HERBERT MARKMAN and POSITEK, INC., Petitioners, v. WESTVIEW INSTRUMENTS, INC. and ALTHON ENTERPRISES, INC., Respondents.

No. 95-26

1995 U.S. Briefs 26

October Term, 1995

November 9, 1995

On Writ of Certiorari to the United States Court of Appeals for the Federal Circuit

BRIEF FOR EXXON CORPORATION, EXXON CHEMICAL PATENTS, INC., AND EXXON RESEARCH AND ENGINEERING COMPANY AS AMICI CURIAE IN SUPPORT OF PETITIONERS

DONALD B. CRAVEN, Counsel of Record, GERALD GOLDMAN, JAMES P. TUITE, JAMES R. LOVELACE, MILLER & CHEVALIER, Chartered, Metropolitan Square, 655 Fifteenth Street, N.W., Washington, D.C. 20005, (202) 626-5800

INTEREST OF THE AMICI CURIAE

The amici curiae hold, directly or indirectly, numerous patents of critical importance to their businesses, and they frequently have become involved in substantial, complex patent infringement suits, including one recently decided by the Federal Circuit. As a consequence, they have vital interests in preserving the integrity of the patent system and in ensuring that sound constitutional and jurisprudential principles govern the litigation of patent disputes. Those interests are directly affected by the court of appeals' broad rulings in the decision under review.

In that decision, a majority of the Federal Circuit, sitting en banc, held that disputes over the literal meaning and scope of patent claims give rise solely to questions of law, involve no factual issues implicating a litigant's Seventh Amendment rights, and are subject to de novo review on appeal. Those principles, the majority proclaimed, not only are consistent with the Seventh Amendment, but also will serve the interests of both patent holders and their competitors by promoting predictable and accurate claim construction. But the amici believe that precisely the opposite is the case. The principles adopted by the Federal Circuit are at odds with the requirements of the Seventh Amendment and are inimical to the just and predictable adjudication of patent disputes.

Counsel for the parties have consented to the filing of this brief, as reflected by the letters filed with the Clerk of the Court.

INTRODUCTION AND SUMMARY OF ARGUMENT

Patent infringement suits for damages, it is unanimously agreed, were decided by juries in 18th Century English courts of law. In fact, the Federal Circuit conceded below that, in light of that history, the Seventh Amendment provides a right to a jury trial in infringement actions for money damages. Pet. App. 44a. The court of appeals proceeded, however, to eviscerate that constitutional right by holding that "part of the infringement inquiry, construing and determining the scope of claims in a patent, is strictly a legal question for the court." Id. That holding goes a long way toward "ejecting juries from infringement cases" because it is generally recognized that "to decide what the claims mean is nearly always to decide the case." Id. at 57a (Mayer, J., concurring in the judgment).

As convincingly shown in the certiorari petition (at 13-17) and in Judge Mayer's concurrence (Pet. App. 65a-68a) and Judge Newman's dissent (id. at 115a-35a), the decision of the majority below clashes directly with this Court's Seventh Amendment jurisprudence. Because juries in 18th Century England decided analogous issues bearing on the scope of patents in infringement cases and similar causes of action, the Seventh Amendment preserves the right to have a jury adjudicate all of the controverted factual issues that arise in construing a patent claim. While the amici agree with that historical Seventh Amendment analysis, it is not the topic of this brief. Rather, the focus here is on additional reasons for overturning the Federal Circuit's decision.

The amici show below that the essential premise underlying the majority's decision--namely, that patent construction gives rise only to issues that are purely legal in nature--is wrong as a matter of logic and precedent. As decisions by this Court and other courts show, the construction of patent claims necessarily entails factual inquiries that frequently give rise to disputed issues that can be answered only by resort to trial testimony and other extrinsic evidence. The amici also show below that subjecting the trial court's findings on these issues to de novo review ignores the inherent limitations of appellate courts and undermines the accurate and predictable adjudication of infringement actions.

I.

