An Analysis of Trends in the Construction of U.S. Patent Claims

An Analysis of Trends in the Construction of U.S. Patent Claims: 1997-2002 John R. Thomas Professor of Law Georgetown University Washington, DC Prepared for The Institute of Intellectual Property Tokyo, Japan Washington, DC TABLE OF CONTENTS EXECUTIVE SUMMARY..............................................................................................i I. CLAIM CONSTRUCTION ......................................................................................…….1 A. Continuing Development of Markman.................................................3 B. Use of the Specification in Interpreting Claims ................................13 C. The Role of the Preamble....................................................................16 D. The Role of the Transition Phrase......................................................19 E. The Claim Differentiation Doctrine...................................................22 F. The Reverse Doctrine of Equivalents.................................................24 II. THE DOCTRINE OF EQUIVALENTS .................................................................….25 A. Warner-Jenkinson and the Insubstantial Differences Test..............26 B. Federal Circuit Developments Post-Warner Jenkinson...................29 C. Prosecution History Estoppel.............................................................32 D. Prior Art Restrictions ........................................................................37 E. The All-Limitations Rule....................................................................39 F. The Public Dedication Doctrine.........................................................42 III. CONSTRUCTION OF M EANS -PLUS -FUNCTION CLAIMS ...................................44 A. Triggering the § 112 ¶ 6 Interpretational Protocol...........................46 B. The Scope of Structural Equivalency.................................................47 C. Functional Claims for Software Inventions.......................................49 D. Step-Plus -Function Claims..................................................................52 IV. CONCLUDING OBSERVATIONS ........................................................................56 APPENDIX OF CITED CASES ................................................................................….58 2 EXECUTIVE SUMMARY The period between 1997 and 2002 marked a dramatic shift in the way U.S. courts interpret patent claims. During this five-year span, the U.S. Court of Appeals for the Federal Circuit (“Federal Circuit”) continued to refine the interpretational analysis pioneered in the 1995 Markman decision. The Federal Circuit also added considerable precision to the construction and range of equivalency of means-plus-function claims. The most remarkable developments during this period concerned the doctrine of equivalents, however. The U.S. Supreme Court issued two significant opinions on the subject, while the Federal Circuit unexpectedly grew far more conservative in finding infringement by equivalents. It can be confidently asserted that the scope of U.S. patent claims is at its most narrow since the creation of the Federal Circuit in 1982. In its Markman decision, the Federal Circuit held that claim interpretation was a matter of law, to be decided by the judge in the case of a jury trial. The court also established two categories of evidentiary inputs for use in claim construction. The first, so-called “intrinsic evidence,” consisted of the claims, the specification, and the prosecution history. Courts were required to consider all of the intrinsic evidence of record to determine the meaning of the claims. All other sources, ranging from dictionary definitions to expert testimony, was classified as “extrinsic evidence.” The trial court judge possessed discretion to consider extrinsic evidence, but need not do so if the intrinsic evidence established the proper interpretation of the claim. In the years following Markman, the Federal Circuit has retreated from the view that extrinsic evidence should rarely, if ever, be used in claim interpretation. In particular, the Federal Circuit has come to favor such source as dictionaries and learned treatises. Such sources, if extant at the time the patent issued, are said to demonstrate the plain meaning of words within the claims. As a result, more recent case law holds that publications may be viewed more freely than the Markman opinion suggests. The Federal Circuit also spoke to the role of the preamble and transition phrase in patent claims. The U.S. patent law has developed no clear standard for whether recitations in the preamble constitute claim limitations or are merely statements of intended use. The case law suggests that lengthier preambles, referenced by the elements within the body of the claim, amount to claim limitations. The court also interpreted the transition phrase “composed of” and maintained its adamant refusal to apply the i so-called “reverse doctrine of equivalents,” a defense to literal infringement. The Federal Circuit also continued to apply the doctrine of claim differentiation. Under this doctrine, courts should presume that each claim of a patent conveys a different meaning. The presumption may be overcome by statements from the specification or prosecution history, however, implying that differently phrased claim limitations bear the same meaning. Recent decisions demonstrate that the Markman decision did not solve all of the ambiguities concerning claim construction. In particular, Markman does not tell the patent community how the varying sources of evidence interrelate, and in particular how much weight each source should be given. In the event the prosecution history is not entirely in accord with the patent’s specification, for example, the resulting claim interpretations tend to be ad hoc. Put differently, the Markman framework of intrinsic and extrinsic evidence is not so much a recipe but a list of ingredients. In addition, a tension exists between two lines of Federal Circuit opinions interpreting patents. One line of cases states that limitations described in the preferred embodiment of the specification should not be read into the claims. The other line of cases holds that the specification can sometimes provide guidance as to the meaning of the claims, thereby dictating the manner in which the claims are to be interpreted. Because no neutral guidepost distinguishes between these two principles, the precise role of the specification in interpreting claims remains uncertain. Whether the Federal Circuit will endeavor to develop more instructive claim interpretation protocols, and whether more detailed interpretative guidelines are even possible, remains to be seen. The most significant developments in U.S. claim interpretation concerned the doctrine of equivalents. The U.S. Supreme Court, only rarely the author of patent decisions since the creation of the Federal Circuit, issued two important cases involving equivalent infringement. The 1997 Supreme Court decision in Warner-Jenkinson was the first of these. In Warner-Jenkinson, the Court resolved many issues concerning the doctrine. Among them were that equivalency should be judged at the time of infringement; that prosecution history estoppel and the all-limitations rule acted to restrict the doctrine; and that the “insubstantial differences” standard was an appropriate verbal formulation governing equivalency. In the years following Warner-Jenkinson, Federal Circuit cases took a major and unexpected turn. There can be no question that the Federal ii Circuit’s approach to equivalency is far more conservative than it was just a few years earlier. Recent Federal Circuit opinions stress that if an accused infringement was foreseeable by a claims drafter of ordinary skill, then it is the duty of that claims drafter to obtain literal protection from the U.S. Patent and Trademark Office (“U.S. PTO”). The Federal Circuit simply will not use the doctrine of equivalents to grant protection that should have been literally claimed. As a result, the Federal Circuit appears unwilling to use the doctrine of equivalents as an error-correction mechanism. Finally, the Federal Circuit has put more stress upon the notice function of claims. Claims are supposed to put others on notice of what activities they can permissibly engage in, the court has explained, and it would be unfair to change routinely the boundaries of those claims through the liberal application of the doctrine of equivalents. These developments culminated in the Federal Circuit’s decision in the Festo case, in which the en banc court held that prosecution history estoppel resulting in a “strict bar” denying the patentee any scope of equivalents. The Supreme Court later overturned the Federal Circuit on that point, but the Supreme Court’s decision largely vindicates the increasingly restrictive equivalency holdings of the Federal Circuit. The Supreme Court also identified three scenarios that would avoid prosecution history estoppel: (1) unforeseeable equivalents for which the applicant could not have fashioned literal claim coverage, (2) amendments with only tangential relations to the equivalent in question, and (3) some other reason frustrating accurate description in claims. The Federal Circuit has continued to hold that the scope of the prior art limits the scope of permissible equivalents of a claim. As a result, a patentee may not obtain, through the doctrine of equivalents, coverage that he could not lawfully have obtained from the PTO through expressly drafted claims. As confirmed by the Supreme Court in Warner-Jenkinson, the Federal Circuit also applies the all-limitations rule to equivalency determinations. Under this rule, a finding of infringement may arise only if each element of a claim is expressed in the accused infringement, either literally or equivalently. The doctrine of equivalents is thus not oriented to the patented invention as a whole, but instead is more strictly employed with respect to each element of a patent claim. Federal Circuit opinions have not definitely resolved what claim language constitutes a limitation for purpose of the all-limitations rule, however, leaving some ambiguities present in applying this principle. In its 2002 decision in Johnson & Johnston, the en banc Federal Circuit iii recognized an additional restraint upon the doctrine of equivalents. Under the “public dedication doctrine,” subject matter that is disclosed in a patent, but not claimed, may not be appropriated through the doctrine of equivalents. Such unclaimed subject matter is considered to have been deliberately disclaimed and therefore dedicated to the public. The Federal Circuit has done much to clarify the interpretation of means-plus-function claims. Significant case law has developed principles for determining whether a claim qualifies as a means-plus-function claim, and therefore is subject to the special interpretational rules found in § 112 ¶ 6 of the U.S. Patent Act. The Federal Circuit has also distinguished between the common law doctrine of equivalents and the statutory equivalents directed by § 112 ¶ 6. In addition, the Federal Circuit has spoken about step-plus-functional claims and the use of means-plus-function claims in the crucial area of software technologies. Interestingly, in a time of sweeping changes, the U.S. patent law is in many ways becoming more certain. With a more constrained approach to equivalent infringement, and more rules built into the basic claim construction determination, U.S. patents today are as narrowly interpreted as they have been since the early 1980's. Absent a prospective defendant’s literal infringement of the claims, patentees are wise to either refrain from bringing suit or to pursue reissue proceedings. As the