As a United States District Court Judge, I enjoy patent cases

The Court of Appeals for the Federal Circuit does not itself report statistics on the number of reversals based on erroneous claims construction by the district judges, but the Federal Circuit reported that during the twelve-month period ending September 30, 2006, of all the 357 cases before that court that were appealed from the district courts and were terminated by a decision of judges of the Federal Circuit, only 13% resulted in reversal.^{^[14]} The average reversal rate of district court decisions appealed to the regional court of appeals for the year ending March 31, 2006 was 9.5%.^{^[15]}

If we United States District Court Judges get it right 90% of the time on average in all other types of cases, why are we not better at construing patent claim language? I know that in each patent case over which I have presided for the more than two decades I have been on the federal district bench, I have done my best to get it right. I also know, as the Chief Judge of my district court, that my colleagues in Chicago do so as well, and I am sure my other district court colleagues across the country are equally conscientious. Despite our individual efforts, however, the fact remains that if a party appeals our claim construction rulings, there is pretty close to a 40% chance our rulings will be determined to be incorrect and be reversed.

Despite the early predictions after the Markman decision that district judges’ decisions on claims construction would result in greater certainty on the soundness of the district judges’ decisions,^{^[16]} the reversal rate appears to persist and remains “steadily high.”^{^[17]} The resultant uncertainty at the district court level encourages disappointed litigants to be more inclined to appeal and invest the additional money for the appeal rather than settle after a district judge’s claim construction ruling. If that ruling is reversed on appeal, the costs go even higher and the litigation goes on with the uncertainty beginning anew unless the Federal Circuit has made a claim construction determination in the case that is dispositive.

I. The Root Bases of the Predicament

A. United States District Court Judges’ Work Loads

The workloads of the 678 United States District Court Judges who are spread across the 94 federal districts of our country is one reason for the predicament. We district judges are the front line in all areas of federal litigation. I feel honored and privileged to be among the district judge ranks, but I must confess that when I pause to reflect on the totality of the job expected of us federal trial judges, I sometimes feel overwhelmed. We 678 people, with the help of our senior judge colleagues, must address all matters that come before us in federal district court. From frivolous pro se civil matters brought by individuals who vent their frustrations in the words of a filed complaint rather wreak violence on society, to cases by the United States government that fall within the full breadth of the federal criminal code which Congress seems to incessantly expand, to complex patent disputes involving the latest and most sophisticated technological advancement., we must be ready to address them all.

Each new civil complaint is assigned at random to a particular judge of the district court in which that complaint is filed. Because we United States District Judges must be generalists to deal with a large and diverse caseload,^{^[18]} we typically cannot concentrate on staying abreast of the changing nuances of any specific area of the law. We must focus on the law pertaining to each case on our respective dockets as we attempt to resolve each case. The sheer volume as well as the diversity of a district judge’s workload demands constant attention to efficiency because, without constant attention to efficiency, the caseload would burgeon beyond the capability of any one judge to provide justice as is necessary in each case. Consequently, changes in the law unrelated to the case at hand may easily go unnoticed by a district judge until he or she must make a decision in that area of the law related to a case on his or her docket.

As a United States District Court Judge, I enjoy patent cases. I also enjoy the diversity of the federal litigation over which I preside. I enjoy the challenge of learning or renewing my acquaintance with each specific area of the law in which I must become adept to resolve each case. I do not have the luxury of spending the time to learn more about a specific area of the law than necessary to resolve a particular case or a particular issue that may need resolution. I may address an issue in depth in one case, especially if it is a patent case, and never see that issue again during the rest of my career on the bench. For example, Dana Corp. v. American Precision Co., is an opinion I wrote during my first four months on the bench regarding the patent law issue of repair versus reconstruction.^{^[19]} I found that the defendant’s processes did not infringe any of the plaintiff’s three patents in suit.^{^[20]} Although my decision was affirmed by the Federal Circuit,^{^[21]} in the more than twenty years that have passed since then, I have never had occasion to address that issue again.

To put this problem in actual figures, in the 12-month period ending March 31, 2006, there were approximately 313,000 total civil and criminal filings across the United States,^{^[22]} and less than 1% (2,812) of those cases involved patents.^{^[23]} If spread across the 678 district judges, there are approximately four patent cases per judge per year as compared to an average of 490 other civil cases and criminal cases per judge per year.^{^[24]} Over 8000 of those civil and criminal cases (greater than 2.5%) were in my district, the Northern District of Illinois, which is headquartered in Chicago.^{^[25]} Of those over 8000 filings in my district, around 460 of those cases were intellectual property filings, making my district the 5th busiest district court in the country for intellectual property litigation.^{^[26]} Yet, only 5.5% of our district’s docket were intellectual property and less than 1% were patent cases.^{^[27]} Due to the diversity of cases handled by us district court judges, we have limited time to learn and sort through the intricacies of the new technology which form the heart of the dispute. Because of the constant press of the rest of our caseload, we often find it difficult to devote the time that, in a perfect world, would be devoted to any case, let alone, a complex patent case. Additionally, the expertise developed dealing with one patent case on a particular district judge’s docket is often fleeting since, as mentioned earlier, patent cases on average are less than 1% of our collective caseload and the law can change quickly.^{^[28]}

B. District Judges are Typically Generalists by Trade and Training

United States District Court Judges are appointed by the President and confirmed by the Senate but typically not because of their prowess in patent litigation. Certainly, no one inquired into that area of my background when I was appointed in 1985. I dare say because we district judges are generalists and must deal with every type of federal case assigned to our respective dockets, none of my fellow district judges in Chicago were asked, before being appointed to the bench by the President, about their acumen in patent law. Many of my colleagues on the bench do not share my enthusiasm and enjoyment to preside over patent cases and other intellectual property cases. Because I have taught an advanced law school seminar entitled “Intellectual Property Trial Advocacy” every year for more than two decades at the University of Illinois and other law schools, I have had the incentive to stay abreast of the patent law as it has evolved and have attempted to incorporate new technological issues into my course materials during that time. Many other judges on the federal district courts, however, pressed as they are to resolve legal issues in other areas of the law, do not remain attuned to the current patent law issues because they do not have an interest in patent law or did not have patent litigation experience before their appointment to the bench.^{^[29]}