A. Although the courts have traditionally viewed claim construction as raising, in an overall sense, a legal question, it does not follow that disputed factual issues are completely absent from the interpretation of a patent. The test to be applied in determining the scope and meaning of a patent requires a determination "of what one of ordinary skill in the art at the time of the invention would have understood the terms to mean." Id. at 48a. This test entails a specific inquiry into inherently factual matters, such as the level of ordinary skill in the relevant technical or scientific community and the meaning of certain terminology to that community at a particular point in time. Those matters often are the subject of dispute, the resolution of which properly turns on the assessment of expert testimony and other extrinsic evidence by a finder of fact. Harries v. Air King Prods. Co., 183 F.2d 158, 164 (2d Cir. 1950) (L. Hand, J.) (describing the inquiry as "plainly a question of fact" subject to "clearly erroneous" appellate review); see also In re Mahurkar Double Lumen Hemodialysis Catheter Patent Litig., 831 F. Supp. 1354, 1359 (N.D. Ill. 1993) (Easterbrook, J.).

B. The effort of the majority below to classify all aspects of claim construction as legal in nature squarely conflicts with this Court's teachings in two cases involving the test of non-obviousness, a condition of patentability. See 35 U.S.C. � 103. Although "the ultimate question" of non-obviousness may be a legal one, it involves several "subsidiary determinations" of a factual nature that are reviewed on appeal under "the clearly-erroneous standard." Dennison Manufacturing Co. v. Panduit Corp., 475 U.S. 809, 811 (1986). One of those subsidiary "factual inquiries" is "the level of ordinary skill in the pertinent art." Graham v. John Deere Co., 383 U.S. 1, 17 (1966). Because the identical inquiry is made in the claim construction context, Dennison and Graham compel reversal here.

C. While conceding that claim construction will often require the consideration of expert testimony and other extrinsic evidence, the majority below refused to come to grips with the logical corollary that the admission of such evidence unavoidably creates the opportunity for factual disputes over the understanding of those skilled in the art. It is pure fantasy to pretend that the trial court, in assessing the quality of such evidence, does not engage in classic fact-finding activities--evaluating credibility and weighing the relative strength and probity of the evidence.

D. The majority's analogy of claim construction to statutory interpretation is ill-conceived. Statutory interpretation requires judges to construe the language of the statute and, in certain circumstances, to analyze other legislative materials--an activity that they are fully equipped to perform without hearing expert testimony or other extrinsic evidence. By contrast, a court cannot determine on its own the meaning of a highly technical patent to one of ordinary skill in the art at a particular point in time. If that meaning is disputed by the parties, the court must hear evidence and engage in fact-finding to resolve the dispute.

II.

The new legal regime for deciding infringement cases ushered in by the decision below will upset the proper roles of trial and appellate courts, undermine the accuracy of claim interpretations, and weaken the stability of the patent system. The blanket declaration that all aspects of claim construction involve exclusively legal issues improperly liberates the Federal Circuit to reformulate claim interpretations without regard to the evidence introduced at trial showing how one skilled in the art would have understood the patent at the relevant point in time. In any event, the inferences to be drawn and the findings to be made based on the trial evidence are the domain of the fact finder and should be reviewed on appeal under a deferential standard. In light of the often complicated and technical nature of the testimony in patent litigation, appellate review on a cold written record provides an especially inadequate basis for assessing the evidence.

According to the majority, patent holders and their competitors will benefit from interpretations by judges "trained in the law" applying "established rules of construction." Pet. App. 28a-29a. But legal training and rules of construction will not answer how one of ordinary skill in the art at the relevant time would have understood the patent claim. That can be determined only as a result of a factual analysis of the relevant evidence. Moreover, by heightening the prospects of judicially devised interpretations unanchored in the evidence regarding the understanding of those skilled in the art, the majority below has injected greater uncertainty into the patent system.

