotropia suggests, the fact that Federal Circuit judges are repeatedly exposed to the same issues might render them prone to sharp disagreement over nuances.[135]

When viewed together, the desire for consistency and/or to vote one’s conviction likely contribute to the formation of camps within the Federal Circuit, which could drive panel-dependent outcomes. (At a high level, panel-dependence could also be indicative of a more fundamental problem with the case law: The precedents associated with a given doctrine may be so indeterminate[136] that they comfortably accommodate inconsistent approaches favored by different camps of judges.) Although the desire to be consistent and to remain steadfast on an issue can, under certain circumstances, have beneficial effects in contributing to the stability of case law, they may become problematic when they impede the timely reconsideration of defective or otherwise suboptimal precedents.

2. The Need to Maintain Collegiality or Détente

Initially, the desire for consistency and conviction-voting might provide the most intuitive explanations for the Federal Circuit’s difficulty with self-correction. However, they provide only part of the story.

The collective need of a group of judges to maintain collegiality or détente[137] on a daily basis,[138] and to avoid fruitless battles with colleagues who might serve with them for decades,[139] may, at times, take priority over correcting suboptimal case law, which might involve resurrecting disputes over which the judges have “agreed to disagree.” Although a certain degree of collegiality is indispensable to effective decision-making,[140] the risk of “settling” behavior nevertheless exists[141]: when a group of judges serves together long enough, its members will become intimately familiar with each others’ views to an extent that would allow them to reach a state of equilibrium where the judges fall into a predictable pattern in their voting and opinion-writing.[142] This predictability, in turn, could lead to the ossification of suboptimal precedent.[143] And because the en banc process is fraught with drama, high cost, and uncertainty,[144] individual panels are left with essentially two options for handling precedents they find troublesome: distinguish them or ignore them.[145] As a result, over time, divergent lines of precedent could emerge that may prove increasingly difficult to reconcile.

The willingness of appellate judges to not only live with, but also allow the creation and maintenance of, suboptimal precedents and divergent case law[146] may be explained by certain behaviors characterized by Judge Richard Posner as “going along” voting and “live and let live” opinion-joining.[147] “Going along” voting in a panel occurs when the judges who are relatively indifferent about the outcome cast their votes with the member having the strongest views.[148] Otherwise, if one or both of the indifferent judges were to vote differently from the opinionated judge, they will need to respond to the spirited arguments of the opinionated judge in either a majority opinion or a dissent.[149] To an indifferent judge, “going along” is less costly than devoting resources to an issue he may not feel strongly about.[150] Relatedly, “live and let live” opinion-joining occurs when a judge joins an opinion that contains remarks he disagrees with, but which are perceived by the joining judge to be dictum.[151] A judge may rationally view the effort to eliminate dictum with which he disagrees as not worth the hassle and potential clashes with the authoring judge.[152]

The “going-along” voting and “live and let live” opinion-joining described by Judge Posner constitute a set of behaviors that may be characterized more generally as “collegial concurrence,” which Cass Sunstein and others define as a form of deference to one’s colleagues[153] that arises out of a sense of realism,[154] where an individual judge rationally views any attempt to “correct” the other members of the panel as costly and futile. Collegial concurrence may also be at work when a draft precedential opinion setting forth a problematic rule is circulated to the entire court,[155] and the non-panel judges acquiesce in its issuance without substantive revisions.[156]

The risk of perpetuating and creating problematic case law also exists if there is a split panel, particularly when the judges belong to opposing camps. In such cases, suboptimal precedents may result when panel members unduly focus on vindicating their respective positions or when lingering resentments or jealousies surface,[157] such that the collective will may not exist to undertake a labor-intensive analysis to reconcile competing considerations and craft a workable rule. Furthermore, when the dissenting judge has been “written off” by the majority, which may often happen to chronic dissenters,[158] it is possible that the majority may take more extreme positions than if it were endeavoring to convert the dissenter into a joiner.