Additionally, judges without a technical background must learn the rudiments of technological analysis, and the advances in the technology that form the basis of the patent dispute on the job, without the assistance of unbiased tutors. We district judges typically are provided our education on the pertinent technology by the respective advocates in the case before us, who understandably emphasize the key points of the facts in their respective client’s favor. Without a technical background to help us parse the respective parties’ arguments, we judges are sometimes drawn to the wrong conclusion more than we otherwise would be if the factual premises underlying the factual basis of the dispute had a familiar ring based upon our prior experience or education.

II. Factors Contributing to the Predicament

A. The De Novo Review Standard

Before we district court judges shoulder all the responsibility for the part of the patent litigation predicament stemming from the high reversal rate, mentioned earlier,^{^[30]} let me just say in our defense, there is another major factor contributing to the number of patent claim construction reversals. The decisions we district judges make regarding patent claims construction currently are reviewed by the United States Court of Appeals for the Federal Circuit under a de novo standard, which allows the United States Court of Appeals for the Federal Circuit to determine the correctness of our rulings as to the definition of patent claim terms without any deference to what we did in the district court.^{^[31]}

As a result of the de novo standard of review applied to our claim construction determinations, we United States District Court Judges feel like the late comedian Rodney Dangerfield, because our opinions “get no respect.”^{^[32]} The de novo standard of review applied to claim construction allows the United States Court of Appeals for the Federal Circuit to look at our determinations anew and determine the correctness of our rulings as to the proper interpretation of patent claim terms based upon what parties present to the Court of Appeals for the Federal Circuit by way of both factual and legal argument, not what was presented to us trial court judges.^{^[33]}

A de novo standard of review is what the Federal Circuit announced in its 1995 en banc decision in Markman as to district judges’ claim construction determinations.^{^[34]} In 1996, The Supreme Court, in affirming the Federal Circuit’s en banc decision in the Markman case, agreed that judges, not juries, should make patent claim construction determinations.^{^[35]} The Supreme Court in its Markman opinion said nothing, however, about the de novo standard of review.^{^[36]} The Federal Circuit revisited the de novo issue en banc two years later in 1998 and ruled once again in Cybor Corporation v. FAS Technologies, Inc.,^{^[37]} that the de novo standard is the proper standard to be applied in reviewing district judges’ patent claim determinations. That remains the standard today.^{^[38]}

While claim construction involves many of what one would consider to be factual determinations, it is reviewed entirely de novo as an issue of law by the Federal Circuit, even though the meaning of a patent claim term to a person of ordinary skill in the art involved in the invention at the time of filing of the patent application has no precedential effect on the meaning to be given to that same claim term used in another patent in another area of art.

B. The Varying Guidance

The law denoted as precedential by the Federal Circuit is supposed to provide guidance to us lower court judges, but that has not occurred as clearly as desired regarding patent claim construction since the Supreme Court’s 1996 decision in Markman for two reasons: (1) the Federal Circuit’s articulation of the methodology to be used in the analysis desired of us district judges has varied over the past ten years, and (2) because of the factual differences inherent among different patent cases, the claim term definition in a particular case involving a particular patent is without precedential value in another case.

While the precedent of the Court of Appeals for the Federal Circuit provides some general guidance to us lower court judges as to the methodology we should employ in going about the task of claim construction, the Federal Circuit’s precedential opinions seem to provide conflicting views in the nuances of the task and the value we are to ascribe to each factor we consider. Shortly after the Supreme Court’s 1996 opinion in Markman, the Federal Circuit issued Vitronics Corp. v. Conceptronic, Inc.,^{^[39]} and established that evidence intrinsic to the patent and the patenting process was to be used and, only on the intrinsic evidence’s failure to provide the answer, was extrinsic evidence to be considered. In 1999, the Federal Circuit in Pitney Bowes, Inc. v. Hewlett-Packard Co.,^{^[40]} clarified that the Vitronics prohibition against the use of extrinsic evidence unless the intrinsic evidence failed to furnish the meaning was a “warning.” The Federal Circuit in Pitney Bowes stated that “Vitronics does not prohibit courts from examining extrinsic evidence, even when the patent document is itself clear.”^{^[41]} In 2002, three years after Pitney Bowes and after a series of opinions^{^[42]} in which the Federal Circuit stressed that claim terms were to be heavily presumed to have their ordinary meaning unless the patentee specificity rebutted the ordinary meaning in the text of the patent, the Federal Circuit issued Texas Digital Sys., Inc. v. Telegenix, Inc.,^{^[43]} which promoted dictionaries and encyclopedias as “objective resources” that are “the most meaningful sources of information to aid judges in better understanding,” the patent claim terminology to be defined.^{^[44]} The Federal Circuit informed us district judges and the patent litigation world that the Vitronics opinion’s relegation of dictionaries and similar sources of the ordinary meaning of a patent claim term to the status of extrinsic evidence was “misplaced.”^{^[45]}

Two years ago in its en banc opinion in Phillips v. AWH Corp.,^{^[46]} the Federal Circuit provided us a further clarification that the patent specification, not dictionaries, was the best source to which district judges should look for the ordinary meaning of a term. The Federal Circuit in Phillips diminished considerably our reliance on dictionaries.^{^[47]} The Phillips opinion gave some guidance for us to follow as we district judges wrestle with the difficulties and pitfalls of claim construction in patent litigation, but Phillips again eschewed the notion that factual determinations were a necessary part of the claim construction process and again reaffirmed the de novo review standard.^{^[48]}