ARGUMENT

I. UNDER THE TRADITIONAL AND AGREED-UPON TEST FOR CONSTRUING PATENTS, THE INTERPRETATION OF CLAIMS FREQUENTLY GIVES RISE TO DISPUTES OVER UNDERLYING FACTUAL ISSUES THAT MUST BE RESOLVED BY THE TRIER OF FACT AND REVIEWED ON APPEAL UNDER A DEFERENTIAL STANDARD

A. Notwithstanding the Legal Nature of the Ultimate Claim Construction Question, Legitimate Disputed Factual Issues Will Arise

This litigation involves an action for patent infringement in which the plaintiffs seek money damages. As the Federal Circuit "has repeatedly said, a determination of patent infringement is a two-step analysis." Genentech, Inc. v. Wellcome Foundation Ltd., 29 F.3d 1555, 1561 n.6 (Fed. Cir. 1994); accord Pet. App. 20a-21a; Lemelson v. General Mills, Inc., 968 F.2d 1202, 1206 (Fed. Cir. 1992), cert. denied, 113 S. Ct. 976 (1993). "The first step" of the analysis is "to construe the [patent] claims," Morton International, Inc. v. Cardinal Chemical Co., 5 F.3d 1464, 1468 (Fed. Cir. 1993), in order "to determine [their] proper scope and meaning." Genentech, 29 F.3d at 1561 n.6. The second step requires a determination "whether an accused device or process is within the scope of the properly interpreted claim." Id.

The second analytical step--the infringement determination--indisputably raises a question of fact. Winans v. Denmead, 56 U.S. (15 How.) 330, 337 (1854); Hilton Davis Chemical Co. v. Warner-Jenkinson Co., 62 F.3d 1512, 1520 (Fed. Cir. 1995) (en banc), petition for cert. filed, -- U.S.L.W. -- (U.S. Nov. 6, 1995) (No. 95-728). The issue presented here, however, is whether the first "part of the infringement inquiry, construing and determining the scope of the claims in a patent," may present factual issues requiring jury determinations under the Seventh Amendment. Pet. App. 44a. That step in the process, the majority below broadly held, never gives rise to such factual questions because it purportedly presents "strictly a legal question for the [trial] court" subject to de novo review on appeal. Id. & n.13.

To reach those conclusions, the majority was forced to overrule numerous precedents. Before the decision below, the Federal Circuit had repeatedly held that, "when the meaning of a term in a claim is disputed and extrinsic evidence is necessary to explain that term, then an underlying factual question arises, and the construction of the claim should be left to the trier or jury under appropriate instruction." Palumbo v. Don-Joy Co., 762 F.2d 969, 974 (Fed. Cir. 1985); accord Pet. App. 59a-61a (Mayer, J., concurring in the judgment) (citing cases); id. at 135a-44a (Newman, J., dissenting) (citing cases); see also id. at 21a-25a (majority opinion acknowledging inconsistency in Federal Circuit precedents).

The majority's revamped approach to claim construction is fundamentally unsound. While claim construction is ultimately a question of law, underlying factual issues must be resolved in interpreting the scope and meaning of a patent. That conclusion flows directly from the standard for claim construction, which is not in dispute. That standard establishes a "test of what one of ordinary skill in the art at the time of the invention would have understood the term to mean." Id. at 48a; cf. 35 U.S.C. � 112 ("The specification shall contain a written description of the invention . . . in such full, clear, concise, and exact terms as to enable any person skilled in the art . . . to make and use the same . . . ."). By its terms, the test requires an inquiry into how particular persons at a particular point in time would have understood the patent claim.

As the majority below acknowledged, a judge "is not the hypothetical person skilled in the art to whom a patent is addressed" and "is not usually a person conversant in the particular technical art involved." Pet. App. 50a. Citing numerous decisions of this Court and other courts, the majority accordingly conceded that it will at times be necessary to hear expert testimony and other extrinsic evidence to decide an issue of claim construction. Id. at 31a, 33a-36a, 50a. Indeed, the trial court may "use certain extrinsic evidence that the court finds helpful" and "reject[] other evidence as unhelpful . . . en route to pronouncing the meaning of claim language as a matter of law." Id. at 36a. In short, as recognized by the Federal Circuit:

The purpose of expert testimony is to provide assistance to the court in understanding, when the claims are technologically complex or linguistically obscure, how a technician in the field, reading the patent, would understand the claims.