* * *

It may appear counterintuitive that the concerns relating to consistency/conviction and the need for collegiality may not only co-exist within a single court, but also could jointly contribute to the generation and maintenance of suboptimal precedent. Indeed, consistency/conviction and collegiality can be influences that may be present concurrently on the same panel because they may arise for different issues for different judges. Listed below are the possible configurations for a three-judge panel (for en banc panels, additional configurations may exist):

(a) All three judges are relatively indifferent regarding the issues on appeal: Collegial concurrence will likely be the dominant influence on the manner in which the opinion is crafted.

(b) One judge has strong opinions on an issue (because of consistency/conviction), and the other two judges are relatively indifferent: The indifferent judges are likely to “go along” with the opinionated judge.

(c) Two judges have strong opinions that are incompatible with each other (because of consistency/conviction), and the remaining judge is indifferent: The indifferent judge will likely “go along” with one of the two opinionated judges, and the opinionated judge who is in the minority is likely to dissent.

(d) All three judges have strong opinions that are incompatible with each other (because of consistency/conviction): Separate opinions are likely, and there may or may not be a majority opinion.

3. Lack of Urgency

It is not often that a critical mass of appellate judges will perceive an urgent need to correct some defective precedent and willingly “rock the boat”[159] in undertaking the laborious en banc process.[160] A sense of urgency arises infrequently because the consequences of inaction are unlikely to be concrete and immediate for a permanent appellate judge. Rather, the impact of suboptimal decisions from the Federal Circuit is most immediate on the practitioners, who may need to update their case strategies, and on the district judges, who must apply the newly-minted precedents in the first instance, under the threat of reversal. For Federal Circuit judges, the impact of problematic case law on their day-to-day work is considerably attenuated, as they can distinguish it in subsequent cases or effectively disregard it[161] largely without fear of reversal, as review by the en banc court[162] or the Supreme Court is rare.[163] Accordingly, the more immediate concerns, such as appearing consistent, saving face, conviction-voting, maintaining collegiality or détente, and avoiding fruitless battles with colleagues who may serve with them for an indefinite period of time, may take precedence over engaging in a potentially costly analysis. Overcoming the “knowing-doing gap,” then, may require changing the adjudicatory environment in a way that decreases the salience of these concerns.

III. Surmounting the Barriers to Self-Correction

To overcome both the curse of expertise and the knowing-doing gap, this Article proposes staffing the Federal Circuit with a rotating group of district judges who serve staggered terms of limited duration. A suitable term served by each district judge could be two[164] to four years, which could help guard against the development of the blind spots and inertia associated with the accumulation of expertise and prolonged tenure, respectively,[165] as well as reduce the likelihood of capture by special interests.[166]

This proposal resembles a common arrangement within the federal judiciary for staffing tribunals that exercise jurisdiction over specialized subject matter. For example, the Foreign Intelligence Surveillance Court, which reviews applications for orders authorizing electronic surveillance within the United States to obtain foreign intelligence information,[167] is staffed by district judges[168] who serve non-renewable, staggered terms of up to seven years.[169] The Judicial Panel on Multidistrict Litigation (JPML), which is empowered to transfer to a single district multiple civil cases whose pretrial proceedings may benefit from consolidation and coordination,[170] is staffed by a mix of district judges and circuit judges.[171] The Bankruptcy Appellate Panels (BAPs) that exist in some circuits[172] are staffed by bankruptcy trial judges who are appointed for limited terms[173] to hear bankruptcy appeals in three-judge panels.[174] As such, the federal judiciary has a variety of existing models as well as the requisite logistical experience for successfully implementing this proposal. In a similar vein, judicial rotation has been suggested as a way of staffing a potential specialized Article III appeals court for immigration, in which district judges and circuit judges serve two-year terms.[175]

Moreover, given that the Federal Circuit, in some respects, behaves not unlike an administrative agency that promulgates substantive rules,[176] it may be appropriate to change its composition regularly, just like the other government agencies involved in the development of patent policy. For example, the U.S. Patent & Trademark Office and the Department of Justice are each headed by political appointees who typically change along with the presidential administration,[177] and the Federal Trade Commission and the International Trade Commission are each led by commissioners who serve staggered terms of limited duration.[178]

