CASRIP Newsletter - Spring 1995, Volume 2, Issue 2
Markman: The Federal Circuit's Long-Awaited En Banc Decision on Claim Interpretation as a Question of Law
On April 5, 1995, the Federal Circuit handed down Markman v. Westview Instruments, Inc., 34 USPQ2d 1321 (Fed. Cir. 1995) (en banc). Markman is a landmark decision on how the scope of protection of patents should be resolved in United States litigation. Many, indeed most, patent infringement disputes arise because the parties disagree on what a patent's claims mean and what the patent therefore covers. The dispute is not over the characteristics of the accused infringer's product or process but over how the patent claim should be interpreted (construed).
Before Markman, three-judge panel decisions of the Federal Circuit divided on whether patent claim interpretation was an issue of law or fact. The law-fact distinction has a major practical impact: if it is one of fact, any genuine dispute between the parties can be resolved only through the full course of litigation, including a jury trial if either party demands one. If the issue is one of law, a dispute can be resolved early in litigation, for example, through a motion for summary judgment. The distinction also controls the allocation of authority between trial court and appeals court. An appeals court freely reviews questions of law but must give substantial deference to a trial court's resolution of questions of fact.
Intertwined with the procedural question on claim interpretation (i.e. law versus fact), are substantive questions on claim interpretation. To what sources can, should or must a court look in deciding what critical words in a patent claim mean? The universe of sources can be tightly confined (e.g. only the words themselves together with any definition in the patent's specification) or broadly expanded (arguments and actions during the patent's prosecution; expert testimony on what the words mean to "persons of ordinary skill in the art"; the inventor's testimony as to what he or she invented and intended to claim).
Federal Circuit decisions before Markman approved the use of extrinsic evidence, that is, sources other than the patent itself and its prosecution history (the official written record in the Patent and Trademark Office), at least if claim language was "ambiguous" or a technical term in the art. The opportunity to present expert testimony meant that the parties (patentee and accused infringer) would likely procure and present witnesses expressing opposing viewpoints. Admitting extrinsic evidence suggested that claim interpretation was a question of fact because a characteristic of fact issues is that they are established by evidence. (The phrase "extrinsic evidence" comes from contract law; case law generally allows evidence on what the parties to a contract intended when the contract is ambiguous.) But treating claim interpretation as a fact issue subject to expert testimony and fact finding in litigation vastly increases uncertainty over a matter upon which the ability to plan and rely is of the utmost importance. Persons and companies make business decisions, as well as decisions about licensing and litigation, based on how they and their counsel interpret many patents held by others. For this reason, patent law requires that claim language be clear and definite.
In Markman , a solid majority (8 of the 11 judges), in an extensive opinion by Chief Judge Archer, resolved the conflicting lines of cases by holding that claim interpretation is a question of law. A judge may, in his or her discretion, admit "extrinsic evidence" as an aid to claim interpretation, but this does not create a fact issue.
The following three passages summarize the court's holding.
1. "Through [the] process of construing claims by, among other things, using certain extrinsic evidence that the court finds helpful and rejecting other evidence as unhelpful, and resolving disputes en route to pronouncing the meaning of claim language as a matter of law based on the patent documents themselves, the court is not crediting certain evidence over other evidence or making factual evidentiary findings. Rather, the court is looking to the extrinsic evidence to assist in its construction of the written document, a task it is required to perform. . . . The district court's claim construction, enlightened by such extrinsic evidence as may be helpful, is still based upon the patent and prosecution history. It is therefore still construction, and is a matter of law subject to de novo review."
2. "When, after considering the extrinsic evidence, the court finally arrives at an understanding of the language as used in the patent and prosecution history, the court must then pronounce as a matter of law the meaning of that language. . . . This ordinarily can be accomplished by the court in framing its charge to the jury, but may also be done in the context of dispositive motions such as those seeking judgment as a matter of law."
3. "[W]e review district court determinations on questions of claim construction under a de novo standard of review, like other legal questions. In this regard, . . . we are reiterating the long-recognized appellate review standard for issues of law in the trial proceeding, regardless of whether the case was tried to a judge or a jury."
