What the Federal Circuit Can Learn from the Supreme Court -- and Vice Versa

I would especially like to thank Michael Carroll for inviting me to deliver this Lecture. It is a real privilege to have been asked, and a pleasure to be here. I would also like to thank him for suggesting that I talk about one of my favorite topics. I have been a student of the Federal Circuit for some time. The statute enacting the court was passed while I was clerking in the Second Circuit for Judge Feinberg, and I have to tell you, the smile on his face when he discovered he would never have to hear another patent case was a beauty to behold. Also, while I was a law clerk to Chief Justice Burger at the Supreme Court, the hard work of getting the court organized was taking place. Justice Burger took a special interest in judicial administration, so during that period, he spent time with Howard Markey. Markey had been the Chief Judge of the Court of Customs and Patent Appeals, and he was slated to become the First Circuit's first Chief. The two of them decided that since I had been a chemist before I [Vol. 59:787 went to law school, and because I was interested in patent law, I should be Markey's inaugural Federal Circuit law clerk. It would have been a lot of fun, but unfortunately my life-my husband and children-were moving back to New York. So I had to decline, and I started to teach at NYU Law School instead. Nonetheless, I remained curious about how the court was faring. And at its fifth anniversary, which was right before I was to get tenure, I decided to write my "tenure piece" on its jurisprudence. 1 Since then, I have revisited the question of the court periodically.

To a student of patents, civil procedure, and legal institutions, the Federal Circuit is a superb subject for academic inquiry because it was actually created as an experiment; there are not too many experiments in law, so this one was pretty special. What happened is this: in the early 1970s, federal appellate dockets had increased to the point where the regional circuits could no longer handle the load. The first impulse was to add new judges to existing circuits. But that would have been of limited help because increasing the number of judges would lead to more intra-circuit inconsistency and would breed more cases for the circuit courts to decide. Adding new circuits was also a possibility, but such an addition would also have been problematic because more circuits would mean new opportunities for inter-circuit splits, and that would breed more cases for the Supreme Court to decide.

The Hruska Commission was convened to study the issue, and in 1973, it conceived the idea of experimenting with specializationwith pulling a class or classes of appeals out of the regional system and funneling them into a special appellate tribunal. 2 The new court would reduce the dockets of the regional circuits, and it could, in theory, do much more. Its presence could also diminish opportunities for forum shopping and take pressure off the Supreme Court. If it were small enough to speak with a single voice, it could bring more coherence to the law it administered. And, with greater expertise in the field, it might decide cases more efficiently. As Judge Markey told Congress, "[I]f I am doing brain surgery every day, day in and day out, chances are very good that I will do your brain surgery much quicker . . . than someone who does brain surgery once every 2010] FINNEGAN LECTURE 789 couple of years." 3 Although the Hruska Commission report actually made several suggestions for fields that would be appropriate for specialization, patent law was, if you will, "the killer app."

Way back in 1911, Learned Hand was called upon to decide whether purified adrenaline was patentable subject matter. 4 It was a significant issue then, and the case retains its importance now, because its holding is thought to support patenting in the biotech sector. Significantly, in his rather brilliant opinion, Judge Hand nevertheless ended as follows:

I cannot stop without calling attention to the extraordinary condition of the law which makes it possible for a man without any knowledge of even the rudiments of chemistry to pass upon such questions as these. . . . How long we shall continue to blunder along . . . no one knows; but all fair persons not conventionalized by provincial legal habits of mind ought, I should think, unite to effect some such advance.

Congress decided, in light of the Hruska Commission Report, that seventy-one years of blundering was enough. The "advance"-the Federal Circuit-would, in one fell swoop, solve the problems the Commission was formed to address, answer Learned Hand's plea for technologically expert judging in patent cases, and bring that smile to Judge Feinberg's face.