A. Combating the Curse of Expertise with District Judges

As previously discussed, the curse of expertise creates a blind spot for the Federal Circuit regarding its perception of the soundness of its precedents: it may render the court prone to misjudging the difficulties encountered by non-expert district judges in applying Federal Circuit case law. In addition, it may render the “expert” Federal Circuit judges resistant to considering debiasing information that may help them recognize potential problems.[179]

One way of combating the curse of expertise at the Federal Circuit may be to replace the experts, i.e., the permanent Federal Circuit judges, with non-experts, i.e., district judges. More specifically, the Federal Circuit could be staffed with district judges who have handled a sufficient number of patent cases so as to have developed a sense of which precedents might be suboptimal, and how new or modified precedents might affect the quality of adjudication. In selecting the district judges to serve on the Federal Circuit, the experience threshold may be based on a variety of metrics such as the number of claim construction orders issued or the average number of patent cases handled per year.

That district judges with patent experience may be particularly suitable for the Federal Circuit is suggested by Hinds’s experimental studies demonstrating that those with an intermediate level of knowledge may outperform both experts and novices in anticipating difficulties faced by novices in completing a task.[180] An expert whose learning experience is a distant memory may not be able to recall his initial difficulty with the task as readily as someone who has learned it more recently,[181] while a novice may not have an adequate understanding of the task to make accurate predictions about the behavior of other novices.[182] Applying these lessons to patent case adjudication, it is possible that district judges who have handled multiple patent cases (i.e., the intermediate users of patent case law) may be better than either permanent Federal Circuit judges (i.e., the experts) or district judges who have little or no experience with patent cases (i.e., the novices) at identifying suboptimal precedents, particularly those that have proven difficult for district judges to apply reliably.

Hinds’s experiments also suggest that those with an intermediate level of knowledge may be more receptive than experts to debiasing information that could help improve the quality of their decision-making.[183] It is likely then, that compared to the current version of the Federal Circuit that is staffed with “expert” judges, a version of the court that is staffed with non-expert, yet experienced, district judges might give greater consideration to the conflicts memos prepared by the STA,[184] the relevant academic literature,[185] the views of expert agencies such as the Federal Trade Commission,[186] analogous case law from the regional circuits, and feedback from other district judges and practitioners. In addition, experienced district judges may be less influenced by the appellate selection effects that would otherwise provide an inaccurate picture of the relative frequency and severity of certain problems at the district court level.[187]

Furthermore, those with an intermediate level of knowledge, i.e., the patent-experienced district judges, may have a greater ability to craft workable precedents than either experts or novices. In another experimental study, Pamela Hinds, along with Michael Patterson and Jeffrey Pfeffer, found that experts tend to use more abstract concepts when imparting specialized, technical information to novices, while beginners tend to use more concrete statements.[188] And while novices instructed by experts demonstrated a greater ability to transfer their knowledge to different, analogous tasks,[189] the novices instructed by beginners learned to complete a specific task more effectively.[190] The results of this experimental study suggest that the pedagogically-optimal mix might be achieved by an individual whose skill level falls in between that of an expert and a beginner, and who, as a result, is more likely to provide an appropriate mix of abstract and concrete guidance. It is possible, then, that opinions authored by patent-experienced district judges—who are neither experts nor beginners—may be more amenable to reliable application at the trial level than those authored by permanent “expert” appellate judges or by district judges who have little substantive experience with patent cases. This is particularly important for patent appeals involving issues (e.g., discovery) that are frequently in contention at the district court level, but are infrequently reviewed on the merits at the appellate level, such that it is critical to get such cases “right” whenever they reach the Federal Circuit because there may be limited opportunities for correcting such precedents in the future.[191]

The proposed arrangement whereby district judges craft appellate precedents—that they will later follow under the threat of reversal—could be viewed as a quality-control measure that bears a conceptual resemblance to a software development process known as “eating one’s own dog food” or “dogfooding,” where software developers use internally the products they are developing in order to improve their ability to test and debug them.[192] Examples include Google’s employees internally using AndroidTM products before making them available to the public,[193] and Microsoft internally using the WindowsÆ operating system.[194] This practice tightens and strengthens the software development feedback loop between the developers and the end users because they include the same people. Likewise, the substitution of permanent Federal Circuit judges with district judges could similarly strengthen the “feedback loop” between the appellate and trial levels by having one of the primary consumers of Federal Circuit case law (i.e., district judges) contribute directly to its creation and revision. More generally, district judges who have struggled to apply Federal Circuit precedents may have a better sense than the current group of permanent Federal Circuit judges (most of whom do not have any experience as trial judges)[195] of how existing precedents should be clarified, modified, limited or overruled so that generalist district judges, as well as litigants, may reliably apply them in a manner that improves the overall quality of patent case adjudication.