U.S. patent system continues to evolve, only the future will reveal the wisdom of these choices upon U.S. innovation policy. iv AN ANALYSIS OF TRENDS IN THE CONSTRUCTION OF U.S. PATENT CLAIMS 1997-2002 -- John R. Thomas* The five-year period between 1997 and 2002 marked a dramatic shift in the thinking about the interpretation of patent claims in the United States. With the Markman revolution already achieved by the mid-1990's, the U.S. Court of Appeals for the Federal Circuit continued to refine its literal infringement jurisprudence. Guided by two significant U.S. Supreme Court decisions, the Federal Circuit also developed its case law in the areas of the doctrine of equivalents and the interpretation of means-plus-function claims. The resulting judicial opinions have marked a major and unexpected turn in the Federal Circuit’s patent jurisprudence. There can be no question that the court interprets claims far more narrowly than it did just a few years earlier. With a more constrained approach to equivalent infringement, and more rules built into the basic claim construction determination, U.S. patents enjoy a more circumscribed scope than they have since the creation of the Federal Circuit. This report tracks considers trends in U.S. patent claim construction from 1997 through 2002. Part I first considers the refinement of the Markman line of cases at the Federal Circuit. The report then addresses several recurring situations in claim construction, including the relationship between the specification and the claims, the role of the preamble and transition phrase, the claim differentiation doctrine and the reverse doctrine of equivalents. Part II of the report turns to the doctrine of equivalents. It reviews two significant U.S. Supreme Court decisions, Warner-Jenkinson and Festo, that explain the scope of equivalency. The report then considers four principal limitations upon the doctrine: prosecution history estoppel, the all-limitations rule, the prior art and the public dedication doctrine. Finally, in Part III, this report reviews the Federal Circuit’s evolving jurisprudence in the area of means-plus-function claims. The report considers case law that identifies means-plus-function claims, refines their scope of coverage, and applies these rules to process and software claims. Part IV offers concluding observations. I. Claim Construction The Patent Act consistently bases the patentee’s exclusive rights upon the “patented invention.”1 U.S. courts have consistently judged this phrase to refer to the invention as recited in the claims. If the accused infringement 1 embodies every limitation of the claim, the claim is said to “read on” that product or process. In such cases the defendant has literally infringed the patent. In determining the meaning of patent claims, the U.S. courts traditionally faced two basic questions. The first is the determination of the materials that may be appropriately considered during claim interpretation. Beyond the asserted patent claim itself, litigants have offered such evidence as the prosecution history, prior art documents, learned treatises, dictionary definitions and the testimony of inventors, technical experts and legal experts. As these sources often provide conflicting evidence of meaning, the courts have required an essentia l interpretational protocol for construing patent claims. Second, whether the issue of claim interpretation is one of fact or law determined whether juries or judges determined the meaning of the claims, as well as the standard of review on appeal. The 1995 en banc opinion of the Federal Circuit in Markman v. Westview Instruments, Inc. provided answers to these fundamental questions.1 As such, it has proven an influential source of both the substance and procedure of contemporary claim interpretation. The Markman court first established two categories of evidentiary inputs for use in claim interpretation. The first, so-called “intrinsic evidence,” consisted of the claims, the specification, and the prosecution history. Courts were required to consider all of the intrinsic evidence of record to determine the meaning of the claims. All other sources, ranging from dictionary definitions to expert testimony, was classified as “extrinsic evidence.” The trial court judge possessed discretion to consider extrinsic evidence, but need not do so if the intrinsic evidence established the proper interpretation of the claim. The Federal Circuit also held that claim interpretation was a matter of law solely for the judge to decide. The majority reasoned that patents, like other written instruments, had traditionally been subject to interpretation by the courts. Further, participants in the patent system were best served by an articulated, reasoned analysis of the claim scope by a trained jurist. As a result, in a jury trial, it is the responsibility of the judge to resolve the meaning of the claims and to so instruct the jury. On appeal, the Federal Circuit would review claim interpretation holdings de novo, without discretion to the trial court. In the years following the issuance of the Markman decision, the Federal Circuit has had numerous occasion to apply these principles to 2 U.S. CLAIM CONSTRUCTION 1997-2002 different factual settings. This report next considers the refinement of the Markman line of cases at the Federal Circuit. The report then addresses several recurring situations in claim construction, including the relationship between the specification and the claims, the role of the preamble and transition phrase, the claim differentiation and the reverse doctrine of equivalents. A. Continuing Development of Markman The Federal Circuit’s decision in Pitney Bowes Inc. v. Hewlett-Packard Co. demonstrates the Federal Circuit’s changing position towards the use of extrinsic evidence – that is, expert testimony, dictionaries, technical writings and other evidence external to the patent and prosecution history – in patent claim interpretation. More particularly, the Federal Circuit appears to have continued to retreat from its earlier position that extrinsic evidence should rarely, if ever, be used in claim interpretation. To understand Pitney Bowes, one must be familiar with the earlier Federal Circuit opinion in Vitronics Corp. v. Conceptronic, Inc.3 Issued shortly after the U.S. Supreme Court published its well-known decision in Markman v. Westview Instruments, Inc.,4 Vitronics provided a detailed review of the Federal Circuit’s position towards the use of extrinsic evidence in claim interpretation. Many subsequent Federal Circuit opinions and legal commentary cited Vitronics as the court’s leading restatement of claim interpretation principles. 2 The most noteworthy aspect of the Vitronics opinion was its strong language discouraging trial courts from using extrinsic evidence to construe patent claims. According to the Federal Circuit, when the patent instrument and prosecution history were unambiguous, expert testimony regarding the meaning of a claim would be entitled to no weight. Otherwise, members of the public would be disadvantaged, because they would not be able to consult expert testimony that does not then exist to interpret the claims. The Vitronics court also discounted the use of inventor testimony to construe patent claims, reasoning that such testimony was most often self-serving. This aspect of the Vitronics decision was not well-received by the district courts. Traditionally, U.S. trial court judges have enjoyed great discretion in managing evidentiary issues during trial. The detailed, micro-management of the Federal Circuit in Vitronics appeared to supplant this usual authority and create special rules for patent trials. The trial courts also doubted whether the advice of Vitronics comported with the usual rule that claims should be read from the perspective of a person of skill in the art. Vitronics seemed to discourage courts from obtaining information about how a 3 U.S. CLAIM CONSTRUCTION 1997-2002 skilled artisan would interpret the claims, so long as the court believed the patent documents to be clear. A subsequent Supreme Court decision, Kumho Tire Co. v. Carmichael,5 seemed to support the views of the trial court judges. Kumho Tire was a tort case and did not involve patents in any way. But the Supreme Court did discuss at length the use of scientific evidence in civil litigation. The Court’s general conclusion was that district courts were trained gatekeepers for expert testimony and enjoyed broad discretion to hear evidence that is proponent demonstrated to be of reasonable reliability. The Kumho Tire decision seemed to repudiate the hard-nosed attitude Vitronics had taken towards extrinsic evidence. The Pitney Bowes litigation provided the Federal Circuit to step back from its seemingly hostile position towards the use of expert testimony in patent claim interpretation. Here, Pitney Bowes owned U.S. Patent No. 4,386,272, entitled “Apparatus and Method for Generating Images by Producing Light Spots of Different Sizes.” The ‘272 patent recognized that sometimes the corners and edges of certain characters generated by laser printers can have an uneven appearance. The laser printer industry referred to this problem as the “jaggies.” The ‘272 patent disclosed two methods for overcoming jaggies by using toner dots of different sizes. In one method, an intensity modulator attached to a light beam source varies the degree of exposure on a photoreceptor by changing the intensity of the beam of light and the length of time that the beam of light remains in contact with the surface of the photoreceptor. As the modulator changes the intensity of the beam of light, the number of electrons displaced on the photoreceptor also change. A second method taught by the ‘272 patent employs two power sources using parallel laser beams, each having a different diameter and corresponding spot size. Because different-sized exposed areas are generated on the photoreceptor, different-sized dots of toner are created on the paper, thereby reducing the problem of jaggies. Pitney Bowes sued HP for patent infringement in the District Court for the Distrct of Connecticut. According to Pitney Bowes, HP’s “Resolution Enhancement technology” class of laser printers infringed the asserted claims of the ‘272 patent. Representative claim 1 of the ‘272 patent provided: 1. A method of producing on a photoreceptor an image of generated shapes made up of spots, comprising: 4 U.S. CLAIM CONSTRUCTION 1997-2002 directing a plurality of beams of lights towards a photoreceptor, each beam of light generating a spot on the photoreceptor and controlling a parameter of the light beams to produce spots of different sizes whereby the appearance of smoothed edges are given to the generated shapes. At trial, the parties differed on their interpretation of the term “spots of different sizes.” HP contended that this term referred to the spots of light generated by the light beam on the photoreceptor. Because HP’s laser printers did not produce light spots of different sizes, this construction would mandate a finding of noninfringement. However, Pitney Bowes argued that the phrase “spots of different sizes” actually described the spots of discharged area on the photoreceptor that result from contact with the light beam. On summary judgment, the district court concluded that the plain meaning of the claims did not fully support either construction. However, the court reasoned that HP’s claim construction was more accurate in view of the specification and prosecution history. The district court in part observed that Pitney Bowes had amended the title of the ‘272 patent to modify the term “spots” with the word “light.” Significantly, the court also relied upon the affidavit of an HP expert, while apparently discounting an affidavit of a Pitney Bowes technical