Therefore, we district judges now fully understand that we must follow the algorithmic procedure of looking first to the patent, its claims, specifications, prior art references, drawings depicting the preferred embodiments, and then turning to the prosecution history, which is the give-and-take with the PTO during the patenting process, before we turn to sources extrinsic to the patenting process such as dictionaries, the testimony of experts (who are persons skilled in the relevant art, into whose shoes we judges are to figuratively step in interpreting claim language), or the testimony of the inventor (even though the inventor is supposed to have the prerogative to be the lexicographer as to the patent’s terms). Yet, the value and weight to be accorded any inconsistent information stemming from the appropriate patent sources, as we consult them seriatim, remains the difficulty that I believe we district judges collectively have in patent claim term construction. The guidance on those points is understandably scant because of the factual differences we face as we move from interpreting the terms of a patent in one case to another, and because the definition of a claim term used in one patent has no precedential effect on the meaning to be given to that same claim term used in another patent in an unrelated field.^{^[49]}

An additional procedural wrench thrown into the claim construction process that seems to grate against the notions of uniformity and predictability of claim construction intended by the Supreme Court’s Markman decision is due in part to the failure of the Federal Circuit to provide us district judges with a set of uniform rules for the timing and format of the claim construction process.^{^[50]} The district courts apply a variety of modes for claim construction hearings, with, for example, 81% holding hearings on claim construction, 58% deciding claim construction before trial but after discovery, and 70% of claim construction proceedings held after the close of discovery.^{^[51]} Furthermore, since the Supreme Court’s decision in Markman, district courts are twice as likely to decide cases based on summary judgment.^{^[52]} The ad hoc manner in which claim construction is decided increases unpredictability and expense, especially if, after discovery and a trial or determination on a motion for summary judgment, a Federal Circuit panel reverses the district court’s claim construction.

C. The Costs of Patent Litigation

The key problem with the high reversal ratesCbesides frustrating us United States District Court Judges who are trying their best to get the cases right the first time, so that we don’t have to see them a second timeCis the cost and uncertainty that reversals add to the expense of patent litigation in the United States. In the United States, litigation costs of taking a patent lawsuit through discovery range from $350,000 to $3 million, and the costs of a patent case taken through to appeal range from $650,000 to $4.5. million.^{^[53]} For comparison, in the United Kingdom, which is the most expensive country in Europe for patent litigation, the costs range from approximately 200,000 pounds (about $370,000 U.S. dollars at current rates) to 1.5 million pounds (about $2.775 million US).^{^[54]} In Germany, which is the least expensive of the European countries, costs range from 15,568 (about $19,677 US) to 41,888 euros (about $52,944 US), although infringement and validity are tried separately.^{^[55]}

The higher costs of patent litigation in the United States as compared to European countries is apparently due to several factors. One is the unpredictability of the outcome in patent litigation in the United States.^{^[56]} This unpredictability may encourage litigants to take one of two tactics: (1) to pursue the litigation process to the hilt through expensive discovery, pretrial proceedings, trial and appeal or (2) in the alternative, to settle early a case upon which they may have otherwise prevailed to minimize the costs of litigation.^{^[57]}

The second apparent reason for the cost differential between the United States and the European patent systems is that the patent litigation pretrial discovery process takes a longer time in the United States. In the United States, claim construction often takes place after the close of discovery but before trial.^{^[58]} Pretrial discovery in a patent case typically takes several months and sometimes years. The patent case then proceeds to potentially dispositive motions and, if denied, to trial. This all must happen under our court procedures before the party who lost the claim construction determination can appeal the claim construction issue to the Federal Circuit. Should the Federal Circuit on appeal determine that the district judge’s claim construction was erroneous, all the money and time spent litigating the case in the district court following that erroneous claim construction by the district court is thus a wasted expense.^{^[59]}

The third apparent reason for the difference in costs between the U.S. and Europe is that the litigation process in the United States can be expensive, regardless what happens on appeal. For example, preparing and briefing a motion for summary judgment in a patent case in the district court could cost the parties hundreds of thousands of dollars. Additionally, the typical United States District Judge does not preside over more than one patent case in a year, and counsel for the parties must bear the costs of providing the judge, who has been assigned a patent case in her or his district, with an update on patent law as well as the factual background of the advancement in the technology at issue in each patent case.^{^[60]}

III. Proposals to Remedy the Predicament

As seen from the discussion above, there are several factors at work in creating the current patent predicament in this country: the de novo standard of review, the varying and inconsistent guidance that has been provided by the Federal Circuit, and the heavy and diverse workload and lack of technical and patent expertise from the district judges. In addition, there has been general dissatisfaction with the quality of patents granted by the U.S. Patent and Trademark Office (“PTO”), creating a greater incentive and need to litigate the validity of the granted patents through litigation.^{^[61]} With these multiple factors contributing to the problem of greater uncertainty and increased costs in patent litigation, there is no singular proposal that can effectively resolve the patent predicament. Instead, this complicated problem requires the application of several, complementary proposals to create a thorough and comprehensive solution. Below are several proposals, some of which have already been put into effect to some extent. I cannot emphasize enough that I believe that implementing all of these proposals so they will work together to directly address the key problems currently affecting the quality of the U.S. patent system and patent litigation is the procedure that should be followed to address the patent litigation predicament.