Two judges (Mayer and Rader) wrote concurring opinions, agreeing that under the facts of Markman no genuine fact issue existed on claim interpretation but insisting that the constitutional right to jury trial should apply when a genuine fact dispute arises as to the meaning of a patent claim. Judge Newman wrote an extensive dissenting opinion, stressing both the importance of the constitutional right to jury trial and the practical problems in resolving disputes over "technologic facts" as questions of law.
The facts of the Markman case illustrate the impact of the court's holding. Markman's patent concerned a system for controlling inventory in a dry cleaning store. A dry cleaner takes in clothes items from a customer; separates them for sorting and combining with like items from other customers for cleaning, which entails movement among different locations; and, after cleaning, re-sorts and returns the items to the (one hopes) appropriate customer. The patent offered a solution to dry-cleaning inventory control problems, including loss of items and theft by employees who keep the proceeds from customers for items cleaned but not recorded. The disclosed embodiment involved placing a tag with a bar code on each item of clothing presented by a customer, optically scanning the codes at various points during the cleaning process, and using a computer to generate reports on location and number of items in the system.
The patent's claim required that the system be able to "detect and localize spurious additions to inventory as well as spurious deletions therefrom."
Markman charged Westview with infringement. Westview's equipment used bar- coding to generate invoices for each customer's order and a list of invoices. It could verify whether any item with an invoice was on the list, but it could not generate an inventory of individual clothes items.
The case was tried to a jury. To establish infringement, Markman offered his own testimony and that of a "patent expert" that "inventory" was not restricted to an inventory of clothes items and could include an inventory of invoices. Westview argued that the patent specification and prosecution history consistently used inventory to refer to clothes items and thus that there could be no infringement. The trial judge instructed the jury to resolve the dispute, and the jury rendered a verdict for Markman. On a post-trial motion, the judge overturned the jury verdict, ruling that, as matter of law, "inventory" referred to clothes items.
On appeal, the majority affirmed. The testimony by the inventor and patent expert did not create a fact issue for a jury. The patent claim language, the specification, and the prosecution history made clear that "inventory" meant clothes items.
The court's analysis indicates that in cases similar to Markman there will be no need to have a jury trial on infringement. The trial court could and should have resolved the matter before trial by ruling on a summary judgment motion.
Markman is being widely interpreted as a decision favoring accused infringers and restricting patentees' rights to force jury trials in patent infringement suits. The court's holding will undoubtedly make less credible the demands of patent holders who rely on unreasonable interpretations and who respond to well-founded noninfringement arguments by threatening to take the case to a jury that may not appreciate technical subtleties. But Markman can work in favor of patent holders with strong, clearly-worded patents. Interestingly, in the first Federal Circuit panel decision to apply Markman, the court held that a patent owner was entitled to a summary judgment of infringement based on the correct legal interpretation of its claims. Transmatic, Inc. v. Gulton Industries, Inc., 1995 WL 251622 (Fed. Cir., May 1, 1995).
Glaxo's Lucrative Second "Zantac" Patent Survives Best Mode, Inequitable Conduct and Anticipation Challenges--But Just Barely!
The Federal Circuit's April 21, 1995 decision, Glaxo Inc. v. Novopharm Ltd., 34 USPQ2d 1565 (Fed. Cir. 1993), illustrated once again how the validity of an extremely valuable patent can turn on close questions of fact, both scientific (the exact form of a compound a process produces) and behaviorial (whether a scientist intended to deceive in submitting to the Patent Office a declaration setting forth testing results.) The case also resolved an unsettled legal issue of great concern to companies and other large organizations seeking patents: does the statutorily required (35 U.S.C. Section 112) disclosure of the "best mode contemplated by the inventor" for carrying out an invention include modes preferred for commercial reasons by persons in the organization but not known or preferred by the individual inventor or inventors?