expert, to determine the terminology employed in the digital printing field. As a result, the district court concluded that in the digital printing field, it was common to define the size of a light spot based upon the intensity of the light beam that produced it. This usage supported HP in that it suggested that “spots” are the spots of discharged areas on the photoreceptor. As a result, the district court issued a summary judgment of noninfringement. Pitney Bowes then appealed to the Federal Circuit. Citing Vitronics, Pitney Bowes argued that the district court erred in relying on expert testimony in construing the claims. A three-judge panel consisting of Judges Michel, Plager and Rader heard the appeal. The resulting opinion from the Federal Circuit was an unusual one. The court issued a lead opinion written by Judge Michel, the author of the Vitronics opinion. The lead opinion was followed by “additional views” authored by Judge Rader and joined by Judge Plager. The opinion of Judge Michel began by interpreting the claims. He noted that the preamble of claim 1 spoke towards generating images on a photoreceptor, not the light spots generated by the beam of light, as the district court had determined. As well, the body of claim 1 referred to beams of light 5 U.S. CLAIM CONSTRUCTION 1997-2002 “generating a spot on the photoreceptor.” Judge Michel reasoned that this language suggested that the term “spots of different sizes” referred to spots of discharged area on the photoreceptor, as suggested by Pitney Bowes. Judge Michel also believed that the trial court had overemphasized the amendment of the ‘272 patent’s title. According to the Federal Circuit, its precedent has never suggested that an amendment to a patent title during prosecution carries an estoppel effect in the manner of a claim amendment. Judge Michel concluded by directing that the noninfringement ruling be vacated, and he remanded the case for further proceedings. The most significant portion of the Pitney Bowes opinion concerns the district court’s use of expert testimony. Judge Michel explained that a proper reading of Vitronics revealed that it did not prohibit the use of expert testimony to interpret claims. Instead, Vitronics simply held that extrinsic evidence should not contradict the meaning of the claims discernable through a careful examination of the claims, written description and prosecution history. Judge Michel went on to provide that consultation of extrinsic evidence is especially appropriate to ensure that a judge’s understanding of the technical aspects of the case comported with those of skilled artisans. Applying this standard to the present case, Judge Michel explained that the district court did not err by considering expert testimony. According to the Federal Circuit, the district court had properly relied upon the written description and amendment of the patent’s title to reach its claim construction. Although the Federal Circuit ultimately concluded that the district court’s construction was erroneous, the source of this error was not its reference to the extrinsic evidence. Judge Michel opinion attempted to downplay some of the more robust language of the Vitronics opinion. A reading of the additional views of Judge Rader, which were joined by Judge Plager, suggests that these two jurists were more quick to point out that Vitronics had gone too far. According to Judge Rader, Vitronics should not have dictated a claim interpretation process that excluded reliable expert testimony. Judge Rader recognized that expert testimony could provide information of considerable value to the trial court, including a proper technological context to understand the claims and an understanding of prosecution procedures. Suggesting that Kuhmo Tire was inconsistent with Vitronics, the additional views applauded the Federal Circuit’s efforts to express more trust in the trial courts to give expert testimony its proper weight in the claim interpretation process. The Pitney Bowes opinion recognizes that the trial courts continued to 6 U.S. CLAIM CONSTRUCTION 1997-2002 accept expert testimony in order to assist in the interpretation of claims, even after Markman. The Federal Circuit has now legitimized this effort and allowed the trial courts to be more forthright as to which sources they relied upon in reaching patent claim interpretations. Pitney Bowes is an opinion that is very much a credit to the Federal Circuit, which remains quite receptive to the views of the trial courts and the patent community on the improved resolution of patent disputes. Although the Federal Circuit appears to be viewing the use of extrinsic evidence with somewhat more favor, the court has maintained the basic Markman formulation. Subsequent opinions demonstrate that the Markman evidentiary structure does not solve all claim interpretation problems, however. Markman does not tell the patent community how the varying sources of evidence interrelate, and in particular how much weight each source should be given. It may be that no more guidance of this sort is possible. The Federal Circuit’s decision in K-2 Corp. v. Salomon S.A. suggests these difficulties.6 K-2 Corp. owns U.S. Patent No. 5,437,466, concerning an in-line roller skate. The patent describes a skate with a soft, pliable inner “bootie” or “shoe” surrounded by molded plastic or straps attached to the base of the skate. This arrangement allows the wearer’s foot to breathe and offers a substantially lighter skate while retaining the structural stiffness required for use of the skate. Representative claim 1 of the ‘466 patent provided in part: 1. In an in-line roller skate having an upper shoe portion and a lower frame portion ... a non-rigid shoe portion adapted to receive and substantially enclose the entire foot of the skater ... support means positioned adjacent selected areas of said non-rigid shoe portion for providing support to aid the skater in maintaining said in-line roller skate in a substantially vertical position ... and a base portion, ... said non-rigid shoe portion being permanently affixed to said base portion at least at said toe area and said heel area for substantially preventing movement therebetween at least in a horizontal plane, wherein at least a portion of said non-rigid shoe portion extends continuously from said base portion to at least the top of said ankle support cuff. 7 U.S. CLAIM CONSTRUCTION 1997-2002 K-2 brought suit against Salomon S.A. and Salomon North America in the District Court for the Western District of Washington. K-2 contended that Salomon’s model TR skate infringed the ‘466 patent. The TR skate included a soft inner bootie surrounded by a rigid plastic structure. The TR skate’s inner bootie is fastened to the lower, rigid portion of the skate by rivets and a screw in the toe area, as well as by a removable hex-head screw in the heel area. The crucial issue in the case was whether the use of a removable screw in the heel area of the TR skate was “permanently affixed’ within the meaning of the ‘466 patent claims. At trial, the district court sided with Salomon. The district court concluded that the claim language “permanently affixed” did not read upon a removable screw. As a result, there was no literal infringement. The district court also observed that K-2 had amended its claims to avoid the prior art Johnson paent, U.S. Patent No. 5,331,752. The Johnson patent disclosed an in-line skate in which the bootie is removable and affixed at the toe and mid-foot. The district court therefore concluded that prosecution history estoppel barred K-2 from asserting equivalent infringement. K-2 appealed to the Federal Circuit. A three-judge panel consisting of Judges Clevenger, Rader and Gajarsa heard the appeal. Over a dissent by Judge Rader, Judge Clevenger wrote the majority opinion affirming summary judgment of noninfringement. Judge Clevenger began by observing that the ordinary and accustomed meaning of “permanently affixed” was that the fastening be unremovable. K-2 argued, however, that the phrase “permanently affixed” should be understood in the context of the claim language immediately following it – “for substantially preventing movement therebetween at least in a horizontal plane.” K-2 suggested that this language converted the meaning of “permanently fixed” to mean “affixed to prevent movement at least in a horizontal plane.” The majority declined to follow K-2's reading of the claim. Judge Clevenger ruled that a more plausible reading of these two limitations was that the fastening be both permanent and sufficiently rigid to prevent horizontal movement of the bootie. Judge Clevenger noted that it was easy to conceive of a fastening that was permanent, but not strong enough to prevent sliding. In the view of the majority, K-2's interpretation amounted to little more than an effort to eliminate the term “permanently” from the claim. As a result, the Federal Circuit affirmed the district court’s claim construction and its finding 8 U.S. CLAIM CONSTRUCTION 1997-2002 of no literal infringement. Judge Clevenger then turned to the doctrine of equivalents issues. K-2 asserted that the removable screw attaching the heel of the bootie in Salomon’s skate was equivalent to the “permanently affixed” fastening required by the ‘466 patent claims. The Federal Circuit agreed with Salomon, however, that prosecution history estoppel resulted in the surrender of any subject matter encompassing a removable connection between the bootie and the base of the skate. The majority also cited French Patent No. 2,688,072, which taught the use of a soft upper bootie attached to the skate base with screws at the toe and heel. Judge Clevenger explained that the Salomon skates were, at most, an obvious variation of the skate taught by the French patent. Because K-2 could not have obtained claims that would have literally covered the Salomon skates at the PTO, it would be inappropriate for the courts to grant the identical claim scope through the doctrine of equivalents. The Federal Circuit therefore affirmed the district court’s holding that the Salomon skate was not an equivalent of the invention claimed in the ‘466 patent. Judge Rader wrote a dissenting opinion, in which he explained that the majority had been too quick to reach its judgment. Claim terms should be interpreted from the perspective of an ordinary person of skill in the art, Judge Rader explained. To illustrate his concerns, Judge Rader noted: “‘Permanent’ in the context of mountain ranges can mean millennia; ‘permanent’ in the context of hair treatments can mean until the next rain storm; ‘permanent’ in the context of creased pants can mean until the next long airline trip in coach class; ‘permanent’ in the context of genetic combinations can mean until the next mutation.” Judge Rader accused the majority of not even considering what meaning an ordinary in-line skate artisan would attach to “permanently.” Instead, the judges simply applied their own linguistic values on a public decision. K-2 demonstrates that the Markman framework of intrinsic and extrinsic evidence is not so much a recipe but a list of ingredients. How much weight a particular evidentiary input should be given in the interpretative process will vary from case to case. Whether the Federal Circuit will endeavor to develop more instructive claim interpretation protocols, and whether more detailed interpretative guidelines are even possible, remains to be seen. The subsequent decision in Wang Laboratories Inc. v. America Online Inc.,1 applied these principles to the drafting and construction of patents 9 U.S. CLAIM CONSTRUCTION 1997-2002 relating to Internet technologies. Wang Laboratories owned the patent- in- suit, U.S. patent No. 4,751,669. The ‘669 patent describes an apparatus for providing users with textual and graphical information