A. The PTO Establishing a Procedure for Post-Grant Review that Includes Claim Term Clarification

A grant of a patent by the PTO initiates the patentee’s legal right for a set period of monopoly that may be enforced through litigation in federal court. To ignore the PTO’s role when attempting to remedy the patent predicament is to cut off one of several valuable avenues in resolving the problems regarding both the accuracy of claim construction determinations and the high cost of patent litigation. The patent predicament does not begin in the district court, but with the grant of a patent by the PTO. The PTO in the last several years has been criticized for allowing patents of less than top quality to issue.^{^[62]} The low quality of patents granted by the PTO is due to a variety of impediments the PTO has had to face, including lack of proper funding, an ever-increasing filing rate and the fast-paced development of new technological subjects that are potentially patentable.^{^[63]} Any viable attempt at reforming the patent litigation system must include addressing the issues of quality at the PTO.

To comprehensively address the inaccuracy, unpredictability, and high costs of patent litigation, the first remedy that I support as part of the solution to the patent predicament, along with other strategies, is a review period following the issuance of a patent, called a “post-grant review” proceeding, during which third parties may challenge the grant of a patent.^{^[64]} The idea of the “post-grant review” had been the subject of a 2005 bill introduced in the 109th session of the House of Representatives,^{^[65]} but unfortunately the bill never proceeded further than being heard by the House Subcommittee on Courts, the Internet, and Intellectual Property. The post-grant review proceeding would provide third parties opposed to the grant of the patent a specified window of time from the date a patent was issued to challenge the patent before the PTO in a forum with limited discovery.^{^[66]} In the proposed legislation, an estoppel exception had been included to ensure that a challenger to the patent would not be prevented from presenting new evidence in a subsequent federal case that could not have been reasonably discovered during the post-grant review proceeding.^{^[67]}

The post-grant review proceeding would help to increase the accuracy of patent litigation results and lower patent litigation costs in several ways. First, a post-grant review proceeding would result in the granting of higher quality patents because the added layer of review would hold the PTO accountable for the patents that it issues and the quality of those decisions.^{^[68]} Allowing the PTO the opportunity to make corrections to newly issued patents would protect potential litigants that are negatively affected by a patent’s issuance, while still allowing for the patentee to be heard further within the PTO through an administrative adversarial process.^{^[69]} Another advantage to a post-grant review process by the PTO is that the district courts could have greater confidence in certain determinations that the PTO made in its post-grant review, such as the construction given to key claim terms by the PTO, leading to a more predictable and accurate patent system. In any subsequent litigation of a patent that has been provided post-grant review within the PTO, the parties to the litigation and the district judge presiding over that litigation will benefit substantially by having a better understanding of how the terms of the patent should be construed. This would ensure greater consistency and predictability with regard to patents claim construction by district judges. The increased uniformity and predictability of claim construction would, in turn, help to lessen the incentive of patentees and alleged infringers to gamble on the outcomes of litigation at the district court level and on appeal to the Federal Circuit by encouraging earlier settlements.^{^[70]}

The post-grant review process should therefore result in lower costs in any subsequent patent litigation because the process would provide parties faced with a potential infringement lawsuit an alternative, less expensive forum for challenging the validity of a patent.^{^[71]} The costs for the post-grant review forum is estimated to be between $500,000 and almost $4 million per party.^{^[72]} Doubts expressed about the procedure pertain to some distrust of the PTO’s ability to correct its own mistakes.^{^[73]} In addition, parties would be estopped from litigating the issue of invalidity a second time in federal court.^{^[74]} The possibility of being estopped from pursuing a claim in federal court could be a disincentive against using the post-grant review procedure, yet the lack of such a provision would not effectively reduce costs because parties would continue to litigate in federal court after using the post-grant review process.^{^[75]} A middle ground could be achieved by allowing review of the PTO’s determinations in United States District Court with a standard of review that requires deference by the court to the PTO.^{^[76]}

B. The Federal Circuit Providing Greater Deference and Guidance to District Judges’ Patent Claim Term Determinations

Of course, solutions aimed solely at the PTO are not enough to stem the tide of rising costs, unpredictability, and inaccuracy in patent litigation. Any effective solution must also focus on the courts and specifically the relationship between the district courts and the Federal Circuit. Remedying the high reversal rate by the Federal Circuit of district court judges on claim constructions necessitates addressing the intertwined issues of the predictability of outcomes and accuracy of decisions. As I previously outlined, the source of the patent predicament is rooted in district judges’ generalist skill sets and the large and varied caseloads each district judge has, which prevents district judges from having sufficient time and resources to master the complicated area of patent litigation. These issues in the district courts are then complicated by the need for guidance from the Federal Circuit with regard to claim construction, along with the unforgiving de novo standard of review applied to claim construction issues by the Federal Circuit. To solely address either the issue of the standard of review or the lack of guidance creates a choice between predictability and accuracy, respectively, and thus any proposal must encompass both these aspects of the patent predicament. Hence, both objectives must be addressed by the Federal Circuit in order to revive Markman’s unfulfilled promise of uniformity and predictability.^{^[77]}

As recently as the en banc decision in Phillips v. AWH Corp.,^{^[78]} the Federal Circuit has reaffirmed its intention to continue to apply a de novo standard of review to claim construction, despite criticism that the unforgiving standard increases the unpredictability of patent litigation by encouraging litigants to gamble on the Federal Circuit’s reversing a district judge’s determination. Several other possible standards of review could be applied instead of the current de novo standard. For example, starting with a very deferential standard, the Federal Circuit could review the district court’s claim construction for abuse of discretion. At the other end of the spectrum, but not quite as far as the de novo standard of review, the Federal Circuit could review questions of law de novo but findings of fact for clear error, to be overturned only if the Federal Circuit is firmly convinced that a mistake was made. This standard would necessitate the recognition by the Federal Circuit of the underlying fact-based issues involved in claim construction, rather than viewing claim construction entirely as a question of law. Finally, the Federal Circuit could create a standard of review somewhere closer to clear error rather than de novo review.^{^[79]}

A more deferential standard of review would reduce the reversal rate. A district judge may put more time and effort into construing a claim if the judge knows that his or her decision will be given some weight and deference.^{^[80]} This is just human nature. When an individual knows that the “buck stops here,” greater effort and time is often put into the decision. This not only applies to the district judges but to the counsel for the parties who appear in patent cases in the district courts. Patent litigation counsel will have a greater incentive to obtain a favorable district court ruling than they have now. Because this approach would result in a lower reversal rate of district judges’ construction of patent claim terms, patent law would offer more predictability and greater stability to litigants.