B. Combating the Knowing-Doing Gap with Rotations

Although district judges may help mitigate the problems arising from the curse of expertise, permanently elevating individual district judges to the Federal Circuit is an incomplete solution. Rather, the district judges should serve staggered terms of limited duration in order to decrease the influence of the three elements that contribute to the knowing-doing gap.[196]

First, to overcome the consistency/conviction element that promotes over-commitment to prior decisions, changes in personnel may be necessary.[197] As discussed previously, a judge’s reputational investment in his prior positions, the convenience of relying on previous analyses, and/or his deep-seated convictions, may make it difficult for him to change course.[198] Rather than waiting for one or more Federal Circuit judges to perform the unusually self-disciplined act of making a public about-face on an issue and engaging in a labor-intensive analysis to limit or overrule precedents they had authored or voted for in the past,[199] the task of precedent correction should be entrusted to a new set of judges who were not involved in creating or perpetuating the precedents at issue, and, as a result, may be less hesitant to make changes when necessary.[200] To be clear, the consistency/conviction element will not be completely eliminated in a Federal Circuit staffed with rotating district judges. However, its ability to hinder self-correction may be substantially attenuated because the judges will be serving for limited terms.

Second, regularly rotating the membership of the Federal Circuit may help prevent the court from reaching an unproductive equilibrium, in which judges fall into predictable patterns of voting that may allow suboptimal precedents to issue and remain uncorrected.[201] Compared to a group of permanent judges with a long history of service together, a group of temporary judges who serve staggered, limited terms might be more amenable to rethinking existing doctrines. Each rotation of temporary judges will introduce new members who may bring fresh perspectives and whose views may not be firm on certain issues. Because a temporary judge may not be fully aware of the ideological or doctrinal alignments of the other temporary judges—and the extent to which their views are set—he may perceive more opportunities (than a permanent Federal Circuit judge) for persuading his colleagues to critically re-evaluate existing case law. The regular rotation of judges may also inhibit the formation of opposing camps that give rise to panel-dependent outcomes and divergent lines of precedent. In addition, judges who serve limited terms may feel freer to “rock the boat” by invoking the labor-intensive en banc procedure because the term-limited nature of their appointments may heighten their sense of purpose—i.e., to improve the adjudication of patent disputes—while decreasing the relative importance of maintaining a predictable adjudicatory equilibrium that allows judges to serve comfortably for an indefinite period of time with the same colleagues.[202]

Some, however, may point to the high level of dissent at the Federal Circuit[203] as indicative of a court with diverse views that is frequently engaged in a critical analysis of its precedents. However, the frequency of dissent and separate opinion-writing may be largely a reflection of circuit culture and norms.[204] Moreover, a high dissent rate may indicate a high degree of entrenchment[205] with established camps of judges that have settled into an equilibrium of “agreeing to disagree,” where, on a given panel, neither the majority nor the dissenter perceives a compelling need to temper its views to reach a unanimous result.[206] Perhaps because of the high concentration of patent cases on its docket, the Federal Circuit, by design, might be particularly susceptible to camps developing among its permanent judges, whose views may have become progressively nuanced and divergent as a consequence of the repeated exposure to the same issues over a prolonged period of time.[207]

Rather than the rate of dissent, the en banc rate may provide a better sense of whether precedential ossification has set in.[208] The Federal Circuit’s en banc rate is relatively low,[209] which may indicate that, despite the frequency of dissents, the majority of Federal Circuit judges are choosing to avoid a costly, time-consuming endeavor whose outcome may be uncertain.[210] Ultimately then, the high rate of dissents, when coupled with the relatively low rate of taking cases en banc, may be indicative of a court where its members have settled into camps that give rise to multimodal, panel-dependent outcomes, and where the collective will to resolve conflicting precedents may be weak.