from computer-controlled databases through interactive, bilateral communication over a telephone network. The Wang system allows users to review, retrieve and store pages or “frames” of data from different information suppliers. The preferred embodiment provided in the written description of the ‘669 patent employed a so-called “character-based protocol.” Such a protocol divides a video display into a grid wherein a character, such as a letter or number, is placed in each cell of the grid. Such systems sends frames of digital information to a subscriber as a stream of encoded bytes identifying the characters to be displayed and containing instructions about the attributes of the displayed frame such as the size of the characters or their background color. Character-based protocols have only a limited graphical capacity, which is obtained through the use of mosaic characters. Wang sued American Online Inc. (“AOL”) and Netscape Communications Corp. for infringement of the ‘669 patent in the United State District Court for the Eastern District of Virginia. Among Wang’s contentions was that claim 20 of the ‘669 patent read upon the well-known “favorite places” service of AOL and the “bookmark” feature provided by Netscape. Claim 20 of the ‘669 patent provided: 20. Apparatus for retrieving selected frames of information from a central videotex supplier of the information frames, each information frame having an associated unique identifier assigned by the supplier for retrieving the frame, the apparatus comprising a display device connected to display the information frames, storage means connected to store the identifier and a unique keyword selected by an operator and associated with the identifier, menu means connected to display on the display device a menu frame containing the keyword, data entry means connected to enter into the apparatus a request for retrieval of a selected information frame by moving a cursor to the keyword associated with the selected information frame, and a processor connected to be responsive to the data entry means for retrieving the information frame in response to the entry of the request by transmitting the associated identifier to the 10 U.S. CLAIM CONSTRUCTION 1997-2002 supplier. During trial, AOL and Netscape pointed out that their systems did not use a character-based system, but rather a bit-mapped protocol. Bit-mapped protocols encode an image with reference to individual pixels of the display monitor. According to the defendants, this different technical characteristic mandated a conclusion of non-infringement. During trial, the district court sided with AOL and Netscape. The district court construed the claims of the ‘669 patent to be limited to character-based systems, as opposed to bit-mapped protocols. This interpretation resulted from the district court’s definition of the claim term “frame” as a “page of information assembled prior to display which is encoded in a character-based protocol . . . to then be displayed on the screen representing a fixed full screen arrangement, such as rows and columns of alphanumeric and graphic characters.” Based upon this claim interpretation, the district court granted summary judgment of noninfringement, both literally and under the doctrine of equivalents. Wang then appealed to the Federal Circuit. A three-judge panel consisting of Judges Mayer, Newman and Gajarsa heard the appeal, with Judge Newman writing the unanimous opinion. Wang’s principal argument on appeal was that the district court’s definition of “frame” was too narrow in view of the ‘669 patent’s specification and prosecution history. The Federal Circuit disagreed, resulting in an affirmance of the district court’s summary judgment of noninfringement. Although the term “frame” can in general usage be applied to bit-mapped displays and character-based systems alike, Judge Newman explained, the ‘669 patent limited the term “frame” to character-based systems. The court observed that the only system supported by an enabling description in the ‘669 patent specification used a character-based protocol. The Federal Circuit acknowledged that the ‘669 patent did mention other protocols, including a bit-mapped protocol. However, many of these citations occurred only in the “Background of the Invention” portion of the patent, and each of them simply acknowledged the state of the art. These fleeting references did not allow the conclusion that bit-mapped protocols fell within the scope of the claims, according to the court. To further support its interpretation, the Federal Circuit noted that Wang itself had not implemented a bit-mapped protocol of the system described in the ‘669 patent. The court also pointed to testimony of the 11 U.S. CLAIM CONSTRUCTION 1997-2002 inventors named in the ‘669 patent that they had been unable to develop a decoder for a bit-mapped protocol. Although Wang had correctly argued that a claim is not invalid because it reads upon subject matter not specifically described in the specification, Judge Newman explained, the subject matter must at least be described as the applicant’s own invention to meet the requirements of section 112. Wang also pointed to language in the ‘669 patent that stated that the character-based protocol was simply a “preferred embodiment.” Wang then cited case law that the embodiment described in the specification does not necessarily set the boundaries of the claims. The Federal Circuit responded by noting that while the “precedent offers assorted quotations in support of differing conclusions concerning the scope of the specification, these cases must be viewed within the factual context in which they arose.” According to the Federal Circuit, whether an invention is fairly claimed more broadly than the preferred embodiment is question specific to the content of the specification, the context in which the embodiment is described, the prosecution history, and if appropriate, the prior art. In this case, Judge Newman concluded, the adjective “preferred” as to the disclosure did not by itself broaden the claims beyond their support in the specification. The Federal Circuit found that the ‘669 patent’s prosecution history was in accord. Judge Newman noted that in an Information Disclosure Statement filed by Wang during the prosecution of the ‘669 patent, Wang distinguished a reference by stating that it “encodes pictorial information . . . on the [picture element] level, rather than on the character level.” This history reinforced the conclusion that Wang focused its invention upon a character-based system. The Federal Circuit therefore agreed with the district court’s conclusion that no literal infringement was possible following this claim interpretation. The court also upheld the district court’s summary judgment of non-infringement under the doctrine of equivalents. Judge Newman agreed that the differences between character-based and bit-mapped protocols were substantial. The record established that the operation, structure and capabilities of the systems were extensively different, according to the Federal Circuit. The court in particular noted Wang’s own technical expert explanation that “huge, huge differences” separated the two systems, and that bit-mapped protocols allowed much more detail and complexity in graphic displays. As a result, no reasonable fact finder could conclude only insubstantial differences between the two systems. 12 U.S. CLAIM CONSTRUCTION 1997-2002 The Wang Laboratories opinion suggests that the Federal Circuit seems to be interpreting patent claims more narrowly than it did in the past. Arguably the Federal Circuit might have decided the doctrine of equivalents issue in this case differently a decade ago. Many electronics engineers recognize that the distinction between character-based and bit-mapped protocols are not large. Further, the trend towards bit-mapped protocols was motivated by subsequent technological advance: namely, the more robust communications media allowing increased rates of data transmission. This sort of situation was traditionally considered an apt case favoring application of the doctrine of equivalents. Wang Laboratories may suggest that the court will be reluctant to apply the doctrine of equivalents in areas of rapid technological advance such as the Internet. B. Use of the Specification in Interpreting Claims Under the U.S. Patent Act, the claims define the scope of patent protection. However, the role of the rest of the patent -- the so-called specification -- in interpreting claims has proven a controversial topic. Courts often declare that the terms of the claims are to be read in light of the specification. Yet they also state that it is inappropriate to import limitations from the specification into the claims. Because no neutral principles govern this distinction, whether a court is interpreting or importing in a particular case is largely a subject exercise. The decision in SciMed Life Systems, Inc. v. Advanced Cardiovascular Systems, Inc.,1 illustrates this difficulty. There the Federal Circuit affirmed a finding of noninfringement. The plainitff, SciMed Life Systems, Inc. (“SciMed”), owns three U.S. patents that concern balloon dilution catheters: U.S. Patents No. 5,156,594; 5,217,482; and 5,395,334. These three patents are related and have nearly the same specifications. However, the claims of these three patents vary, being directed towards different aspects of the catheter. SciMed filed suit against Advanced Cardiovascular Systems, Inc. (“ACS”) in the United States District Court for the Norther District of California, charging ACS with infringement of each of the three patents. The SciMed patents describe catheters having three sections: a first shaft section, a second shaft section, and a transition section between the two. The catheters claimed in the SciMed patents contain two passageways, or lumens. According to the SciMed patents, one of these passageways is positioned inside the other. This arrangement is referred to as the “coaxial lumen” configuration. The accused devices marketed by ACS were constructed in a somewhat different fashion. They used the “dual lumen” 13 U.S. CLAIM CONSTRUCTION 1997-2002 configuration. In this arrangement, the two lumens are positioned side-by-side within the catheter. As the litigation proceeded, ACS filed a motion for summary judgment. The position of ACS was that the trial judge should determine that ACS did not infringe any of the patents, without needing to take the case to a full trial. The trial judge agreed and granted the motion of ACS. The trial judge reasoned that the asserted patent claims were limited to coaxial lumen catheters. The trial judge also held that ACS did not infringe under the Doctrine of Equivalents. SciMed appealed to the Federal Circuit. A three-judge panel consisting of Judge Bryson, Senior Judge Plager and Judge Dyk heard the appeal. Judge Dyk wrote the opinion, which was joined by the other two judges. Judge Dyk also wrote a separate, short concurring opinion. The Federal Circuit began by identifying the sole issue in the case: whether the shared specification of the three patents limits the scope of the asserted claims to catheters with coaxial lumens. SciMed argued that none of the asserted claims actually included the term “coaxial lumen.” According to SciMed, the trial judge had impermissibly imported a limitation from the patent specification’s preferred embodiment into the claims. Citing many cases that such a claim interpretation is inappropriate, SciMed urged the Federal Circuit to reverse the lower court’s judgment. The Federal Circuit disagreed with SciMed that the trial judge had impermissibly read a limitation from the written description into the claims. Judge Dyk instead explained that claims must be read in view of the specification. When the specification makes clear that the invention does not include a particular feature, that feature is judged to be outside the scope of the claims. This principle holds even where the claims, read without reference to the specification, might be considered broad enough to include that particular feature. Judge Dyk then engaged in a detailed reading of the three SciMed patents. He first noted that the abstracts of the patents described the lumens to be in a coaxial relationship. Second, the patents distinguished the prior art based upon the use of dual lumens, explaining that coaxial lumens were preferable. Third, in the “Summary of the Invention” portion of the patents, several textual passages expressly characterized SciMed’s invention as including a coaxial configuration. 