But it is important to remember that this was an experiment, and it was risky. Specialization had been tried in the past and, for the most part, failed. The poster child was the Commerce Court, which was created in 1910 to review decisions of the Interstate Commerce Commission. The court was so hated-by the railroads, by shippers, by the public-that it was disbanded within three years of its founding (and one of its five judges was also impeached, but that may be a different story).

In fact, there were many reasons to be concerned about specialization. Isolating patent law could favor special interests. Appointments to regional courts-that is, to generalist courts-are so highly contested by so many special interest groups that they dilute each other's effectiveness. But when there is only one field to fight about, those who are better organized and have the most moneywhich, in this context, is probably patent holders-can "capture" the appointments process so that judges are predisposed to their interests. Even without capture, patent holders would, it was thought, have an advantage. As repeat players, they could manipulate the way that important issues were framed for litigation.

There were also fears that the judges might develop tunnel vision. The saying is, if you have a hammer, everything looks like a nail. In an effort to support innovation, the judges might be so focused on patents, they would ignore non-patent incentives to innovate, such as intellectual curiosity, the availability of prizes, or competition. Furthermore, people were worried about the Federal Circuit's isolation. They were concerned that the court's exclusive jurisdiction would take patent law out of the judicial mainstream and deprive the law of the benefits of cross-fertilization.

7 Finally, the bar was worried that there would be difficult boundary problems on the allocation of cases among the appellate courts.

Congress took these concerns seriously. While the Federal Circuit was given authority over all, or, as we will see, nearly all patent appeals, it is not specialized in the traditional sense because there are many other sources of its judicial authority. These include such areas as diverse as contracts, torts, export controls, labor law, and energy issues.

Other industries and bar groups are therefore involved in lobbying for appointments, and the judges must stay abreast of non-patent legal developments.

Indeed, from most perspectives, the Hruska Commission's experiment has been a raging success. 9 The court has now passed the quarter-century mark. The patent industries and the patent bar are delighted with it, and-in what might be the biggest compliment of all-many other countries are copying it. 10 On the whole, the concerns people expressed about specialization have not eventuated. There has been no capture of the appointment process. If anything, there is concern that not enough appointees have had patent 12 -the Federal Circuit had "case" rather than "issue" jurisdiction: once a case is properly before the Federal Circuit, it decides all of the issues, not just the patent ones. 13 At the same time, in Christianson v. Colt Industries Operating Corp.

14 and Holmes Group, Inc. v. Vornado Air Circulation Systems, Inc., 15 the Supreme Court withheld Federal Circuit jurisdiction when the patent issues in a case appear only in the defense or in a counterclaim.

These decisions have the benefit of giving the court a somewhat broader perspective on innovation policy, while also creating an interchange with other national courts.

Most important, however, the Federal Circuit's success can be attributed to the many positive contributions it has made to patent law. Most obviously, for patent appeals, the Federal Circuit is almost the only game in town. As a result, it has eliminated forum shopping and has attained a high degree of national uniformity, and that is a value that the industry positively cherishes. No one will make heavy commitments of time or money if there is uncertainty about what law is going to apply. There are some exceptions here, including, as Kimberly Moore has shown, some forum shopping at the district court level. 17 But the situation has vastly improved. The Federal Circuit has also made a key procedural innovation. In Markman v. Westview Instruments, Inc., 18 the court, with the Supreme Court's approval, eliminated jury trials on claim construction. 19 ruling created more predictability in the interpretation of patent claims, yet another value that the industry holds in high esteem. To be sure, there are still complaints about the continuing indeterminacy of claim construction, but in an empirical study, Jeffrey Lefstin demonstrated that the level of uncertainty in this area specifically is no different than that for contract interpretation generally. 20 Indeed, in another empirical work, Lefstin showed much more: that the Federal Circuit's developed expertise in patent law has made the law more predictable across a whole range of issues, including infringement, validity, and inequitable conduct. 21 And according to Robert Gomulkiewicz, the court has also become highly knowledgeable about the technology business. 22 Because the court sees so many cases about patent transactions, it is now an influential voice within the federal judiciary as a whole on questions involving licensing.