Id. at 35a, quoting Advanced Cardiovascular Sys., Inc. v. Scimed Life Sys., Inc., 887 F.2d 1070, 1076 (Fed. Cir. 1989) (dissenting opinion) (emphasis omitted).

This approach to construing patent claims often requires an analysis of complex science or engineering questions, as well as an appreciation of the technical terminology used in the discipline. To resolve those types of questions, judges need to consider the testimony of experts and other extrinsic evidence.

Judge Learned Hand, "one of the Nation's great patent judges" (Gerald Gunther, Learned Hand: The Man and the Judge 138 (1994)), highlighted this characteristic of patent litigation in Harries v. Air King Products Co., 183 F.2d 158 (2d Cir. 1950). In that case, the trial court had to determine as part of the infringement inquiry "how the art understood" certain patent claims. Id. at 164. That issue--which in Judge Hand's view involved "plainly a question of fact" subject to "clearly erroneous" review on appeal--was one that the court was "altogether incompetent to decide" since "even the terminology [was] beyond [its] acquaintance." Id. For that reason, expert testimony was indispensable to a proper adjudication, notwithstanding the inherent imperfections of partisan testimony. He stated: "While Congress sees fit to set before us tasks which are so much beyond our powers, suitors must be content that we shall resort to the testimony of experts, though they are concededly advocates with the inevitable bias that advocacy engenders." Id. Thus, expert testimony plays an essential role in achieving a fair and accurate adjudication in patent cases involving complex subjects and technical terminology beyond the ken of judges.

More recently, Judge Easterbrook, sitting as the trial judge in a patent infringement case, struck a similar theme. He noted that, although the ultimate scope of a patent claim may be "an issue of law," "judges should not pretend that all nominally 'legal' issues may be resolved without reference to facts." In re Mahurkar Double Lumen Hemodialysis Catheter Patent Litig., 831 F. Supp. 1354, 1359 (N.D. Ill. 1993). Because patent claims "are not self-defining," it would be "folly to suppose" that their "language," particularly in scientifically or technologically complex areas, would have "only [one] possible understanding" to both a judge and one skilled in the art. Id. "What seems clear to a judge may read otherwise to a skilled designer. That is why," Judge Easterbrook explained, "we have . . . trial[s]" in these types of cases. Id.

Thus, while claim construction may be considered, in an ultimate sense, to be a legal issue, disputed fact questions will arise during the interpretive process that can be resolved only on the basis of expert testimony or other extrinsic evidence. Those disputed issues must be resolved by a trier of fact, whose findings may properly be considered on appeal only under a clearly erroneous or similarly deferential standard of review.

This is precisely the teaching of this Court in two closely analogous cases. In Graham v. John Deere Co., 383 U.S. 1 (1966), the defendant in an infringement action argued that the patent was invalid because it failed to satisfy the test of non-obviousness, which is one of the conditions of patentability. See 35 U.S.C. � 103. That test, expressed in language strikingly similar to the test at issue here, requires a determination whether "the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art." Id. (emphasis added). While this Court characterized that question, like claim construction, as "one of law," it also explained that the non-obviousness test "lends itself to several basic factual inquiries"--"the scope and content of the prior art are to be determined; differences between the prior art and the claims at issue are to be ascertained; and the level of ordinary skill in the pertinent art resolved." Graham, 383 U.S. at 17. In Dennison Manufacturing Co. v. Panduit Corp., 475 U.S. 809, 811 (1986), the Court elaborated that, regardless of whether "the ultimate question" is legal or factual, the "subsidiary determinations" that must be made under the test for non-obviousness are undeniably factual in nature and should be reviewed on appeal under "the clearly-erroneous standard."

This Court's decisions in Graham and Dennison are controlling here. They hold that the level of ordinary skill in the pertinent art is a question of fact even if it arises under what is ultimately a question of law. That same inquiry is an integral element of claim construction, because concededly claim construction must be conducted from the viewpoint of one of ordinary skill in the art at the relevant time. It necessarily follows, therefore, that interpreting a patent claim involves the resolution of a factual issue, which may itself become the subject of dispute. Graham and Dennison thus contradict the Federal Circuit's premise that claim construction does not involve factual issues, thereby requiring reversal of the decision below.