Third, having district judges serve limited terms may help create a sense of urgency at the Federal Circuit for fixing suboptimal precedents. As previously discussed,[211] permanent judges may view the process of correcting precedents as a labor-intensive endeavor that may be conveniently deferred by defaulting to their prior positions,[212] engaging in forms of collegial concurrence such as “going-along voting” and “live and let live opinion-joining,”[213] and studiously avoiding the en banc process.[214] Unlike a judge with a permanent appointment at the Federal Circuit, a district judge who serves for a limited time knows that after returning to the district court level, she will be required to follow the precedents she created—under the threat of reversal. Accordingly, district judges serving temporarily at the Federal Circuit may feel a heightened need to “get it right” compared to permanent judges,[215] such that they might be more willing to undertake the effort to correct, or prevent the issuance of, suboptimal precedents.

At a more general level, the rotation of judges at the Federal Circuit may create a form of de facto percolation through regular changes in personnel, which may be a faster form of percolation involving more judges than increasing the number of circuit courts that hear patent appeals (i.e., percolation through space)[216] or waiting for Supreme Court intervention (i.e., percolation through time).[217] That is, with regular changes in personnel, the process of case law development at the Federal Circuit is likely to be more responsive than it is currently. If a rule is sound, subsequent instantiations of the Federal Circuit with a new slate of district judges will likely maintain it.[218] And if a rule is unsound, subsequent instantiations may be less hesitant to reconsider it and make adjustments. Ultimately, by allowing percolation to occur through regular changes in personnel, Federal Circuit precedents can be developed by individuals from a much wider variety of backgrounds and perspectives than is currently possible, while at the same time preserving a single appellate venue for patent cases.[219]

IV. Concerns and Objections

A. Stability of Case Law

Some might object to staffing the Federal Circuit with a rotating group of district judges on the ground that it could potentially destabilize patent case law.

This proposal is certainly not without costs, and the potential for doctrinal fluctuations is one of them. However, because Federal Circuit case law has limited opportunities for the type of corrective percolation that occurs among the regional circuits,[220] and, at the same time, must adapt to changes in technology,[221] the benefits associated with a more responsive system for reconsidering and updating precedent, as provided by the judicial rotation proposal, are expected to outweigh the costs associated with any temporary doctrinal fluctuations. Moreover, the current status of the Federal Circuit as “the de facto administrator of the Patent Act”[222] may tilt the cost-benefit analysis in favor of implementing some mechanism for regularly rotating its membership. Whether judicial rotations may be cost-justified for any of the regional circuits is an issue left to future research.[223]

The entrenchment of suboptimal precedents may be a more serious problem for the Federal Circuit than the transient doctrinal swings that may result from the court’s attempts to further refine its case law based on fresh insights that new members may bring with each rotation. The development of sound precedent is necessarily an iterative process, and the use of term-limited, rotating judges could cause fluctuations in precedent to occur more frequently within a shorter time period. At the same time, the case law associated with a particular doctrine may be less prone to getting stuck in a suboptimal state. In contrast, with permanent judges, the process of self-correction by the court may be much slower such that suboptimal case law might have the appearance of stability or doctrinal “consistency” because it is not being actively reconsidered, as opposed to enduring on its merits.

The potential for doctrinal instability under the rotation proposal could be mitigated, in part, by amending the Federal Circuit’s IOPs. For example, to avoid the potential loss of institutional or historical knowledge regarding Federal Circuit case law when a group of temporary judges rotates off the court, the IOPs could be amended so that the STA would once again provide the judges with reports analyzing every draft precedential opinion for potential conflicts prior to issuance. And, as compared to the permanent Federal Circuit judges, the rotating district judges, as non-experts, might pay closer attention to the STA’s reports.[224] If the proliferation of too many precedential opinions in the course of self-correction is a concern,[225] the IOPs could be changed so that the rate of issuance of precedential opinions decreases, while allowing more non-precedential dispositions and Rule 36 judgments to issue, so as to dampen any precedential whipsawing that might occur from the accelerated percolation resulting from the rotations. Currently, an election to issue a Rule 36 judgment requires panel unanimity, while a majority is required to issue an opinion as nonprecedential.[226] To decrease the proportion of dispositions classified as precedential, the IOPs could be amended so that an opinion may be issued as precedential only if all panel members agree on that designation.[227] Alternatively, the election to make an opinion precedential could be taken away from the panel that decided the case, and instead given to a different panel of judges.