Given the happiness within the bar and within the patent industries, and given the eagerness of other countries to copy the Hruska Commission's experiment, it is perhaps a surprise that lately the Supreme Court has changed its practice. In the first twenty or so years, its review of the Federal Circuit was largely intermittent and confined to procedural issues-cases like Hohri, Christianson, Holmes, and Markman. However, the Court has recently begun to intervene regularly; it has begun to address the substance of patent law; and it has reversed, vacated, or questioned nearly every decision: MedImmune, Inc. v. Genentech, Inc., on standing to challenge a patent; 32 on a dissent from the dismissal of certiorari, Justice Breyer cast doubt on the Federal Circuit's subject matter jurisprudence. 33 And there is yet another subject matter case pending in the Supreme Court-Bilski v. Kappos 34 -where there are similar doubts about the Federal Circuit's rule.

Significantly, we are in an era in which Supreme Court review is otherwise declining. Furthermore, heightened review of patent cases is happening in the absence of the circuit splits that usually attract Supreme Court attention. Accordingly, one really must wonder about this level of activity and whether it is an implicit criticism of the Federal Circuit's work.

I, however, am going to argue tonight that heightened review should not be taken as a criticism of the Federal Circuit. The Supreme Court's interest in patent law is at worst neutral: every circuit comes into focus periodically, and now that the Federal Circuit has come of age, it is taking its turn. More to the point, Supreme Court involvement in Federal Circuit decisions should be regarded as highly salutary, for these two tribunals have a great deal to learn from one another.

What I mean is this. The Federal Circuit must, obviously, learn from the Supreme Court, for it is bound by the Supreme Court's decisions. But the relationship between these two courts is not simply Both of these courts are caught in the Hruska Commission's experiment; they must both figure out how a judiciary largely committed to generalist adjudication should deal with a court that is so differently constituted. And both courts are, in a sense, courts of last resort (at least, the Federal Circuit is when the Supreme Court is not so focused on its activities). In that capacity, both have weighty responsibilities regarding the substance of national law and for supervising the courts below them. Of course, they see these problems from different perspectives. Sharing their views-learning from one another-could enhance the operation of the patent system, shed light on the costs and benefits of specialization, ease the path for other specialized courts, and improve judicial administration more generally.

Let me start with the first area of specialization, where there are two distinct problems: the Federal Circuit's relationship to the generalist Supreme Court, which reviews its work, and its relationship to the generalist trial courts, whose work it reviews. I will describe both problems and then discuss the lessons to be learned.

Of these, the harder question is the Federal Circuit's relationship to the Supreme Court. Obviously, the Federal Circuit is subject to Supreme Court review-the same as the other circuit courts-but review here seems particularly intrusive. The judges on the Federal Circuit have built up experience over their years of service, while the Justices of the Supreme Court do not even have a generalist's knowledge of patent law. After all, their own experience on lower court benches could not possibly have given them any perspective on patent law because-and there is some irony here-all the patent cases had been diverted to the Federal Circuit by the time most of them were appointed. Justice Stevens is an exception, and he is approaching retirement.

On the one hand, Supreme Court involvement dilutes the Federal Circuit's hard-won expertise, but on the other hand, the Supreme Court's involvement may be more important in the case of the Federal Circuit than it is for the other courts it reviews. Consider the common law. Although patent law is nominally statutory, it leaves wide gaps for judge-made law. And common law judges make law in an evolutionary and collaborative fashion.

Take the Evarts Act, 36 which established the regional circuits. At the time the Act was passed, the question was whether these new regional circuits would be bound by each other's law. An approach requiring appellate courts to follow one another's precedent would have had the benefit of guaranteed national uniformity. Nonetheless, the decision was made to give each circuit judicial independence-that is, to forego national uniformity-so that the law would "percolate," allowing the fittest rule to survive. "Survival of the fittest" is no longer possible for patent law, because apart from cases in the Holmes or Christianson posture, the only circuit court hearing patent cases is now the Federal Circuit.