B. The Federal Circuit's Efforts to Explain Away the Existence of Factual Issues in Claim Construction Are Unpersuasive

The Federal Circuit's refusal to recognize any factual dimension to claim construction rests on two propositions. First, the majority reasoned that extrinsic evidence in patent construction, even though necessary, does not give rise to any genuine factual dispute because it merely "inform[s] the court about the language in which the patent is written" and that the court's consideration of such evidence does not involve the "crediting [of] certain evidence over other evidence." Pet. App. 36a, 50a. Second, the majority analogized claim construction to statutory interpretation--"a matter of law strictly for the court." Id. at 51a. Neither of these propositions withstands analysis.

1. As noted above, the majority itself acknowledged the need for extrinsic evidence to address the standard of how the patent was understood by those skilled in the art at the relevant time. But the decision below fails to recognize the logical implications of that concession--namely, that factual disputes will arise that can properly be resolved only by a trier of fact, subject to a deferential standard of review. The majority dismisses the potential for factual disputes on the ground that extrinsic evidence plays an informational role only; it is needed, the majority asserts, because of the "unfamiliarity of the court with the terminology of the art to which the patent is addressed." Id. at 50a. Thus, under the majority's view, witnesses providing extrinsic evidence in a patent infringement case are nothing more than talking dictionaries instructing the court as to the proper meaning of the terms used in a patent claim.

This is a crabbed and unrealistic view of the role of extrinsic evidence in the claim construction process. Extrinsic evidence is necessary because the touchstone for interpreting the claim is the understanding of a person "of ordinary skill in the art at the time of the invention." Id. at 48a. Such evidence is not limited to the abstract interpretation of the specific terms in which the claim is cast but instead encompasses proof of "how a technician, in the field, reading the patent, would understand the claims." Id. at 35a (internal quotation marks omitted). Moreover, the evidence is presented by interested participants in the adversary process and not as a neutral tutorial. It is inevitable, therefore, that the parties' presentations will be conflicting and that disputes will arise regarding the understanding of those skilled in the art.

The majority's assertion that, in considering the parties' conflicting evidence, the trial court "is not crediting certain evidence over other evidence or making factual evidentiary findings" simply defies reality. Id. at 36a. The majority acknowledged that the trier of fact must consider conflicting testimony and assess what is "helpful" and what is "unhelpful." Id. When it engages in that activity, what else can it be doing other than "crediting certain evidence over other evidence"? For example, testimony elicited on cross-examination of the defendant's witnesses in an infringement action that supports the plaintiff's construction of the claim would likely be credited as "helpful." On the other hand, testimony that is self-contradictory or impeached by documentary evidence may be rejected as "unhelpful." And the witness's demeanor and similar considerations may well provide important guidance in determining whether his or her testimony should be regarded as "helpful." Such judgments in a patent case are no different from those that a trier of fact would make in resolving any other kind of factual issue.

Indeed, "the jury . . . weighs the contradictory evidence and inferences, judges the credibility of witnesses, receives expert instructions, and draws the ultimate conclusion as to the facts. The very essence of its function is to select from among conflicting inferences and conclusions that which it considers most reasonable." Tennant v. Peoria & P.U. Ry. Co., 321 U.S. 29, 35 (1944); see also Inwood Labs., Inc. v. Ives Labs., Inc., 456 U.S. 844, 856 (1982) ("Determining the weight and credibility of the evidence is the special province of the trier of fact."); Rogers v. Missouri Pac. R.R., 352 U.S. 500, 504 (1957).