Finally, the opposite concern might also arise: whether the district judge rotation proposal could actually make doctrinal change less likely. That is, when in doubt, might the district judges be inclined to defer to an existing body of law created by “the experts”? It is possible that this inclination could exist at the very beginning of the tenure of a district judge who has rotated onto the Federal Circuit. However, this inclination may be tempered as the district judge settles into his appellate role and delves into the body of Federal Circuit case law on a regular basis. Overall, the potential for undue deference to prior case law is likely to be weak because the district judges who are selected for the rotations, while not experts, are not novices either.[228] Moreover, based on their experiences with adjudicating patent cases at the trial level, the district judges may relish the opportunity to revisit those doctrines have been troublesome for them, especially because they will have to live with the precedents they create when their rotations end.

B. Quality of Adjudication

Some may argue that the participation of district judges in Federal Circuit appeals could compromise the quality of adjudication, because their level of patent law expertise is lower than that of permanent Federal Circuit judges, and that the selection criteria applied to individuals who become district judges may be somehow less rigorous than that of Federal Circuit judges.[229] However, as suggested by Nard & Duffy’s proposal to expand the number of circuit courts to hear patent appeals to allow inter-circuit percolation,[230] and Judge Wood’s endorsement thereof,[231] patent law expertise at the appellate level may not be nearly as important as ensuring that robust mechanisms exist to facilitate reconsideration and correction of case law. The Federal Circuit’s susceptibility to expertise-induced blind spots[232] and lack of receptivity to debiasing information[233]—combined with the institutional inertia arising from lifetime appointments[234]—may well result in suboptimal case law that is likely to endure. The tradeoff then, is whether an expert court that might be difficult to correct is preferable to a non-expert court that might be easier to correct. In view of the unique requirements of patent case law—namely, uniformity and the ability to adapt to changing technologies—the latter option might be preferable.

In addition, staffing a “specialized court” with a group of generalist district judges may check the tendency of that court to develop case law that unnecessarily deviates from the mainstream practice of the regional circuits.[235] Permanent Federal Circuit judges may see themselves as “boosters” of patent law,[236] such that the appropriate patentee-public balance in patent case law might be better maintained by a Federal Circuit that is staffed by temporary judges whose reputations are not solely dependent on their work at that court, and, as a result, would be less inclined to develop precedents in a direction that would enhance the court’s influence at the expense of other governmental institutions and the public.[237] Furthermore, because the generalized nature of their dockets regularly exposes district judges to issues that lie at the federal-state court interface, they may be more sensitive to the need to strike an appropriate jurisdictional balance between federal and state courts in cases involving state law claims that raise issues related to patent law.[238] Finally, whether someone is appointed to a circuit court versus a district court is less a function of qualifications and more a function of politics and chance.[239]

Another concern with staffing the Federal Circuit with only district judges is that they might be reluctant to reverse a fellow district judge on appeal based on sympathy or other reasons unrelated to the merits.[240] However, this may not be a substantial risk: according to one study of appeals terminated in the federal appellate courts from 1987 to 1992, the reversal rate for appeals from district courts where the panel included a district judge sitting by designation (18.54%) was indistinguishable from that of all panels in appeals originating from the district courts (18.57%).[241] It is possible, however, that the similarity in reversal rates might be partially attributable to the designated district judges deferring to the appellate judges on their panels,[242] such that a panel comprised of only district judges might feel freer to affirm more often. To decrease the likelihood of undeserved affirmances under the judicial rotation proposal, the cases assigned to a panel should not include any appeal from the home district of any panel member. Also, if the identity of the author is masked through the issuance of a “per curiam” opinion, a district judge might feel more comfortable authoring an opinion reversing a fellow district judge. To the extent that the affirmance rate might increase by a nontrivial margin in a district-judge-only Federal Circuit, it could reflect a salutary development that counterbalances the previous expansion of the number of issues subject to de novo review.[243] In addition to the standard for reviewing claim construction,[244] the Federal Circuit recently extended de novo review to the objective prong of the willfulness standard,[245] and the objective reasonableness determination for an exceptional case under 35 U.S.C. § 285.[246] If the Federal Circuit were staffed by only district judges, the court may be less amenable to expanding the number of issues that are reviewed de novo and may even reverse this trend, which could ultimately bring greater stability—not less—in the adjudication of certain issues on appeal.[247]