If, then, we are going to get evolution in patent law, it has to be through a different mechanism. Supreme Court involvement in patent decisions is one such avenue. 37 There is another reason Supreme Court involvement is necessary: despite congressional attempts to give the Federal Circuit cases outside patent law, patents remain at the core of its docket, at least in the innovation area. The court has little chance to see how patents fit into the economy as a whole. The Supreme Court does have that perspective.

And in the recent group of cases, we see the difference Supreme Court involvement can make at both the micro and the macro level. Thus, at the micro level the Supreme Court has made smallish doctrinal adjustments intended to keep patent law in the mainstream. In eBay, the Court claimed it was making sure that the standards for injunctive relief stay the same across all federal causes of action; 38 in MedImmune, it made the test for standing uniform; 39 in Illinois Tool, it equalized the treatment of antitrust defendants.

At the macro level, the Supreme Court has, essentially, pressed the reset button. Although I earlier described the establishment of the Federal Circuit as an outgrowth of administrative concerns, one can also read the legislative history as revealing a strong interest in strengthening patent value and stemming what was then perceived as 2006) (holding that "in all cases involving a tying arrangement, the plaintiff must prove that the defendant market power in the tying product"). [Vol. 59:787 a flight to trade secrecy. The Federal Circuit took this commitment to heart (which, in part, may be why the patent bar has been so pleased with its performance). However, its success has turned into something of a mixed blessing: legal scholars, economists, the Federal Trade Commission, the National Academies, and even some in the patent industries have expressed concern that there are now too many patents, that they cover too much economic activity, that patent quality is declining, and that the high cost of patent litigation is chilling innovation.

The Supreme Court has moved in very effectively. For example, by giving district courts discretion over awarding injunctive relief, eBay's limit on permanent injunctive relief should reduce incentives to litigate;

42 KSR, which raised the standard of nonobviousness, relieved concerns about patent quality; Justice Breyer's dissent in LabCorp clearly flagged the problem of patent proliferation. As he stated: "[S]ometimes too much patent protection can impede rather than 'promote the Progress of Science and useful Arts.'" 43 Of course, many of these changes could have been made by Congress, but as its recent prolonged attempt at patent reform suggests, there can be wisdom in relying on a judicial approach.

In sum, that is one problem: figuring out how the Supreme Court can use the generalist knowledge derived from its unique position in a way that takes account of the Federal Circuit's expertise in technology, patents, and licensing.

The other problem is determining the Federal Circuit's role as a specialist appellate court reviewing a generalist trial court. The problem here is that appellate courts generally defer to the factual But the Federal Circuit does not have that problem. Besides, it chooses clerks for their technical backgrounds and it can hire staff to advise it on technical matters. Accordingly, a strong argument can be made that the relationships between these courts should be different. And significantly, the countries that have copied the idea of specialized patent courts have mostly established their patent courts at the trial court level, which is some indication that the real gains from specialization are reaped with respect to fact-finding. If that is true, it would be highly advantageous to find a way for the Federal Circuit to make an equivalent contribution, even though it is an appellate court.

So, those are the two specialization issues-what can the courts learn from one another? Although the Supreme Court is the older institution-and supreme-the fact of the matter is that the Federal Circuit faces these specialization issues more regularly. As a result, it has the most to teach. Starting at the end, with the question of trial court review: since its earliest days, the Federal Circuit has been attentive to the question of effective review of fact-finding. Likely, its concern initially arose because it was Chief Judge Markey's view that the way to establish the court's reputation would be to straighten out the mess that was nonobviousness, where the disparate views on the regional circuits had given rise to the most extremely corrosive form of forum shopping. In its earliest nonobviousness cases, the Federal Circuit therefore undertook a detailed examination of the patents in issue, at the prior art, and at their relationship. If And that Rule applied even though the Federal Circuit's grasp of the facts was clearly better than the trial court's.