14 U.S. CLAIM CONSTRUCTION 1997-2002 Judge Dyk then turned to a portion of the specification titled “Catheter Intermediate Sleeve Section.” That portion of the specification expressly described the catheter’s two lumens as having a coaxial positioning. The specification then stated that the “structure defined above is the basic sleeve structure for all embodiments of the present invention contemplated and disclosed herein.” Such broad and unequivocal language made it difficult to imagine how the patents could have more clearly stated that a coaxial lumen configuration was a necessary element of the invention. The Federal Circuit therefore concluded that SciMed’s invention was defined in such a way that the dual lumen arrangement could not be covered. Judge Dyk reasoned that the use of coaxial lumens was not merely the preferred embodiment of the invention. Instead, the coaxial positioning of the lumens applied to all embodiments of the invention. The Federal Circuit therefore affirmed the trial judge’s summary judgment that the ACS catheters did not literally infringe the SciMed patents. The Federal Circuit then turned to the Doctrine of Equivalents. Judge Dyk agreed with the trial judge that ACS did not infringe the SciMed patents by the Doctrine Equivalents. Referring to his earlier reading of the SciMed patents, Judge Dyk observed that these patents had specifically identified and criticized the dual lumen configuration. Indeed, the SciMed patents had regarded the dual lumen configuration as significantly inferior to the coaxial lumen configuration used in the invention. With such an express disclaimer present, Judge Dyk reasoned, SciMed should not be allowed to recapture the excluded subject matter under the Doctrine of Equivalents. To do so would allow SciMed to destroy the ability of its patents to notify others what the scope of the claims was. Such a step would be unfair to SciMed’s competitors, including ACS. As a result, the Federal Circuit concluded that SciMed had totally disclaimed the dual lumen configuration, whether under the principle of literal infringement or under the Doctrine of Equivalents. The judgment of the trial court was affirmed. Although he had authored the majority opinion, Judge Dyk also wrote a short concurring opinion. Judge Dyk there observed that tension existed between two lines of judicial opinions interpreting patents. One line of cases stated that limitations described in the preferred embodiment of the specification should not be read into the claims. The other lines of cases stated that the specification can sometimes provide guidance as to the meaning of the claims, thereby dictating the manner in which the claims are to be interpreted. Judge Dyk stated that the problem was that Federal Circuit 15 U.S. CLAIM CONSTRUCTION 1997-2002 decisions provided little guidance as when it is appropriate to look to the specification to narrow the claim by interpretation and when it is inappropriate to do so. Judge Dyk predicted that litigants and trial courts will continue to be confused until the Federal Circuit provides better guidance as to the role of the specification in interpreting the claims. C. The Role of the Preamble In General Electric Co. v. Nintendo Co., 2 the Federal Circuit explained how the preamble relates to the ultimate interpretation of a claim. General Electric (“GE”) owns three patents, Nos. 4,097,899; 4,169,659; and 4,270,125, that relate to electronic control circuitry used in connection with television systems. GE filed suit claiming that Nintendo’s video game systems, including the Nintendo Entertainment System, Super Nintendo Entertainment System, and Gameboy, infringed the three patents. Nintendo responded by filing summary judgment motions asserting that its systems were noninfringing, and also that certain claims of the ‘899 patent were invalid. The District Court of the District of New Jersey granted Nintendo each of the summary judgments it sought, resulting in an appeal to the Federal Circuit. Judge Michel authored an unanimous opinion for a panel that included Judges Newman and Gajarsa. The Federal Circuit first turned to the ‘899 patent. The ‘899 patent described a switch that allows television users to switch between video information received from an antenna and other video information, such as the output of a video record player, VCR or similar device. Claim 12 of the ‘899 patent provided in part: Video record player apparatus comprising: a player RF signal input terminal; a player RF signal output terminal; a player power supply developing supply potentials when selectively enabled; [and] means, responsive to the absence of supply potential development by said player power supply, for establishing a second signal path between said player RF signal input terminal and said player RF signal output terminal; said second signal path being disrupted in the presence of supply potential development by said player power supply. Nintendo argued both that its systems were noninfringing and that the asserted claims were invalid. Regarding noninfringement, Nintendo argued 16 U.S. CLAIM CONSTRUCTION 1997-2002 that its systems do not “disrupt” the signal path as required by the final element of claim 12 and related claims. Nintendo pointed to language in the ‘899 patent specification that specifically distinguishes between disrupting (establishing a high series impedance) and bypassing (creating a path of low resistance). The Federal Circuit agreed that the Nintendo systems bypass rather than disrupt the signal path. Not only was this finding preclude a possibility of literal infringement, Judge Michel also reasoned that there was no possibility of liability under the doctrine of equivalents either. The court concluded that no reasonable jury could find that “disrupting” was an equivalent function to “bypassing” because disrupting signal path results in an alternation of the signal path. The court therefore affirmed the judgment that Nintendo systems do not literally infringe the ‘899 patent because they did not perform the claimed function. The Federal Circuit then turned to Nintendo’s contention that the asserted claims of the ‘899 patent were invalid. The district court had found that the claims were anticipated by a Japanese patent publication of September 3, 1975, SHO 49-7211. GE argued that this reference was not anticipatory because it did not disclose automatic signal switching once a secondary devices such as VCRs began to output electronic signals. Nintendo did not dispute that this limitation was not expressly disclosed by the reference, but instead argued that a skilled artisan would understand this feature was inherent to the reference. Judge Michel disagreed, stating simply that the reference was not disclosed or inherent to the reference. The Federal Circuit also noted that the district court had not made an obviousness determination and that this issue was not susceptible to review. The district court’s summary judgment of invalidity was therefore reversed. The court next turned to the ‘659 patent. The ‘659 patent is directed towards a signal generator that provided timing information for use by video source equipment such as television cameras or video tape recorders. Claim 1 of the ‘659 patent provided in part: A television synchronizing signal generator comprising: a source of clock pulses; [and] a vertical counter coupled to an output of said synchronous horizontal counter for producing signals . . ., said vertical counter being clocked by a signal which is advanced in phase . . . . 17 U.S. CLAIM CONSTRUCTION 1997-2002 The crucial issue was whether the Nintendo systems included vertical counters “clocked by a signal which is advanced in phase.” For a vertical counter to be clocked, the signal must go through the clock line. Nintendo’s devices send the signal through the input line, not the clock line. Therefore no literal infringement was possible. Judge Michel also upheld the district court’s judgment that there was no equivalent infringement. The Federal Circuit observed tha t GE did little more than present conclusory statements that Nintendo’s input line usage allowed precise timing control, just as if the signal had been clocked. According to the court, because GE pointed to no evidence of equivalence, GE had failed to demonstrate a genuine issue of material fact sufficient to overcome summary judgment. Last, the court considered the ‘125 patent. The ‘125 patent concerned a method for displaying computer generated information on a display screen. It employed direct memory access (“DMA”) to allow the computer to transfer data without diverting resources from the CPU. The crucial issue concerning the ‘125 patent was whether its single asserted claim, claim 1, required a bit map display device. Bit- mapped systems encode an image with reference to individual pixels of the display monitor. The Nintendo systems did not generate individual bits, but instead transferred whole characters made up of fixed compilations of bits. Claim 1 of the ‘125 patent is quite lengthy. However, it referred to bit mapping at only point – in its preamble. The preamble of claim 1 recited in part: “A system for displaying a pattern on a raster scanned display device by mapping bits from a display location in a memory associated with a computer onto the raster as contrasting spots depending on the value of each bit . . . .” The issue before the Federal Circuit was whether this language required that the Nintendo systems perform bit mapping in order to literally infringe. The U.S. patent law has developed no clear standard for whether recitations in the preamble constitute claim limitations or are merely statements of intended use. If the preamble terms “give meaning to the claim and properly define the invention,” then they are judged as necessary claim limitations. In re Paulsen.3 Otherwise the preamble terms are considered merely setting forth the intended technical setting in which the invention will be used, and therefore not elements that restrict the scope of the claim. In this case, Judge Michel concluded that the preamble of claim 1 amounted to a claim limitation. The court reasoned that the ‘125 patent specification made clear that the inventors were working on the particular problem of displaying binary data on a raster scan display device and not general improvements to all display systems. The court also noted that 18 U.S. CLAIM CONSTRUCTION 1997-2002 during the prosecution of the ‘125 patent, the applicant had distinguished a reference by noting that it mapped characters, not bits. The Federal Circuit concluded that claim 1 required bit mapping, and therefore the Nintendo systems infringed the ‘125 patent neither literally or equivalently. The General Electric decision is notable for its approval of multiple summary judgments in a complex patent infringement litigation. Such a resolution was rare prior to the Supreme Court’s decision in Markman v. Westview Instruments, Inc. 4 Following the Markman Court’s conclusion that claim interpretation is a matter of law to be resolved