Another potential objection to staffing an appellate court with only district judges is that they are allegedly less comfortable than circuit judges with the type of group decision-making that takes place at the appellate level, as opposed to the solitary decision-making process at the trial level.[248] This claim appears weak in light of the frequency with which district judges sit by designation on, or are elevated to, appellate courts. As for including appellate judges from the regional circuits in the Federal Circuit rotation, it would likely make the proposal less effective, as very few regional circuit judges have any experience with patent cases (let alone an intermediate level of experience that is recommended for the rotations).[249] At the same time, regional circuit judges may be less accountable because they are unlikely to be in a position to apply the precedents they create at the Federal Circuit as binding authority when they return to their home courts. In addition, the hierarchical difference between the regional circuit judges and the district judges might adversely affect the district judges’ independence of judgment required to critically re-evaluate precedents.[250]

While the analysis of the district judge rotation proposal has focused primarily on its potential impact on the Federal Circuit’s patent case law, the effect of the rotations on the non-patent portions of the Federal Circuit’s docket is expected to be approximately neutral. Much of the Federal Circuit’s non-patent docket originates from agency tribunals, the Court of Federal Claims, and the Court of International Trade.[251] The variety of cases that a generalist district judge typically handles is far greater than the Federal Circuit docket—in terms of the types of government agencies, tribunals, parties, and issues,[252]—such that it is unlikely that the rotating judges will have difficulty with, or materially affect the quality of, the adjudication of the Federal Circuit’s non-patent docket, particularly when the scope of appellate review is often restricted in such cases.[253]

C. Are There Less Radical Alternatives?

Some may question the need for regular rotations at the Federal Circuit, given that turnover does occur, albeit at a slower pace. Since 2010, six new judges have been appointed to the court: Kathleen O’Malley,[254] Jimmie Reyna,[255] Evan Wallach,[256] Richard Taranto,[257] Raymond Chen,[258] and Todd Hughes.[259] The new judges constitute half of the twelve authorized active judgeships.[260] Based on this turnover, might the culture of the Federal Circuit change so that it is more amenable to timely self-correction? In drawing an analogy to the assimilation of immigrants, Rochelle Dreyfuss suggests that it might take three generations of Federal Circuit judges to shed the defensive culture of the Markey-era and internalize the general norms of the rest of the federal judiciary.[261] But the behavioral elements[262] that impair timely self-correction will likely remain, as they are grounded not in the culture of the court, but rather in the organizational structure based on the types of individuals involved (i.e., experts) and their operating environment (i.e., working with colleagues having lifetime appointments).

Others may wonder if changes to the Federal Circuit’s IOPs may be sufficient to improve the court’s ability to timely address suboptimal case law, so as to obviate the need for structural changes. For example, the impact of the consistency/conviction element on the ability of a court to reconsider precedent may be weakened by revising the IOPs so that the panel judges are excluded from participating in the en banc consideration of an opinion issued by their panel. In addition, if all opinions were issued “per curiam,” it may better focus the panel judges’ attention on reaching the right result by decreasing the influence of considerations related to the authoring judge’s need to publicly save face or seek vindication of prior positions.[263] Furthermore, the IOPs could be amended to make it more difficult to designate opinions as precedential (e.g., requiring a majority vote of non-panel judges) in order to prevent the proliferation of divergent or suboptimal precedents. To further enhance intra-circuit percolation, the IOPs could be changed so that the presiding judge on the merits panel would be assigned randomly, without regard to seniority, to ensure that the authorship of significant opinions is as varied as possible.[264] And finally, the STA’s review of all precedential opinions could be reinstated.