A majority of the current group of judges on the Federal Circuit appear willing to reconsider the rule of no deference to district judges’ patent claim determinations mandated by the de novo standard of review. Counting the positions stated or joined in by the Federal Circuit judges in the order denying the petition to hear Amgen v. Hoechst Marion Roussel, Inc.,^{^[81]} a careful observer can see that, in an appropriate case, reconsideration of the de novo review standard may occur. It appears that there now may be sufficient votes to modify the de novo standard of review that the Federal Circuit reaffirmed nine years ago sitting en banc in Cybor Corp. v. FAS Technologies, Inc.^{^[82]}. Chief Judge Michel specifically cited “a steadily high reversal rate; a lack of predictability about appellate outcomes; … a loss of the comparative advantage often enjoyed by district judges who heard or read all the evidence and may have spent more time on the claim construction than [the appeals panel], and inundation [of the Federal Circuit] with the minutia of construing numerous disputed claim terms in … nearly every patent case” as reasons to reassess the de novo standard for claim construction.^{^[83]} In addition to Chief Judge Michel, Judge Rader, Judge Newman and the Federal Circuit’s newest judge, Judge Moore, all dissented from the denial of the petition for a rehearing en banc based on a desire to address the de novo standard of review.^{^[84]} Further, the concurring Judges Gajarsa, Linn and Dyk wrote separately to state that their concurrence “should not be read as a endorsement of the panel’s claim construction in this particular case, nor as an unqualified endorsement of the en banc decision in Cybor Corp. v. FAS Techs., Inc.”^{^[85]}

Revising the standard of review alone, without more, however, will presumably sacrifice some accuracy at the district court level for greater certainty in the litigation process, because the precision of district judges’ claim term interpretations will not be scrutinized as closely.^{^[86]} To ensure an appropriate level of correctness in district judges’ claim term determination, the Federal Circuit will have to change more than just the de novo standard of review. The Federal Circuit must also fulfill the promise implied in Markman of providing United States District Judges with a uniform and clear system of guidance with regard to claim construction.^{^[87]} As mentioned earlier, many district judges, regardless of our desire to do it right, have limited experience and familiarity with patent law and, due to the press of our other judicial business, are not able to learn the subtleties of technology as much as we would like.^{^[88]} In order to avoid an exchange of diminished accuracy for greater certainty, a more deferential standard must go hand in hand with greater guidance to the district courts.^{^[89]} The Federal Circuit must pronounce clearer canons of construction to better assist district court judges. Problematically, the judges of the Federal Circuit, based on their written opinions, currently appear to disagree among themselves regarding the precise methodology that should be used in claim construction.^{^[90]} But to help resolve the patent litigation predicament, our patent litigation system must provide uniform guidance to us district judges when we preside over patent cases.

C. District Judges’ Greater Use of Case-Specific Specialized Advisors

To benefit fully from the better guidance the Federal Circuit will be providing us as to the patent law and procedure, we district judges should increase our use of specialized advisors, such as special masters, to assist us in better understanding the technology involved in the specific patent cases before us. A special master is generally a court-appointed attorney or professor authorized under our Federal Rule of Civil Procedure 53 to assist judges in various pretrial proceedings, such as discovery, settlement, recommendations for findings of law and fact, the formulation of remedies, or the calculation of damages.^{^[91]}

Because we generalist district judges in the United States are expected to decide claim term meanings from the viewpoint a person of ordinary skill in the field of invention and to determine how such a person would understand the claim at the time of invention, we typically could use some assistance in understanding the technology that is relevant to the patent in the suit.^{^[92]} The premise behind a district judge’s designation of a special master to assist the judge with claim construction or other tasks in patent litigation is to delegate to an individual with a better technical background an otherwise judicial task and obtain a recommendation from that person that can be evaluated by the parties and the judge before the judge decides to accept or reject the special master’s recommendation.^{^[93]} The special masters selected for such tasks would have technical expertise that would assist us district judges, who typically do not have the extensive background that is necessary to address a particular patent.^{^[94]} Employing the expertise of special masters more than we district judges do at the present time would help us to be better informed when making our claim language determinations and, hopefully, to lower the reversal rate.^{^[95]} For example, Judge John T. Ward in the United States District Court of the Eastern District of Texas, who has helped to make the Eastern District of Texas one of the busiest federal districts for patent litigation filings, in part due to the enactment of the Eastern District of Texas’s local patent rules, appoints technical advisors to provide recommendations as to claim construction.^{^[96]} More of us should follow Judge Wards’s lead and appoint special masters or technical advisors.

There are drawbacks, however, to the patent litigation system that must be considered regarding the appointment of special masters. Our system of justice places a high value on the concept of generalist judges addressing a wide range of issues, under a theory that the decisions of many generalists judges synthesize to give rise to a better system of law.^{^[97]} By using a special master, whose expertise is patent or technical issues, the parties may lose some of the benefits provided by having a generalist judge experienced in all procedural aspects of litigation. A special master typically will not have that expertise. Generalist judges also protect against what has been referred to as the “capture” of a specialized judge, or here, a special master, who deals only with one area of law, to a particular specialized view of a special interest group.^{^[98]} I note that the “capture” concept is often raised by commentators as a concern associated with specialized courts and judges as well.^{^[99]} However, I certainly believe that we district judges can keep an open mind and decide each case on its individualized merits. That is what our oath of office demands, and I believe we have and will continue to uniformly perform our judicial duties in that manner.