by the court, however, patent infringement cases will increasingly conclude through the grant of summary judgment. General Electric also counsels claims drafters not to use lengthy language in the preamble. Whether such language rises to the level of a claim limitation is often ambiguous, and it may take a judicial opinion to resolve the issue conclusively. As claims drafters in Japan typically draft more prolix preambles than are typical in the United States, consideration should be given to adapting claims to U.S. style when filing applications here. D. The Role of the Transition Phrase Ordinarily, U.S. patent professionals use one of three transition phrases between the preamble of the patent claim, which identifies the invention and gives its technical context, and the body of the pat ent claim, which sets forth the specific boundaries of the patentee’s proprietary interest. These three transition phrases are: • “Comprising” – If this term is used, the claim may cover devices that employ additional, unrecited elements. • “Consisting of” – If this phrase is used, the claim is understood to exclude any elements, steps or ingredients not specified in the claim. • “Consisting essentially of” – Unlike the open term “comprising,” or the closed phrase “consisting of,” the phrase “consisting essentially of” is a hybrid. If this phrase is used, the claim is understood to cover additional elements that do not materially affect the basic and novel characteristics of the claimed invention. In AFG Industries, Inc. v. Cardinal IG Co., 5 the Federal Circuit addressed the meaning of a transition phrase that is not commonly used: 19 U.S. CLAIM CONSTRUCTION 1997-2002 “composed of.” The plaintiff in that case, AFG Industries, Inc. is associated with Asahi Glass Co. As a result, AFG is a joint owner of U.S. Patent No. 4,859,532. The ‘532 patent relates to windows with “low-emissivity” coatings. Low-emissivity coatings generally consist of thin, alternating layers of metals coated onto a glass pane. When properly engineered, the metallic coatings are clear, or may have a desired color. These coatings, while permitting visible light to pass, reflect radiant heat. Prior to 1987, the state of the art disclosed the use of thin layers of silver for coating glass. However, the use of a single layer of silver led to undesirable qualities in the coated glass. If the silver layer was too thin, the coating did not sufficiently reflect radiant heat. If the silver layer was too thick, the window became reflective of visible light, resembling a mirror. In the late 1980's Asahi scientists Mamo ru Mimhashi, Takuji Oyama and Koichi Suzuki discovered that layers of silver could be alternated with layers of metal oxides. By alternating these layers, the silver would become increasingly reflective of radiant heat without sacrificing its transparency to visible light. Asahi filed a patent application claiming these invention, which resulted in the issuance of the ‘532 patent. With emphasis added, claim 1 of the ‘532 patent reads as follows: A transparent laminated product comprising a transparent substrate and a 5-layered transparent coating composed of: a first ZnO layer formed on the substrate, a second Ag layer formed on the first layer, a third ZnO layer formed on the second layer, a fourth Ag layer formed on the third layer and a fifth ZnO layer formed on the fourth layer, and having a visible ray transmission of at least 60%, wherein the thickness of each Ag layer is from 60 to 250Å. The specification of the ‘532 patent further explains that "interlayers" could be laid between the silver and metal oxide layers. The patent explains that these interlayers, which are sufficiently thin to avoid substantially changing the optical properties of the silver and metal oxide layers, make the layers more durable and increase their adhesivity. The relevant part of the ‘532 patent states: For the purpose of improving the adhesion or durability of the coating layers, an interlayer having a thickness not to substantially affect the optical properties may be inserted at the 20 U.S. CLAIM CONSTRUCTION 1997-2002 interface with the substrate or at the interface between adjacent layers or at the interface with air. AFG later sued Cardinal IG Co. for infringement of the ‘532 patent. Cardinal’s low-emissivity coatings contain layers of zinc oxide and silver. However, in some of Cardinal’s products, multiple deposits of zinc oxide are applied sequentially on top of each other, with no intervening layers of silver. In addition, some of the layers in Cardinal’s products are separated by thin deposits of titanium dioxide, referred to by Cardinal as "barrier" layers. Also, some of Cardinal’s accused products have an "overcoat" or "top coat" of a thin deposit of silicon nitride or zinc oxide on top of the other layers of the coating. At the district court, the trial judge concluded that the accused Cardinal products could not infringe the ‘532 patent. The trial judge took the position that the ‘532 patent only covered coatings containing five metallic layers, and does not cover coatings that contain other layers or interlayers. AFG then appealed to the Federal Circuit. Judges Michel, Lourie and Bryson heard the appeal. All three judges joined the majority opinion written by Judge Michel. The Federal Circuit began by considering the phrase “composed of” in the ‘532 patent. Because the drafter of the ‘532 patent did not use one of the standard transition phrases, it was uncertain whether claim 1 of the ‘532 was open to additional layers or not. Judge Michel decided the best way to proceed was to interpret the phrase “composed of” in light of the ‘532 patent as whole. Observing that the specification of the ‘532 patent recognized the existence of interlayers, Judge Michel concluded that in this case, the term “composed of” should be considered to be in hybrid format. In other words, claim 1 was open to additional layers that did not materially affect the basic and novel properties of the invention. The Federal Circuit next considered the correct interpretation of the term “layer” in claim 1 of the ‘532 patent. To interpret this term, the trial judge had consulted a standard Webster’s dictionary. Judge Michel quickly indicated that this approach was suspect. Earlier Federal Circuit cases had often cautioned against the use of such dictionaries, Judge Michel noted. In this case, the Fede ral Circuit concluded that ordinary dictionary definitions did not provide for the specific technological context considered by the ‘532 patent. Judge Michel also reasoned that the dictionary definitions provided by 21 U.S. CLAIM CONSTRUCTION 1997-2002 the trial court was inconsistent with expert testimony offered at trial. This testimony indicated that as a form of shorthand, the glass industry did not refer to interlayers when considering the final glass product. Such interlayers were not employed for their optical properties, but only t enable the coating of o glass with metallic layers that did have optical properties. As a result, the industry would generally only refer to interlayers during the manufacturing process. The Federal Circuit concluded that this testimony established that claim 1 of the ‘532 patent could cover glass coatings with interlayers between the expressly recited metallic layers. The trial judge’s opinion was vacated and the matter sent back to the district court for further proceedings. AFG Industries, Inc. v. Cardinal IG Co. suggests that the Federal Circuit has somewhat relaxed from its earlier views regarding the role of expert testimony in interpreting patent claims. Although the Federal Circuit was originally quite hostile to the use of expert testimony to interpret patent claims, this case suggests that sometimes experts have a role to play. This shift in attitude has likely resulted from the court’s experiences over the past several years. It has proven difficult for judges, who most often lack technical training, to interpret patent claims to complex technologies without appropriate assistance. AFG Industries, Inc. v. Cardinal IG Co. indicates that expert testimony may be appropriately considered when judges construe patent claims. This decision also decides how the Federal Circuit will interpret the phrase “composed of” when used in patent claims. Unfortunately, the court did not decide this matter with great clarity. What the phrase “composed of” means will depend upon the entirety of the patent instrument, and may vary from case to case. It would be wise not to use the phrase “composed of” when drafting patent claims. Instead, claims should employ one of three standard transition phrases – either “comprising,” “consisting of,” or “consisting essentially of” – to ensure a more predictable scope of patent protection. E. The Claim Differentiation Doctrine The doctrine of claim differentiation is a staple of U.S. claim construction. Under this doctrine, courts should presume that each claim of a patent conveys a different meaning. In an exemplary decision applying the claim differentiation doctrine, Ecolab Inc. v. Paraclipse Inc., 6 the Federal Circuit considered a patented lighted insect trap. Claim 16 of the asserted patent called for “a source of insect attractant light.” Claim 17 then specified the “trap of claim 16 wherein the insect attractant light comprises a source of ultraviolet light.” The Federal Circuit applied the doctrine of claim 22 U.S. CLAIM CONSTRUCTION 1997-2002 differentiation to hold that claim 16 did not require an ultraviolet light. The claim differentiation doctrine is not inviolate, however, as demonstrated by Fantasy Sports Properties, Inc. v. Sportsline.com.7 The plaintiff in that case owned U.S. Patent No. 4,918,603, relating to a method for playing a fantasy football game on a computer. The game is played by individuals who act as owners of imaginary teams composed of actual football players. Points are awarded to each team based upon the performance of its players in actual professional football games. The total points for each team are tabulated utilizing a database containing the relevant statistics for each player. The ‘603 patent contained 3 claims, which read: 1. A computer for playing football based upon actual football games, comprising: means for setting up individual football franchises; means for drafting actual football players into said franchises; means for selecting starting player rosters from said actual football players; means for trading said actual football players; means for scoring performances of said actual football players based upon actual game scores such that franchises automatically calculate a composite win or loss score from a total of said individual actual football players’ scores; said players’ scores are for quarterbacks, running backs and pass receivers in a first group and kickers in a second group; and wherein said players in said first and second groups receive bonus points. 2. The apparatus according to claim 1, wherein said bonus points for said first and seconds groups are based upon complex or difficult plays. 