While the current version of the Federal Circuit could, in theory, implement these suggested changes to its internal procedures, they are highly unlikely to be adopted. This is because such changes might be perceived as impairing the ability of individual judges to fully participate in the development of precedents (especially if their ability to participate on an en banc panel is restricted) and to establish a legacy (especially if they cannot be identified as the author of a specific opinion), which may be unacceptable to those permanent Federal Circuit judges who, unlike the district judges on temporary rotations, may have their judicial identities closely—and exclusively—associated with the Federal Circuit. And, just as the STA’s conflict memos were scaled back, the court may change its IOPs at any time, so any changes may not last.

Finally, some may object to the proposal on the ground that a less radical option exists: have more district judges sit by designation at the Federal Circuit[265] and vice versa.[266] However, increased designation practice may not be an adequate substitute.

Having district judges sit by designation more often at the Federal Circuit might increase awareness among the Federal Circuit judges that certain precedents are problematic. However, visiting judges do not participate in the Federal Circuit’s en banc process,[267] and the additional insights supplied by visiting district judges may have little effect on closing the knowing-doing gap. As discussed previously, an environment that is conducive to self-correction is not one in which an unproductive equilibrium exists with predictable voting patterns, but rather one in which the judges share a sense of urgency in undertaking the effort to reconsider and take corrective action on problematic case law.[268] The current adjudicatory environment of the Federal Circuit that is maintained by the permanent judges is unlikely to be materially changed by the district judges who sit by designation, as the latter will be both outnumbered and outranked by the former. Staffing panels with a mix of judges at different levels of the judicial hierarchy may result in district judges subconsciously deferring to the circuit judges, such that the former may dissent less often,[269] while the latter may subconsciously assert their superior position in the hierarchy.[270] Indeed, the hierarchical disparity between permanent Federal Circuit judges and visiting district judges may adversely affect the latter’s independence of judgment necessary to engage the other judges in a critical, rigorous analysis directed to reconsidering existing precedents.[271]

Conversely, if Federal Circuit judges were to sit by designation in the district courts more often, it is possible that it could improve their understanding of the impact of Federal Circuit case law at the district court level. However, a Federal Circuit judge may need to spend at least a year at a district court handling a full district court docket in order to meaningfully experience the challenges district judges face in managing patent cases with the rest of their caseload. In this regard, the current practice whereby some circuit judges preside over cherry-picked district court patent cases[272] is wholly inadequate. Furthermore, it is unlikely that a critical mass of Federal Circuit judges will sit by designation at the district court level because they may have neither the interest nor the necessary skill set (e.g., greater familiarity with trial practice, enhanced project management skills, and facility with a wide range of subject matter, especially criminal procedure). But even if every Federal Circuit judge were to sit by designation at the district court level, there is no guarantee that they will use their newly-acquired knowledge to correct suboptimal precedents when they return to the Federal Circuit—the knowing-doing gap will likely exist so long as the environment of the Federal Circuit remains one where a group of experts serve terms of indefinite duration.

Conclusion

If the courts are to solve the patent crisis,[273] the precedents issued by the Federal Circuit must strike the proper balance between private and public interests. And if the precedents become obsolete, unworkable, or are otherwise suboptimal, the Federal Circuit should take prompt remedial measures.

In response to concerns about the Federal Circuit’s limited ability to experiment and adapt its precedents to changing needs,[274] this Article explores the cognitive and situational considerations that may underlie the Federal Circuit’s inability to correct its precedents in a timely manner, and proposes a solution. By staffing the Federal Circuit with a rotating group of district judges, the blind spots and institutional inertia arising from the curse of expertise and the knowing-doing gap, respectively, could be mitigated. Otherwise, by maintaining the present staffing arrangement whereby a group of judges are permanently assigned to the Federal Circuit, timely self-correction may occur only if the judges behave in ways that are unusually self-aware and self-disciplined.