D. Congress’ Creation of Patent Case Reassignment Procedures and Training for District Judges

A further, recent potential innovation for ensuring better accuracy and predictability in patent litigation involves a proposal in Congress to enact case reassignment procedures along with training provisions for district court judges, set forth in the new pilot patent bill, H.R. 34 (formerly H.R. 5418), sponsored by U.S. Representative Darrell Issa and co-sponsored by U.S. Representative Adam Schiff, that recently passed the House.^{^[100]} On February 12, 2007, the U.S. House of Representatives voted to pass H.R. 34, and on February 13, 2007, the bill was received in the U.S. Senate, and read twice, and referred to the Committee on the Judiciary, where as of the date of this writing it remains.^{^[101]}

H.R. 34 proposes to create a pilot program in district courts nationwide, where United States District Judges of the pilot project district courts would choose whether to opt into hearing patent cases.^{^[102]} The patent cases involved in the pilot program would still be randomly assigned.^{^[103]} But if a patent case was assigned to a judge who desires not to hear patent cases, the case could be reassigned until it was assigned to a judge that had chosen to hear patent cases.^{^[104]} In addition, the legislation proposes to fund $5 million a year for the education and professional development in the area of patent law of designated judges.^{^[105]} Those funds would also be available for judicial training and for compensation of specialized patent law clerks with technical expertise.^{^[106]}

This solution would ensure greater accuracy of district judge decisions in patent litigation. The volunteering judges would create a pool of judges with better knowledge of patent law, greater exposure to those types of cases, and access to funds to further educate themselves in patent litigation, as well as hire clerks with technical backgrounds to assist.^{^[107]} Additionally, this pilot program may result in greater efficiency for deciding patent cases because judges with greater expertise in patent law can more quickly resolve patent matters.^{^[108]} Furthermore, several district courts have already begun to implement patent-specific local rules to ensure better procedural efficiency.^{^[109]} With better knowledge and procedural efficiency, the proposal may result in greater predictability of decisions, and from that, parties will hopefully have lower costs in connection with patent litigation.

At the same time, the designation of district judges to hear patent cases alleviates many of the concerns identified by opponents in the United States to a specialized patent trial court. The judges overseeing patent cases would still be generalists who hear a full caseload of other civil and criminal matters. The designated judges, armed with additional training, would be able to better understand the technological issues and the nuances associated with the advancements in technology without isolating litigants at the trial level in a forum apart from other litigation matters. These designated judges would thus continue to be a part of the traditional jurisprudence in the United States, encouraging cross-pollination among different areas of law and protecting against the balkanization of areas of law. As mentioned earlier, the diverse caseload of the designated judge also prevents any isolation or myopic approach to patent law and protects against the idea that a specialized judiciary could be captured by the patent bar.^{^[110]} I favor the H.R. 34 bill proposed by Congressmen Issa and Schiff, which will allow the self-designation of willing United States District Court Judges to hear patent cases as another remedy to help resolve the patent litigation predicament.^{^[111]}

There are, of course, worries voiced about the proposed pilot program. Among them is a concern that the pilot program will foster forum shopping by encouraging litigants to file patent cases in those districts.^{^[112]} While I believe that forum shopping generally should be discouraged, in this instance, where the choice of forum is based on a district’s development of an expertise in the complicated area of patent law, the federal judicial system of our country will not be harmed by encouraging this type of forum selection. On the other side of the forum shopping concern is the speculation that smaller districts not eligible to participate in the pilot program, that have developed local patent rules and expertise, such as the Eastern District of Texas, may see less cases. I find the chance of this outcome occurring to be negligible. I believe that litigants seeking the predictability provided by local patent rules in those districts will still wish to file patent cases there.^{^[113]}

^{^[1]} Chief Judge, United States District Court, Northern District of Illinois, (1985 to present); J.D.,University of Illinois College of Law, 1971; B.S., University of Illinois, 1968.

^{^[2]} Senior Law Clerk to Chief Judge Holderman (2005-2007); J.D., Harvard Law School, 2003; B.A., Duke University, 1999.

The authors gratefully acknowledge and thank Amy Dickerson, Executive Law Clerk to Chief Judge Holderman, and Christopher Stine, Law Clerk Extern to Chief Judge Holderman, for their invaluable assistance on this article.

^{^[3]} Judge James F. Holderman, Judicial Patent Specialization: A View from the Trial Bench, 2002 U. Ill. J.L. Tech. & Pol’y 425.

^{^[4]} H.R. Rep No. 96-1307 (1980), available at http://ipmall.info/hosted-

resources/lipa/patents/House_and_Senate.pdf

^{^[5]} Title 28, United States Code, ' 1295 (2005). In addition to patent appeals from the United States District Courts, the United States Court of Appeals for the Federal Circuit hears appeals from the United States Board of Contract Appeals, the United States Court of International Trade, the United States Court of Federal Claims, the United States Court of Appeals for Veterans Claims, the Department of Veteran Affairs, the International Trade Commission, the Merit Systems Protection Board, the United States Office of Compliance of the United States Congress, the Patent and Trademark Office (“PTO”), as well as from decisions of the Secretary of Agriculture, the Secretary of Commerce, and the Office of Personnel Management. Id.

^{^[6]}Markman v. Westview Instruments, Inc., 517 U.S. 370 (1996).

^{^[7]}Markman v. Westview Instruments, Inc., 52 F.3d 967 (Fed. Cir. 1995).

^{^[8]}136 F.3d 1448, 1473 (Fed. Cir. 1998) (Rader, J., concurring in part).