The decision below simply ignores these litigation realities. Judge Schwartz of the District of Delaware aptly reacted to the majority's description of the trial court's treatment of extrinsic evidence as being "somewhat startling" and creating a "fiction." Lucas Aerospace, Ltd. v. Unison Indus., L.P., 890 F. Supp. 329, 333-34 n.7 (D. Del. 1995), modified on other grounds, No. 93-525, 1995 U.S. Dist. LEXIS 13361 (D. Del. Sept. 5, 1995). Judge Schwartz explained:

As I understand Markman, because claim construction presents a purely legal question, trial judges must ignore all non-transcribible courtroom occurrences such as a witness's body language, inability to maintain eye contact when confronted with a telling question, hesitance or delay in giving an answer, an affirmative answer in a voice revealing the truthful answer is "no," or the changing demeanor of a witness when shifting from sure to treacherous footing. . . . [But] when two experts testify differently as to the meaning of a technical term, and the court embraces the view of one, the other, or neither while construing a patent claim as a matter of law, the court has engaged in weighing evidence and making credibility determinations.

Id. at 333 n.7.

In short, contrary to the Federal Circuit's assertions, where a trial court hears extrinsic evidence to understand the meaning of the patent claim to particular persons at a particular point in time and chooses to rely on some of the evidence and to disregard other evidence, it is not making legal determinations that can be reviewed de novo on appeal. Instead, it is engaging in a classic exercise of fact-finding. See Lavender v. Kurn, 327 U.S. 645, 652- 53 (1946).

2. The second analytical linchpin of the decision below is its analogy of claim construction to statutory interpretation. See Pet. App. 51a-52a. That analogy, however, is wide of the mark.

When possible, this Court uses the "plain meaning" approach to statutory construction and confines itself to interpreting the statutory text. See, e.g., City of Chicago v. Environmental Defense Fund, 114 S. Ct. 1588, 1591-93 (1994). Even on those occasions when the Court goes beyond the statutory text to consider legislative history in the form of Congressional reports and other documents, the materials subject to examination are equally available in the trial and appellate courts and can be best construed by generalist judges, whose training and experience qualify them for that task. See, e.g., Allied-Bruce Terminix Cos. v. Dobson, 115 S. Ct. 834, 839-40, 842 (1995).

This is not at all the situation when a judge must determine what a complex technical patent meant to one skilled in the art at a particular point in time. That exercise requires, in the words of Judge Hand, "resort to the testimony of experts," who may differ regarding the existence of primary facts, or to other extrinsic evidence such as articles and other scientific writings, which may give rise to conflicting inferences. Harries, 183 F.2d at 164.

Thus, the majority's analogy to statutory construction is no more persuasive than its analysis of the trial court's use of extrinsic evidence. Neither justification succeeds in transmogrifying a question of fact into a question of law.

II. THE APPROACH TO CLAIM CONSTRUCTION ADOPTED BELOW IMPROPERLY VESTS THE FACT FINDER'S ROLE IN THE FEDERAL CIRCUIT AND JEOPARDIZES THE ACCURATE AND CERTAIN PROTECTION OF PATENTS

Under the decision below, at least as applied in subsequent cases, the Federal Circuit effectively has taken responsibility for making all of the factual determinations that are necessary to construe patent claims. The majority accomplished this by re-labeling factual determinations as "legal" issues that are reviewable de novo. Pet. App. 44a & n.13. As a result, panels of appellate judges have asserted free rein to second guess determinations made at the trial level that were based on an evaluation of conflicting testimony and other evidence and that, accordingly, have been traditionally subject only to deferential review.

Although Congress intended in establishing the Federal Circuit to centralize appeals of patent infringement cases, it did not intend that tribunal to perform the function of trial courts. Indeed, Congress specifically recognized that "factual issues in a patent case must be tried and decided by the trial judge or a jury in precisely the same manner as such issues are tried in any other kind of a lawsuit. The technical aspects of a patent case are factual issues . . . . Thus, it is the settled practice of the circuit courts of appeals in patent cases to honor and respect" the findings of fact below. H.R. Rep. No. 312, 97th Cong., 1st Sess. 37 (1981). Under the Court of Appeals for the Federal Circuit Act of 1981, this practice was not to change. Id. at 37-38.