That was a major loss. But the Federal Circuit did not give up on the enterprise both for nonobviousness and in general. Instead, it adopted two other approaches. First, it required the trial courts to apply specific analytical techniques to factual questions. For nonobviousness, for example, it required courts to examine secondary considerations-such things as commercial success and long-felt need. 47 Furthermore, it required proof of a teaching, suggestion, or motivation for combining prior art.

In addition, it started classifying many of the more complex technical issues as questions of law, rather than issues of fact, so that Rule 52 would not bar de novo review.

The Federal Circuit has, in short, efficiently canvassed the ways in which it can bring its expertise to bear on the facts that affect the outcome of technologically complex cases. Admittedly, by requiring these analytical techniques, it has sacrificed flexibility for predictability. But as we saw, improving predictability has very much pleased the patent industries.

But there is a catch. True, in Markman, the Supreme Court approved the idea of recharacterizing some factual questions as legal determinations. 50 But at least in part, it did that because the move fit nicely with the Supreme Court's own agenda about limiting factfinding by juries. 51 In fact, the Supreme Court is busy dismantling the analytical requirements. In KSR, its own case on nonobviousness, the Supreme Court began by "rejecting the rigid approach of the Court of Appeals." 52 What the Supreme Court has not done, however, is face the larger question of expertise head-on. If it does not like rigid rules, perhaps it should use its power under the Rules Enabling Act 54 to change Rule 52. Or, if it does not want to engage in that particular experiment with Federal Circuit exceptionalism, it could help the Federal Circuit find the "sweet spot" between rigid rules and standards. Either way, an acknowledgement of the Federal Circuit's attempts to deal with deficiencies in lower courts' handling of technologically complex factual issues could improve patent jurisprudence. Taking lessons from the Federal Circuit might also help the Supreme Court improve adjudication of technical issues in other complex cases, such as antitrust and environmental law.

What about the harder question, the one about the Federal Circuit's relationship to the Supreme Court? How do we get the benefits of Supreme Court intervention without sacrificing the advantages of relying on the Federal Circuit's growing expertise? There are two sub-issues here: When should the Supreme Court intervene-or more accurately, who should decide when the Supreme Court should decide? And, does the Supreme Court owe the Federal Circuit any special regard on substantive patent law questions?

On the first of these issues, I again think that the Supreme Court could learn a great deal from the Federal Circuit. In the past, the Supreme Court has sometimes wasted its time. An example is Pfaff v. Wells Electronics, Inc., 55 where the Court reached a perfectly reasonable position on when an invention was "on sale," 56 but the decision was no better than the one that the Federal Circuit would have found for itself.

Since the Supreme Court's resources are highly limited, 53. Brief Amicus Curiae of Franklin Pierce Law Center in Support of Certiorari at 3, Bilski v. Kappos, No. 08-964 (U.S. filed Mar. 2, 2009), 2009 WL 2445759 (claiming that Congress, the Supreme Court, and independent entities that have studied patent law "do not advocate limiting the scope of patentable subject matter"); Brief of Amicus Curiae Medistem Inc. in Support of the Petition for a Writ of Certiorari at 6, Bilski v. Kappos, No. 08-964 (U.S. filed Feb. 27, 2009), 2009 WL 564646 ("Narrowing the scope of patentable subject matter forces innovators to use other means of protecting their inventions, such as maintaining the invention as a trade secret."); Brief of John P. Sutton Amicus Curiae Supporting Petitioners at 3, Bilski v. Kappos, No. 08-964 (U.S. filed Feb. 25, 2009), 2009 WL 507782 (supporting certiorari to "clarify the law," but not to "make commodity trading into patentable subject matter").

54 And I have just said that on micro and macro issues, the Supreme Court ought to be the teacher. But should its views always trump? I would argue that the Supreme Court's testiness about the Federal Circuit's departures from its precedents is often inappropriate, and that it adversely affects Supreme Court litigation as well. Litigants are forced to rely on language from ancient case law that no one wants to resurrect, when they should be suggesting formulations that address contemporary problems.