^{^[9]}Id. at 1476.

^{^[10]}Kimberly A. Moore, Are District Court Judges Equipped to Resolve Patent Cases?, 15 Harv. J.L. & Tech 1,2 (2001) [hereinafter Are districts Court Judges Equipped]

^{^[11]}Christian A. Chu, Empirical Analysis of the Federal Circuit’s Claim Construction Trends, 16 Berkeley Tech. L.J. 1075, 1092 (Fall 2001).

^{^[12]} Id. at 1104.

^{^[13]} See e.g., Gretchen Ann Bender, Uncertainty and Unpredictability in Patent Litigation: The Time is Ripe for Consistant Claim Construction Methodology, 8 J. Intell. Prop. L. 175, 206-07 (Spring 2001) (40% reversal rate from Markman opinion through 2000); Kimberly A. Moore, Markman Eight Years Later: Is Claim Construction More Predictable?, 9 Lewis & Clark L. Rev. 231,236-39 (2005) (a 40.8% reversal rate if summary affirmances are excluded, a 34.51% if summary affirmances are included, and a 37.5% reversal rate in cases in which the Federal Circuit held one or more patent claims wrongly construed); Andrew T. Zidel, Patent Claim Construction in the Trial Courts: A Study Showing the Need for Clear Guidance from the Federal Circuit, 33 Seton Hall L. Rev. 711, 745-46 (2003) (41.5% reversal rate (39 out of 94) of district court claim construction decisions during 2001).

^{^[14]} U.S. Court of Appeals for the Federal Circuit-Appeals Filed, Terminated, and Pending tbl. B-8 (2006), available at http://www.fedair.gov/pdf/ao0906.pdf.

Table B-8.

U.S. Court of Appeals for the Federal CircuitBAppeals Filed, Terminated, and Pending

During the Twelve-Month Period Ended September 30, 2006

Terminations

Pending By Percent Pending

Source of Appeals 01-Oct-05 Filed Total Judges Other Reversed 30-Sep-06

. . .

U.S. District Courts 376 522 496 357 139 13 402

. . .

^{^[15]} U.S. Courts of Appeals—Appeals Terminated on the Merits, by Circuit, During the 12—Month Period Ending March 31, 2006; tbl. B-5 (2006), available at http://www.uscourts.gov/caseload2006/tables/B05Mar06.pdf.

^{^[16]} Janice M. Mueller, Taking “Inventory” After Markman: The Supreme Court Confirms a New Era in Patent Litigation, The Law Works (June 1996).

^{^[17]} Amgen v. Hoechst Marion Roussel, Inc., 469 F.3d 1039, 1040 (Fed. Cir. 2006) (denial of petition for rehearing en banc).

^{^[18]} Holderman, supra note 3, at 428.

^{^[19]} 618 F. Supp. 288 (N.D. Ill. 1985).

^{^[20]} Id. at 295.

^{^[21]} Dana Corp. v. Am. Precision Co., 827 F.2d 755 (Fed. Cir. 1987).

^{^[22]} See Admin. Office of the U.S. Courts, Federal Judicial Caseload Statistics, 36 tbl.C, 57 tbl.D (2006), available at http://www.uscourts.gov/caseload2006/tables/C00_Mar_06.pdf, http://www.uscourts.gov/caseload2006/tables/D00CMar06.pdf [hereinafter Federal Judicial Caseload Statistics].

^{^[23]} See id. at 44 tbl.C-2, available at http://www.uscourts.gov/caseload2006/tables/C02_Mar_06.pdf.

^{^[24]} See id.; Admin. Office of the U.S. Courts, Judicial Caseload Profile Report (2006), http://www.uscourts.gov/cgi‑bin/cmsd2006.pl.

^{^[25]} See Federal Judicial Caseload Statistics, supra note 19, at 36 tbl.C, 57 tbl.D, available at http://www.uscourts.gov/caseload2006/tables/C00_Mar_06.pdf, http://www.uscourts.gov/caseload2006/tables/D00CMar06.pdf.

^{^[26]} Id. at 36 tbl.C, 48 tbl. C-3, 57 tbl.D, available at http://www.uscourts.gov/caseload2006/tables/C00_Mar_06.pdf, http://www.uscourts.gov/caseload2006/tables/C03Mar06.pdf, http://www.uscourts.gov/caseload2006/tables/D00CMar06.pdf; Judicial Caseload Profile Report, Illinois Northern, available at http://www.uscourts.gov/cgi‑bin/cmsd2006.pl.

^{^[27]} Federal Judicial Caseload Statistics, supra note 19, at 44 tbl.C-2; United States District Court for the Northern District of Illinois, Civil Case Report (Jan. 1, 2006 to Jan. 1, 2007), CME/CF Live, available at https://ecf.ilnd.circ7.dcn/cgi‑bin/login.pl?686576583596168‑L_835_0‑1.

^{^[28]} Holderman, supra note 3, at 429.

^{^[29]} Holderman, supra note 3, at 429.

^{^[30]} See supra notes 7-10 and accompanying text.

^{^[31]} Phillips v. AWH Corp., 415 F.3d 1303, 1328 (Fed. Cir. 2005) (en banc); see Kevin J. Culligan & John P. Hanish, (Im)pure Questions about Claim Construction in Patent Cases after Phillips, 6/06 Metropolitan Corp. Couns. 9 (June 2006); Steve Pollinger, A Battle for Definitional Supremacy in Patent Language Disputes, Tex. Law. Sept. 27, 2004.

^{^[32]} Brainy Quote, Rodney Dangerfield Quotes, http://www.brainyquote.com/quotes/quotes/r/rodneydang165663.html (last visited Mar. 14, 2007).