Glaxo's '431 patent concerned its commercially successful "Zantac" antiulcer medication, an aminoalkyl furan derivative ("ranitidine"). In 1978, Glaxo obtained a patent ('658) claiming ranitidine and its hydrochloride salt. The '658 patent disclosed a method for preparing ranitidine hydrochloride (Example 32) developed by Glaxo scientist David Collin. Between 1977 and 1980, Glaxo prepared ranitidine hydrochloride for toxicology and clinical studies using a method ("Process 3A"), later improved ("Process 3B"), which differed from Example 32. Processes 3A and 3B yielded ranitidine identical to that originally produced using Example 32. On April 15, 1980, Glaxo's David Crookes used Process 3B to produce a form ("Form 2") of ranitidine visibly different from all previous batches. Infrared spectroscopy and x- ray powder diffraction tests confirmed the difference.
Glaxo used Form 2 for commercial production because it had better filtration and drying properties. In 1980, it applied for a patent on Form 2, which issued in 1985.
Seeking to market a generic version of Form 2 in 1995 upon expiration of the '658 patent, Novopharm challenged the validity of the '431 patent.
ANTICIPATION. Novopharm argued that the '658 patent's Example 32 inherently and necessarily produced Form 2 and, therefore, anticipated the '431 patent's claims. Its experts "performed the process disclosed in Example 32 . . . thirteen times and each time they made Form 2 crystals . . ."
The district court found that Example 32 could yield either Form 1 or Form 2. It found that Collin and Glaxo's expert made Form 1 using Example 32. Though they "employed procedures that sometimes departed from the strict letter of Example 32", the district court found that "one skilled in the art would understand that these procedures were consistent with the teaching of Example 32."
The court of appeals affirmed because it was not persuaded that the district court's fact findings were clearly erroneous.
INEQUITABLE CONDUCT. In 1983, the patent office examiner had rejected Glaxo's Form 2 claims as anticipated by or obvious in view of the '658 patent. In response to Glaxo's argument that its claims were to a different form, the examiner asked for a showing that Form 2 was patentably distinct from the prior art '658 composition. Glaxo submitted a declaration by John Harold Hunt, head of Glaxo's spectroscopy group. (The court noted that "The case was complicated by the trial court's inability to hear testimony from Hunt due to his death in 1985.") Hunt's declaration compared infrared spectra and x-ray diffraction patterns of Forms 1 and 2 and concluded that there were significant differences between the two.
The district court found the declaration materially misleading because the Form 1 data Hunt submitted was not obtained from ranitidine prepared according to Example 32, but it refused to infer fraudulent intent, finding that Hunt "did not believe there were any differences between material produced using Example 32 and the material from which he obtained the data . . . ."
The court of appeals affirmed: the district court's finding that the defendant failed to carry its burden of proving intent to deceive by clear and convincing evidence was "debatable" but "that is not sufficient reason to reverse in the absence of firm and definite belief that the district court erred." "Intent is often inferred from surrounding circumstances when a material misrepresentation is shown. . . . But an inference is not required in every case, even when the misrepresentation is in affidavit form."
BEST MODE. Form 2 had better filtration and drying properties but was hampered by poor flow properties "which made the material difficult to measure and dispense in its pure form." To solve the flow problem, Glaxo scientists developed a novel azeotroping process. Glaxo filed a British patent application for the process but abandoned it in order to maintain trade secrecy.
Glaxo officials were aware of a potential "best mode" problem if they sought a United States patent without disclosing the azeotroping process. Brereton, the patent officer initially in charge of U.S. prosecution, recommended that Glaxo not claim pharmaceutical compositions of the Form 2 salt, apparently believing that disclosure of the azeotroping process would not be necessary if claims were only to the compound but would be necessary to support composition claims because the process was the best way to make the Form 2 salt for use in preparing pharmaceutical compositions. After Brereton moved to another position at Glaxo, the U.S. application was amended to include pharmaceutical composition claims even though there was no disclosure of the azeotroping process.