3. The apparatus according to claim 2, wherein said complex and difficult plays include extra points for a quarterback who receives or runs for said touchdown, extra points for said running back for throwing or receiving a touchdown pass, and extra points for said pass receive for passing the ball or running for a touchdown. The claim interpretation dispute in that appeal centered upon claim 1’s 23 U.S. CLAIM CONSTRUCTION 1997-2002 recitation of the term “bonus points.” The specification of the ‘603 patent explained that “bonus points” may be awarded based upon the difficulty of the play. The accused infringers took the position that they did not infringe the ‘603 patent because they did not award “bonus points” in this manner. For example, one of the accused products did not award additional points for scoring plays not typically associated with the position of a scoring player – such as a quarterback scoring a running touchdown. In contrast, the patent owner took the position that the “bonus points” limitation of claim 1 should be broadly construed to cover any points awarded in addition to those given in an actual football game. The patent owner in part turned to the doctrine of claim differentiation. The patent owner noted that claim 2 limited the definition of “bonus points” to “complex or difficult plays,” while claim 3 further limited that definition in terms of specific examples, such as “extra points for a quarterback who receives or runs for [a] touchdown.” As a result, the patentee argued that the term “bonus points” in claim 1 must be given a broader reading. Judge Lourie of the Federal Circuit denied the patent owner’s argument. He observed that during prosecution, the patentee had acquiesced to the USPTO examiner’s rejection based upon a 1987 publication. That publication had explained that fantasy football leagues could award additional points for player performance, such as distance scoring and total yardage. By cancelling originally filed claims with broader wording, Judge Lourie reasoned, the patentee had disclaimed any interpretation of the term “bonus points” that was described in the 1987 publication. Judge Lourie next reasoned that the claim differentiation doctrine creates only a presumption that my be overcome by the patent owner’s disclaimer of subject matter in the prosecution history. Consequently, the “bonus points” language was given the same meaning in all of the claims of the ‘603 patent. F. The Reverse Doctrine of Equivalents In its 1950 decision of Graver Tank v. Linde, the U.S. Supreme Court explained that the doctrine of equivalents may also act against the interest of the patentee.8 When an accused product or proce ss is literally covered by the words of a patent claim, but is “so far changed in principle” that it performs in a “substantially different way,” the court may reach a finding of noninfringement. The so-called “reverse doctrine of equivalents” amounts to a defense to literal infringement. Although the reverse doctrine is equitable in nature, whether an accused technology goes sufficiently outside the principles of the patented invention is a question of fact.9 24 U.S. CLAIM CONSTRUCTION 1997-2002 The reverse doctrine of equivalents has bee n of greater interest to scholars than to the courts. 10 Defendants have only rarely attempted to invoke this defense, and not a single Federal Circuit opinion squarely applies it. Given this paucity of precedent, it is more important to recognize situations where the reverse doctrine does not apply. One such case is where a defendant merely puts a claimed invention towards a new use. For example, suppose that the Major Drug Company holds a patent claiming a chemical composition and a method of using the composition to cure hypertension. An independent researcher named Harry Hirsute discovers that the compound also prevents male pattern baldness, a previously unknown use. If Hirsute manufactures the compound, then he has undoubtedly infringed the Major Drug Company patent. The reverse doctrine of equivalents would not apply even if Hirsute intends to sell the compound only to balding men with normal blood pressure, and even if his invention is separately patentable as a process. The reverse doctrine of equivalents was instead intended to apply to extraordinary cases. It provides courts with something of an escape hatch, useful when a finding of literal infringement would work an unwarranted extension of the claims. The reverse doctrine might pertain to rapidly progressing fields of high technology, where radical subsequent advances allow predecessor patents to appropriate subject matter entirely beyond the scope of their technical contribution. Such circumstances appear so uncommon that, in its two decades of existence, the Federal Circuit has yet to encounter them. In Tate Access Floors, Inc. v. Interface Architectural Resources, Inc.,11 the Federal Circuit made abundantly clear that the reverse doctrine of equivalents was of extremely limited applicability. Judge Gajarsa observed that: “Not once has this court affirmed a decision finding noninfringement based on the reverse doctrine of equivalents.” Describing the doctrine an “an anachronistic exception,” the court refused to apply or extend the reverse doctrine of equivalents. II. The Doctrine of Equivalents The exclusive rights provided by a patent are founded upon, but not exclusively limited to, the text of its claims. Although the courts have long recognized the value of clear and certain claims, they have refused to confine the infringement inquiry to their precise choice of words. Instead, the scope 25 U.S. CLAIM CONSTRUCTION 1997-2002 of protection associated with a patent may be expanded beyond the literal wording of the claims under the doctrine of equivalents. Under the current formulation of the doctrine, an accused product or process that presents insubstantial differences from the claimed invention will be judged an equivalent and therefore an infringement.12 The doctrine of equivalents arose from judicial efforts to stop competitors who would introduce insignificant modifications into the claimed invention in order to avoid literal infringement.13 However, courts have not limited use of the doctrine to cases of bad faith, copying or piracy. Every patent infringement case potentially involves the doctrine of equivalents. When courts apply the doctrine of equivalents, they attempt to balance fair protection for the patentee with appropriate notice to competitors of the scope of the patentee’s exclusive rights. This part considers the current contours of the doctrine of equivalents in light of the1997 Warner-Jenkinson decision of the U.S. Supreme Court. It then considers four limitations that restrict the scope of the doctrine of equivalents. The first, prosecution history estoppel, prevents patentees from receiving a scope of protection that had been surrendered at the PTO in order to persuade the examiner to allow the claim. Patentees are also restricted by the prior art, and may not obtain a claim construction that would embrace technologies known to the art or their obvious variants. Finally, under the all- limitations rule, each element recited in the claim is considered material to defining the invention. As such, the doctrine of equivalents is applied not to the invention as a whole, but to the individual elements recited within the claim. A. Warner-Jenkinson and the Insubstantial Differences Test The Supreme Court’s 1997 opinion in Warner-Jenkinson Co. v. Hilton Davis Chemical Co. provides a definitive analysis of the modern doctrine of equivalents. 14 Both the plaintiff-patentee, Hilton Davis, and the accused infringer, Warner-Jenkinson, manufactured dyes. The patent- in-suit claimed a method of removing impurities by filtering the dye through a porous membrane. As originally presented to the PTO, the claims did not speak towards the pH level at which this ultrafiltration process should be performed. Later during prosecution, in order to distinguish a prior art patent that operated at a pH above 9.0, Hilton Davis added the limitation “at a pH from approximately 6.0 to 9.0.” The accused process operated at a pH of 5.0. A jury trial resulted in a finding of infringement under the doctrine of equivalents. The jurists of the Federal Circuit agreed to hear the case in banc and affirmed 26 U.S. CLAIM CONSTRUCTION 1997-2002 in a set of lengthy, deeply divided opinions. The Supreme Court then granted certiorari and offered a sweeping review of the doctrine of equivalents. Warner-Jenkinson’s first series of arguments contended that the doctrine of equivalents did not survive the enactment of the 1952 Patent Act. Its position was that the doctrine of equivalents was inconsistent with the statutory requirements for definite claims and broadening reissue proceedings, as well as the role of the PTO in determining claim scope. The Court tersely responded that each of these points had been raised and rejected in its earlier decisions, and there was no basis for overturning that decision today.15 A fourth, more interesting argument was that Congress had wholly codified the doctrine of equivalents when it authorized functional claiming in the sixth paragraph of § 112. This statute allows applicants to draft so-called “means plus function” claims that are“construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.” According to Warner-Jenkinson, congressional authorization of an equivalency provision for functional claims amounted to the rejection of a general doctrine of equivalents for other sorts of claims. Justice Thomas rejected this position as well, reasoning that Congress specifically intended § 112 ¶ 6 to overturn the Supreme Court’s 1946 opinion in Halliburton Oil Well Cementing Co. v. Walker, which had cast grave doubts on the validity of functional claiming. Such limited congressional action could not be seen to have such sweeping consequences for the venerable doctrine of equivalents.16 Warner-Jenkinson also urged that equivalents should be limited to variants described in the patent’s specification or, alternatively, to those variants known at the time the patent issued. Justice Thomas quickly dismissed both of these positions, reasoning that the perspective of the person of ordinary skill in the art placed sufficient limits on the doctrine. The Court also upheld the ample precedent proclaiming equivalency to be a factual determination, to be decided by the jury in a jury trial.17 The Court also addressed a position that had been gaining increasing currency in the patent community: that the doctrine of equivalents should apply only to cases where the accused infringer engaged in bad faith, copying or intent. This proposal too was readily rejected. Although the Graver Tank opinion had reasoned that the doctrine of equivale nts discourages piracy and the “unscrupulous copyist,” the Court did not read Graver Tank as limiting the doctrine to cases where those benefits were obtained. Ample precedent had described the doctrine in more neutral terms and had not limited equivalency to actors in bad faith. Following Warner-Jenkinson, then, intent 27 U.S. CLAIM CONSTRUCTION 1997-2002 is not an element of a doctrine of equivalents analysis.18 The final portion of the Supreme Court’s Warner Jenkinson opinion addressed the appropriate standard for the doctrine of equivalents. Concerned that the function- way- result test usually associated with earlier doctrine of equivalents cases not easily apply to sophisticated, nonmechanical technologies, the in banc Federal Circuit had engaged in a thorough rereading of these opinions. Observing that the most recent Supreme Court decision on the doctrine of equivalents, Graver Tank & Mfg. Co. v. Linde Air Products Co., 19 had several times described cases of “insubstantial differences” as appropriate for a finding of equivalency, the Federal Circuit explicitly held that “the application of the doctrine of equivalents rests on the substantiality of the differences between the claimed and accused product or processes, assessed according to an objective standard.”20 The Supreme Court was surprisingly nonchalant towards the Federal Circuit’s reinterpretation of Graver Tank. Declining to select the appropriate standard of equivalency, the Court observed that “[d]ifferent linguistic frameworks may be more suitable to different cases, depending on their particular facts.” According to Justice Thomas, the essential inquiry was whether the accused product or process contained elements identical or equivalent to each element recited in the patent claim. The Court left for the Federal Circuit the further refinement of appropriate equivalency standards on a case-by-case basis.21 Given the Court’s detached view of the appropriate standard of equivalency, Justice Thomas did not elaborate upon the newly approved insubstantial differences standard. However, the Warner-Jenkinson opinion and subsequent Federal Circuit opinions have suggested several mechanisms for determining whether insubstantial differences exist between the claimed invention and accused technology. As in Graver Tank, known interchangeability of substitutes for an element of the patented invention and independent experimentation by the accused infringer might be probative.22 The familiar nonobviousness analysis may also be pertinent to the equivalency inquiry. 