^{^[33]} Merck & Co. v. Tevo Pharms. USA, Inc., 395 F.3d 1364, 1370 (Fed. Cir. 2005).

^{^[34]} Markman v. Westview Instruments, Inc., 52 F.3d 967, 977-979 (Fed. Cir. 1995).

^{^[35]} Markman v. Westview Instruments, Inc., 517 U.S. 370, 376 (1996).

^{^[36]} See id.

^{^[37]} 138 F.3d 1448, 1456 (Fed. Cir. 1998) (en banc).

^{^[38]} Analyzing the positions expressed by the individual Circuit Judges in the November 22, 2006 order denying a petition to hear Amgen, 469 F.3d at1040-46, it appears a majority of the Federal Circuit Judges would consider the issue again.

^{^[39]} 90 F.3d 1576 (Fed. Cir. 1996)

^{^[40]} 182 F.3d 1298 (Fed. Cir. 1999)

^{^[41]} Id. at 1308.

^{^[42]} See, e.g., Johnson Worldwide Assocs. v. Zebco Corp., 175 F.3d 985 (Fed. Cir. 1999); Rexmond Corp. V. Laitram Corp., 274 F.3d 1336 (Fed. Cir. 2001).

^{^[43]} 308 F.3d 1193 (Fed. Cir. 2002)

^{^[44]} Id. at 1203.

^{^[45]} Id. at 1203-04.

^{^[46]} 415 F.3d 1303 (Fed Cir. 2005).

^{^[47]} Id. at 1321.

^{^[48]} See generally id.

^{^[49]} Medrad, Inc. v. MRI Devices Corp., 401 F.3d 1313, 1318 (Fed. Cir. 2005); PIN/NIP, Inc. v. Platte Chem. Co., 304 F.3d 1235, 1244 (Fed. Cir. 2002).

^{^[50]} See Cheryl L. Johnson, Why Judges are Destined to Flunk Their Markman Tests: The History of Their Claim Construction Assignment, 873 PLI/Pat 9, 58 (2006).

^{^[51]} Id. at 57-60.

^{^[52]} Id. at 61.

^{^[53]} Savatore Anastasi and Kevin Alan Wolff, Am. Intellectual Prop. Law Ass’n, Report of the Economic Survey 2005, I-108 to I-110 (Sept. 2005).

^{^[54]} Ray Black, Never Mind the Quality, Feel the Pinch, Managing Intell. Prop., May 1, 2005, at 26.

^{^[55]} Id. at Chart 1*need page number.

^{^[56]} See Moore, supra note 7 at 1-2.

^{^[57]} Cybor Corp. v. FAS Techs., Inc., 138 F.3d 1473, 1475-76 (Fed. Cir. 1998) (J. Rader, concurring in part); see also Moore, Are District Court Judges Equipped, supra note 7, at 14.

^{^[58]} See Brian Banner, ed., Claim Interpretation Proceedings and Appellate Review, Comm. No. 601, Subcomm. A, in Section of Intellectual Property 2002-2003 Annual Report 2 (2002-2003), available at http://www.abanet.org/intelprop/annualreport05/content/02‑03/COMMITTEE%20NO%20601.pdf. 78% of all claim construction hearings occur after discovery but before trial. 16 % of hearings are conducted prior to or during discovery, while only 8% are held before discovery. The smallest percentage of hearingsC 6%Ctake place at trial. Id.

^{^[59]} See Moore, supra note 7, at 2-3.

^{^[60]} See John B. Pegram, Should the U.S. Court of International Trade Be Given Patent Jurisdiction Concurrent with that of the District Courts?, 32 Hous. L. Rev. 67, 76 (1995).

^{^[61]} See Stuart Minor Benjamin & Arti K. Rai, Who’s Afraid of the APA? What the Patent Law System Can Learn from Administrative Law, 95. Geo. L.J. 269, 270 (2007); Johnson, supra note 47, at 70-71.

^{^[62]} Patient [sic] Overhaul; Hearing Before the Subcomm. on Courts, the Internet, and Intellectual Property of the H. Comm. on the Judiciary, 109th Cong. (2005) (statement of Richard Levin, President, Yale), 2005 WL 1010525; Benjamin & Rai, supra note 39, at 270; Johnson, supra note 29, at 70-71; John R. Thomas, The Responsibility of the Rulemaker: Comparative Approaches to Patent Administration Reform, 17 Berkeley Tech. L.J. 727, 728 (2002).

^{^[63]} Thomas, supra note 40, at 728.

^{^[64]} See Priya Ganapati, Patent Law Reform Proposed, Inc.com, May 10, 2005, http://www.inc.com/criticalnews/articles/200505/patents.html.

^{^[65]} H.R. 2795; 109th Cong. (1stSess. 2005), available at http://www.thomas.gov/cgi-bin/query/z?c109:H.R.2795.IH: (last visited Mar. 14, 2007).

^{^[66]} David B. Cochran, Toward a Post-Grant Patent Opposition Proceeding in the United States, Mondaq Bus. Briefing, Apr. 21, 2006, available at http://www.mondaq.com/article.asp?articleid=39280 .

^{^[67]} H.R. 2795, 109th Cong. § 336(a)(2) (1stSess. 2005), available at http://www.thomas.gov/cgi-bin/query/z?c109:H.R.2795.IH: (last visited Mar. 14, 2007); Cochran, supra note 45.

^{^[68]} Johnson, supra note 29, at 70.

^{^[69]} Steven R. Ludwig, A U.S. Post-Grant Patent Opposition System--Will it be Used?, Intell. Prop.Today, March, 2005, at 13-14.

^{^[70]} Are District Court Judges Equipped, supra note 7, at 27-28.

^{^[71]} Ludwig, supra note 48, at 14.

^{^[72]} Id.

^{^[73]} Cochran, supra note 45.

^{^[74]}