A good example is the concept of "synergy." 63 In certain respects, then, the Supreme Court ought to conceptualize its relationship with the Federal Circuit as more of a dialogue than the product of hierarchy-as I said earlier-as the substitute for percolation. The mechanism for doing that is certainly there. Consider patentable subject matter: first, there was-in Justice Breyer's words-the Federal Circuit's "statement" about patentability in State Street. Then came Justice Breyer's dissent in LabCorp. That spurred a set of Federal Circuit cases, culminating in Bilski, which the Supreme Court then decided to review. Next came Prometheus Laboratory, Inc. v. Mayo Collaborative Service, 64 another subject matter case, and Prometheus will, of course, be followed by the Supreme Court decision in Bilski. So, there is plenty of opportunity for a really good interchange of ideas, and the Supreme Court seems receptive to that approach. Indeed, even when the Supreme Court reverses the Federal Circuit's decisions, the Court rather significantly leaves implementation questions to the Federal Circuit's discretion. One could also think about this institutionally. Is the Supreme Court the best institution to be setting mid-range policy, by which I mean crafting the policies relevant to the administration of patent law? In other technical areas, there is an administrative agency that fulfills that function. Now, I have left the U.S. Patent and Trademark Office (USPTO) out of this discussion because it was founded before the Administrative Procedure Act; lacks rule-making authority; and only sees the issues that arise when a patent is issued, not the ones that come up when patented information is used. 65 Perhaps things will change under the new Commissioner, David Kappos, but as currently constituted, the PTO just cannot play the institutional role I am discussing.

66 And as we have seen, Congress is not in a position to do much on this either. So, we have a technically complex set of problems, key to our economic health, and something of a vacuum on the institutional end. Now that all patent cases are before the Federal Circuit, it is uniquely positioned to take on the job of filling that void. But for that to happen, the Federal Circuit has to act like a teacher: it has to explain what policies it is adopting. This, indeed, is a place where the Federal Circuit could learn from the Supreme Court: not what the mid-level policy ought to be-I would leave that to the Federal Circuit-but how to make it evident what mid-level policies it has chosen and why it has decided to further them. In other words, the Federal Circuit must articulate the theory on which it is relying.

The Supreme Court works very hard to explain what it does. It often describes the alternative ways in which it could decide a case, it identifies the policy choices associated with each alternative, and it explains the theory behind the choice it is making. In part, that may be an outgrowth of confronting circuit court splits-it must explain to each circuit why it chose the rule that it did-but it is even true of In contrast, although the Federal Circuit routinely recites policy justifications for the statutory requirements of patent law, it rarely provides insight into the policy rationale for its own decisions. Indeed, some of the judges have publicly suggested that it would be wrong to explain (or even to be motivated by) policy. 70 Now, I have written elsewhere that the reason for denying policy motives may have something to do with the experimental nature of the court. Perhaps it did not want to make waves while it was "on probation" in the public's mind. 71 But after more than twenty-five years, I think it is safe to say that the experiment is over. The court is now part of the fabric of the U.S. judiciary.

Unless the Federal Circuit does a good job articulating, explaining, and justifying policy, it cannot play the institutional role I envision. For example, I just mentioned the potential dialogue on patentable subject matter. But there is a small flaw in the argument: when the Federal Circuit decided Prometheus, it never engaged Breyer's dissent in LabCorp-even though both of the cases were about the same type of invention. [Vol. 59:787 results.

73 This is true for other issues as well: Had the Federal Circuit explained why it was ignoring Rule 52 in Panduit, or why it was adopting analytical rules, perhaps the Supreme Court might have taken the problematic nature of its relationship to the trial courts more seriously.