The Federal Circuit's assumption of the responsibilities properly performed by the trier of fact disserves the fair and accurate adjudication of patent infringement claims. As this Court has recognized, a trier of fact who has heard the witnesses and seen the other extrinsic evidence first-hand is better equipped to make factual findings than an appellate court, which must rely upon the cold written record from the court below. "Only the trial judge can be aware of the variations in demeanor and tone of voice that bear so heavily on the listener's understanding of and belief in what is said." Anderson v. City of Bessemer City, 470 U.S. 564, 575 (1985). Moreover, "the district court may have insights not conveyed by the record into such matters as whether particular evidence was worthy of being relied upon." Pierce v. Underwood, 487 U.S. 552, 560 (1988); see also First Options of Chicago, Inc. v. Kaplan, 115 S. Ct. 1920, 1926 (1995).

It is precisely because trial courts are better suited than courts of appeals to engage in fact-finding that appellate courts owe deference to factual findings made at the trial level. Indeed, the "clearly erroneous" standard of Fed. R. Civ. P. 52(a) "recognizes and rests upon the unique opportunity afforded the trial court judge to evaluate the credibility of witnesses and to weigh the evidence." Inwood Labs., 456 U.S. at 855 (citing Zenith Radio Corp. v. Hazeltine Research, Inc., 395 U.S. 100, 123 (1969)); see also Graver Tank & Mfg. Co. v. Linde Air Prods. Co., 336 U.S. 271, 275 (1949) (noting "the advantages possessed by the trial court in appraising the significance of conflicting testimony"). Thus, "appellate courts must constantly have in mind that their function is not to decide factual issues de novo." Zenith, 395 U.S. at 123; see also First Options, 115 S. Ct. at 1926.

The differentiation in the respective roles played by the fact finder and appellate court is particularly vital in complex matters like patent infringement cases, in which there often is conflicting expert testimony concerning difficult and intricate issues of fact. Thus, this Court has properly emphasized in patent cases--"where the evidence is largely the testimony of experts as to which a trial court may be enlightened by scientific demonstrations"--that an appellate court is not "to even essay an independent evaluation" of the expert testimony, which "is the function of the trial court" in its capacity as the finder of fact. Graver Tank & Mfg. Co. v. Linde Air Prods. Co., 339 U.S. 605, 611 (1950); see also id. at 610. As Judge Newman succinctly explained in her dissenting opinion below, "when the technologic issues are complex, appellate fact finding is probably the least effective path to accurate decisionmaking . . . . Appellate briefs and fifteen minutes per side of attorney argument are not designed for de novo findings of disputed technologic questions." Pet. App. 85a; accord Henry J. Friendly, Federal Jurisdiction: A General View 157 (1973).

At bottom, the effort below to reclassify all claim construction issues as questions of law subject to de novo review reflects a policy judgment that patent cases are best left in the hands of judges "trained in the law." Pet. App. 28a-29a. According to the majority, judges applying "established rules of construction to the language of the patent claim" are more likely than trial fact finders to "arrive at the true and consistent scope of the patent owner's right to be given legal effect." Id.

The majority's approach, however, is misguided because claim construction does not and cannot turn simply on the application of judicially created canons. Canons of construction cannot answer the test of what one of ordinary skill in the art at the relevant time would have understood the patent to embrace. Moreover, that test is fraught with underlying issues that are factual in nature and thus can only be accurately resolved by finders of fact based on a consideration of the evidence. Contrary to the majority's approach, that requires a trial in the district court with only deferential review by the Federal Circuit. See Pall Corp. v. Micron Separations, Inc., 66 F.3d 1211, 1995 U.S. App. LEXIS 27366, *39 (Fed. Cir. Sept. 26, 1995) (Mayer, J., concurring) ("To suggest that appellate judges, precious few of whom are trained in science, will always arrive at the 'true' meaning of words embodying complex concepts endows them with knowledge and enlightenment far beyond those who have training and experience in the field. They are in no position to declare the state of knowledge in the art or that scientific hypotheses are correct as a matter of law.").