Opposing Novopharm's motion for summary judgment of invalidity for failure to set forth the best mode, Glaxo submitted inventor Crookes' affidavit stating that he personally did not know of the azeotroping process and worked in a different department from those who developed it. The district court denied the motion, noting that it could not hold as a matter of law that the inventor, Dr. Crookes, knew of the azeotroping process, and reserving for trial the issue whether other Glaxo employees' knowledge could be imputed to him. At trial, Novopharm presented no evidence as to Crookes' personal knowledge of the azeotroping process. After trial, the court indicated that it would find a best mode violation if knowledge imputation were proper but concluded that the statute did "not permit using imputed knowledge in a best mode analysis."
In a 2-1 decision, the court of appeals affirmed. Judge Rich's majority opinion relied on the literal language of the statute, which refers only to the "inventor", and on a prior decision, Texas Instruments, Inc. v. U.S. Int'l Trade Comm. , 871 F.2d 1054, 10 USPQ2d 1257 (Fed. Cir. 1989). "[W]e held in Texas Instruments[, Inc. v. U.S. Int'l Trade Comm., 871 F.2d 1054, 10 USPQ2d 1257 (Fed. Cir. 1989)] that there was no violation of the best mode requirement of Section 112 by reason of knowledge of the purported best mode on the part of T.I. employees, other than the inventor, in the manufacturing group when the inventor did not know of or conceal this best mode."
In a strongly-worded dissent, Judge Mayer complained that "the court blesses corporate shell games resulting from organizational gerrymandering and willful ignorance by which one can secure the monopoly of a patent while hiding the best mode of practicing the invention the law expects to be made public in return for its protection.": "I would hold that if there truly was . . . a pattern of deliberate concealment of information that would otherwise have been known to the inventor, the knowledge of those who sought to conceal that information and who now attempt to enforce the patent may be imputed to the inventor."
Responding to the dissent, Judge Rich noted that (1) "whether [the inventor's employer] deliberately walled off the inventor is irrelevant to the issue of failure of his application to disclose the best mode known to him" and (2) "because an agency relationship does not exist for purposes of what is disclosed in a patent application, it would be improper to impute a patent attorney's knowledge of a best mode to the inventor for purposes of finding a best mode violation."
As to Novopharm's argument that "looking solely to the inventor's knowledge makes a mockery of the best mode requirement, and fosters a 'head in the sand' mentality for corporate applicants", Judge Rich noted that "the practical reality is that inventors in most every corporate scenario cannot know all of the technology in which their employers are engaged. Therefore, whether intentionally or not, inventors will be effectively isolated from research no matter how relevant it is to the field in which they are working." "Separating scenarios in which employers unintentionally isolate inventors from relevant research from instances in which employers deliberately set out to screen inventors from research, and finding a best mode violation in the latter case, would ignore the very words of Section 112, first paragraph . . . ."
The majority and dissent both commented on a controversial item of testimony. Glaxo's in-house patent agent testified that the inventor "was not consulted 'at any time' during the preparation of the application that matured into the '431 patent." Judge Rich noted that this "completely ignores the requirement that patents are applied for 'in the name of the actual inventor or inventors' according to 37 C.F.R. Section 1.41(a) (1983)." "The inventor(s) must submit an oath or declaration attesting that they have 'reviewed and understand[ ] the contents of the specification' and believe 'the named inventor or inventors to be the original and first inventor or inventors of the subject matter which is claimed and for which a patent is sought.' 37 C.F.R. Section 1.63(b)(1), (2) (1992); see also 37 C.F.R. Section 1.51(a)(2) (1992). . . . [The accused infringer] has not alleged that these requirements were violated. It is therefore presumed that Crookes, the inventor and applicant, must have reviewed the specification and signed the required declaration before the application was filed."
Judge Mayer found it "incongruous to rely on the inventor's actual knowledge here if Glaxo indeed thought that knowledge was so insignificant that it did not even merit consultation during preparation of the application." "The court finds comfort in the regulations requiring that the inventor sign an oath attesting that he has reviewed the application, . . . reasoning that [employer-corporate assignee] must have at least let [the inventor] review the application before it was filed. But none of these regulations speaks to the best mode requirement. Nor are we told how [the inventor] could sign such an oath if he was never consulted before the application was filed. Perhaps the court has hit upon grounds for a charge of inequitable conduct against [the employer] that everyone else missed."