I hasten to add that, to an extent, the Federal Circuit is learning this lesson. An example is the Federal Circuit's decision in Bilski, which very deliberately and repeatedly referenced Diamond v. Diehr, the Supreme Court's 1981 decision on the patentability of software. 74 That approach neatly teed up the problem of relying on outdated case law. We will have to see whether the Federal Circuit made the issue of its authority as a specialized court to stray from Supreme Court precedents clear enough for the Supreme Court to consider the problem.

The opinion-writing issue allows me to segue into another area where the courts have a great deal to teach each other, and that is on dealing with the special problems that come from being a court of last resort, with supervisory and administrative responsibilities. Once again, I think the Federal Circuit has something important to teach. The lesson goes back to the question of reviewing district court fact-finding. While I am also somewhat skeptical about over-reliance on rigid rules, the Federal Circuit deserves credit for taking its role in supervising the lower courts seriously. That is what made the court such a success in patent law circles, and there is a lesson there for the Supreme Court. As two of my colleagues, Sam Estreicher and John Sexton noted twenty-five years ago, the Supreme Court's own docket is cluttered with cases that arise directly from that Court's failure to provide clear analytical directions. 75 Now that the Supreme Court is taking fewer cases, that failure may become even more serious. Here, the Federal Circuit's responsiveness could act as a template for the Court.

In other respects, however, this is a place where the Supreme Court has more experience, and so the Federal Circuit has much to learn. First, there is the issue of writing informative opinions. One reason to think of the Federal Circuit as filling an institutional vacuum is because it is a court of last resort. As I tell my students, for patent law, the Federal Circuit usually is the Supreme Court. But if it is, it has to act that way. It cannot play a policy role unless it tells us what the policy is. Well-articulated policy is also important for supervising the lower courts. Indeed, it is a tool that might replace at least some of the Federal Circuit's famous rigidity. 76 That is, the better the trial court understands the policy that the Federal Circuit is trying to achieve, the more likely it will do what the Federal Circuit thinks is required.

A change might also help to clear the court's dockets. Federal Circuit judges have complained about appeals that are built around nothing more than minor changes in the wording of its holdings. But if the litigants better understood the underlying policy, these linguistic variations might seem less salient to them.

Another issue concerns what I call the repeat player disadvantage. A problem for courts of last resort, or for a court that has an institutional role in setting policy, is that the law it hands down can require revision. Justice Brennan used to call this "damage control." A new rule could be wrong, it might be confusingly formulated, or it may wind up applying to situations the court did not foresee. To do damage control, the court must take the issue up again. And therein lies the problem: litigants have to persuade the court to reconsider an issue it has already laid to rest, and that is not always comfortable. Repeat players-attorneys who appear before a court regularly-may not want to annoy the judges and jeopardize their credibility in future cases. 77 So, for example, John Duffy and Craig Nard note that the number of PTO certiorari petitions plummeted after the Federal Circuit was established. 78 As the quintessential repeat player, perhaps it has been trying to avoid that kind of friction.

Other litigators may also be facing this issue. It seems to me that one such example is the common law experimental use defense. In Madey v. Duke University, 79 the Federal Circuit seemingly reduced 76. See Dreyfuss supra note 9, at 803 ("The elaboration of policy would make the law more comprehensible, and thus easier to apply reproducibly.").

To sum up, the Supreme Court's recent interest in patent law is highly intriguing. It has caused consternation in patent circles. But it should not. There are many questions that these courts need to work through jointly: questions on when specialization is necessary, how it should be provided, under what circumstances a specialized court should be able to "pull rank" and claim that its expertise gives it a superior perspective. By teaching each other the lessons that come from their unique perspectives, these two courts can bring the Hruska Commission's experiment to fruition and make a truly significant contribution to judicial administration both here in the United States and abroad. Equally important, the Federal Circuit and the Supreme Court can together update patent law to the emerging needs of the "Knowledge Economy." As a student of both patent law and institutional design, it is a pleasure to watch these cases and this dialogue unfold. Thank you again for giving me the opportunity to talk about one of my favorite subjects.