The majority's endorsement of judge-made interpretations rather than fact finder determinations based on the evidence will undermine not only the accuracy of claim interpretations but also the stability of the patent system. The test of the understanding of those of ordinary skill in the art has provided a guiding principle on which patent writers and readers have long been able to depend. While paying lip service to that principle, the ruling below does serious damage to its application in practice by papering over the factual disputes it raises. The result can only be uncertainty in the construction of countless patent claims already written and confusion about how to write patents for the future.

In fact, by virtue of the Federal Circuit's application of the decision below, trial court adjudications now stand a far greater chance of being overturned merely because an appellate panel, despite having seen and heard none of the witnesses, disagrees with the trial court's construction of the patent claim. Moreover, parties will view the appeal as a clean slate on which they can re-litigate all claim construction issues, thereby rendering the trial a superfluous exercise. As Judge Mayer stated in his opinion below, "the effect of this case is to make of the judicial process a charade, for notwithstanding any trial level activity, this court will do pretty much what it wants under its de novo retrial." Pet. App. 68a (concurring in the judgment); see also id. at 104a (Newman, J., dissenting) (decrying the "'omniscience of the learned man' theory of dispute resolution in the Federal Circuit").

The consequences of the decision below already are apparent. In Exxon Chemical Patents, Inc. v. Lubrizol Corp., 64 F.3d 1553 (Fed. Cir. 1995), a divided panel of the court of appeals applying the decision below engaged in unfettered de novo review of the district court's construction of a patent claim. The panel stated: "No matter when or how a judge performs the Markman task, on appeal we review the issue of claim interpretation independently without deference to the trial judge." Id. at 1556. The panel then adopted an interpretation of the claim that had been advanced by neither party at any point in the litigation nor by any witness in trial testimony. Id. at 1564-65 (Nies, J., dissenting). Instead, the court's decision was based solely on the appellate judges' parsing of the words of the claim without any regard to the extensive extrinsic evidence presented at trial concerning what those skilled in the art would have understood the claim to mean. See also, e.g., Hoover Group, Inc. v. Custom Metalcraft, Inc., 66 F.3d 299, 1995 U.S. App. LEXIS 26350, *14 (Fed. Cir. Sept. 19, 1995) (reversing trial court's claim construction based solely on patent claim, drawings, and specification, without addressing disputed extrinsic evidence); Mason v. Tampa G Mfg. Co., No. 95-1184, 1995 U.S. App. LEXIS 28368, *10-11 (Fed. Cir. Oct. 12, 1995) (construing patent based solely on panel's examination of patent claim, drawings, and specification without any ruling on construction by trial court and without reference to extrinsic evidence).

Finally, to the extent that the decision below is motivated by the fear of irrational jury decisions, trial judges have mechanisms at hand for dealing with groundless jury determinations. As this Court recently observed in another case involving scientific evidence:

In the event the trial court concludes that the scintilla of evidence presented supporting a position is insufficient to allow a reasonable juror to conclude that the position more likely than not is true, the court remains free to direct a judgment, Fed. Rule Civ. Proc. 50(a), and likewise to grant summary judgment, Fed. Rule Civ. Proc. 56.

Daubert v. Merrill Dow Pharmaceuticals, Inc., 113 S. Ct. 2786, 2798 (1993); see also id. (decrying "pessimism about the capabilities of the jury, and of the adversary system generally"); Curtis v. Loether, 415 U.S. 189, 198 (1974). Certainly, the use of such "conventional devices" is preferable to the "wholesale" assumption of fact-finding powers by the court of appeals. See Daubert, 113 S. Ct. at 2798.

CONCLUSION

The judgment of the court of appeals should be vacated and the case remanded for reconsideration in light of the petitioner's entitlement to a jury determination of the factual issues underlying claim construction and the deference owed to such determinations by the appellate court.

Respectfully submitted, DONALD B. CRAVEN, Counsel of Record, GERALD GOLDMAN, JAMES P. TUITE, JAMES R. LOVELACE, MILLER & CHEVALIER, Chartered, Metropolitan Square, 655 Fifteenth Street, N.W., Washington, D.C. 20005, (202) 626-5800

November 1995