23 That a skilled artisan would have found the substitution of one ingredient of the claimed invention for the ingredient appearing in the accused product is probative of insubstantial differences. Finally, the function-way-result test is not dead after Warner-Jenkinson, but has evolved into one way of articulating whether the differences between the claimed and accused technologies are insubstantial. 28 U.S. CLAIM CONSTRUCTION 1997-2002 B. Federal Circuit Developments Post-Warner Jenkinson Perhaps the most notable development in U.S. claim construction occurred at the Federal Circuit following the Supreme Court’s Warner-Jenkinson opinion. There can be little question that the Federal Circuit became much more conservative in its approach to the doctrine of equivalents. Once known as a court that gave a quite generous scope to patent claims, suddenly the court became far less willing to allow patentees to obtain a scope of protection outside the literal scope of the claims. The 2000 decision in Moore U.S.A., Inc. v. Standard Register Co. is exemplary of this change in thinking. 24 Moore owned U.S. Patent No. 5,201,464, relating to a mailer-type business form. The ‘464 patent claims called for “longitudinal strips of adhesive . . . extending the majority of the lengths” of two side of the form. The accused form featured adhesive strips covering 47.8% of its lengths. Using strong language, the Federal Circuit upheld the lower court’s summary judgment of noninfringement. Judge Michel offered two reasons for the affirmance: First, to allow what is undisputedly a minority (i.e., 47.8%) to be equivalent to a majority would vitiate [the claim language.] If a minority could be equivalent to a majority, this claim language would hardly be necessary . . . . Second, it would defy logic to conclude that a minority – the very antithesis of a majority – could be insubstantially different from a claim limitation requiring a majority, and no reasonable juror could find otherwise.25 With respect to the Federal Circuit, although 47.8% may not be a majority, it is a very close to one. Presidents of the United States have been elected with voter support of this level or less. To say that no reasonable decision maker could find equivalency is a strong statement indeed. At all events, this sweeping language suggests that the Federal Circuit’s view on equivalency have changed. The reasons for this sea change in reasoning are difficult to trace. When the Federal Circuit was created, the court perceived its mission as strengthening the patent system. With the doctrine of equivalents effectively serving as the poster child of a robust patent regime, the court was quick to find that patents had been infringed, often by seemingly disregarding 29 U.S. CLAIM CONSTRUCTION 1997-2002 the precise wording of the claims. Twenty years have since passed. The jurists on the Federal Circuit have changed and, perhaps, the sense of the court’s mission has shifted. The justice of a broad doctrine of equivalents seems less apparent today. Repetition may be one reason for the shift – after a judge has heard patent appeals for many years, he may tire of equivalency arguments that amount to the position that “black means white” or “up means down.” In the words of the traditional saying, familiarity breeds contempt. The Federal Circuit’s 1997 decision in Sage Products, Inc. v. Devon Industries Inc. presented an important development. 26 It influenced not only many of the Federal Circuit doctrine of equivalents opinions to follow, but also the Supreme Court in its Festo decision. In Sage Products, the claimed invention concerned a disposal container useful for hazardous medical waste, including hypodermic needles. The claim called for a “disposal container” with “an elongated slot at the top of the container body” and “a first constriction extending over said slot.” This latter constrictive barrier restricted access to the interior of the container so that a person disposing waster products would not touch materials already within the container. The accused product had an elongated slot, but it was not on the top of the container body. Rather, the accused product’s elongated slot was within the container body. In affirming the district court’s judgment that there was no infringement either literally or equivalently, Judge Rader explained: The claim at issue defines a relatively simple structural device. A skilled patent drafter would foresee the limiting potential of the “over said slot” limitation. No subtlety of language or complexity of the technology, obfuscated the significance of this limitation at the time of its incorporation into the claim. . . . If Sage desired broad patent protection for any container that performed a function similar to its claimed container, it could have sought claims with fewer structural encumbrances. . . . [A]s between the patentee who had a clear opportunity to negotiate broader claims but did not do so, and the public at large, it is the patentee who must bear the cost of its failure to seek protection for this foreseeable alteration of its claimed structure.27 A number of important points flow from this passage. First, the court stresses that if an accused infringement was foreseeable by a claims drafter of ordinary skill, then it is the duty of that claims drafter to obtain literal protection from the U.S. PTO. The Federal Circuit simply will not use the doctrine of equivalents to grant protection that should have been literally claimed. Of course, what is foreseeable, like what would have been obvious, 30 U.S. CLAIM CONSTRUCTION 1997-2002 is not always an easy question. One wonders if future patent cases will increasingly turn on “forseeability hearings” that employ the testimony of patent and technical experts in order to determine what an person of ordinary skill should have drafted literally in her claims. Second, the Federal Circuit appears unwilling to use the doctrine of equivalents as an error-correction mechanism. If the patentee unwittingly drafted its claims so that foreseeable infringements are not within their literal language, then the patentee must return to the U.S. PTO to seek a reissue proceeding. Not only does reissue lack the flexibility of the doctrine of equivalents, the Patent Act imposes a number of significant limitations on reissue. The patentee must profess an error to be allowed to pursue a reissue application, 28 all broadening reissues must be filed within two years of the issuance of the patent, 29 and the principle of intervening rights may benefit the patentee’s competitors. 30 All said, the Federal Circuit has made the life of a patentee more difficult than in its salad days, where the doctrine of equivalents could overcome claim drafting errors. Finally, the Federal Circuit stresses the notice function of claims. Claims are supposed to put others on notice of what activities they can permissibly engage in, the court has explained, and it would be unfair to change routinely the boundaries of those claims through the liberal application of the doctrine of equivalents. One difficulty with this approach is that a stinting approach to the doctrine of equivalents may encourage competitors to operate just outside of the boundary of the claims – subjecting the patentee to the formalities of verbalism and significantly decreasing the value of the patent right. At all events, although the insubstantial differences test is the starting point for the doctrine of equivalents, but the fulfillment of this standard does not necessarily compel a conclusion of infringement. Four significant constraints place limits upon the application of the doctrine of equivalents. These are prosecution history estoppel, the prior art, the all- limitations rule and the public dedication doctrine. These limiting principles are frequently tested in patent infringement litigation and are worthy of further discussion here. 31 U.S. CLAIM CONSTRUCTION 1997-2002 C. Prosecution History Estoppel The principle of prosecution history estoppel precludes a patentee from obtaining a claim construction before a court that would include subject matter surrendered at the PTO during prosecution. It is named for the "prosecution history" or "file wrapper," the publicly available papers that document the dialogue between the inventor and examiner during the patent acquisition. If the court concludes that an applicant relinquished certain subject matter in order to secure the allowance of her claims, then as a patentee she may not employ the doctrine of equivalents to recapture the renounced subject matter.31 For example, suppose that Professor Gadget filed a patent application claiming a vaccine. 32 As filed, Gadget’s claims contain no limitations concerning the number of times the vaccine should be administered in order to provide an effective immunization. During prosecution, the examiner cites a prior art reference that discloses a vaccine similar to Gadget’s invention. In order to distinguish his vaccine from the prior art, Gadget inserts a limitation into his claims. As amended, Gadget’s claims recite a “single administration vaccine” such that “said patient may be effectively immunized with one dose.” Satisfied that Gadget’s vaccine presents a patentable advance over the prior art, the examiner allows the claims to issue. Suppose further that Gadget later sues Dr. Nefarious for patent infringement. Nefarious manufactures a vaccine very similar to that of Gadget, but the accused vaccine must be administered in two separate doses at least one month apart. Nefarious may order a copy of the prosecution history of the Gadget patent from the PTO and place it into evidence before the trial court. The court would then realize that Gadget expressly limited his claims to a single administration vaccine in order to avoid the prior art. Nefarious would further assert that as a result, the “one dose” limitation should not be expanded to “two doses” via the doctrine of equivalents. Because there is no literal infringement and prosecution history estoppel bars the application of the doctrine of equivalents, the court should dismiss Gadget’s infringement suit. In addition to estoppel through claim amendment, courts have also recognized estoppel by argument. 33 If an applicant makes an argument to the examiner characterizing the claimed invention or distinguishing it from the prior art, then prosecution history estoppel may apply as well. No claim need be amended in these circumstances. Courts consider whether a competitor reading the administrative record would reasonably believe the applicant surrendered subject matter. Absent a strong indication from the prosecution history, PTO examiners are assumed to rely upon applicants arguments when 32 U.S. CLAIM CONSTRUCTION 1997-2002 deciding whether to allow patents to issue or not. The doctrine of prosecution history estoppel formally differs from the use of prosecution histories as intrinsic evidence of the meaning of a claim. As described in the Markman opinion, prosecution histories complement the patent instrument as an essential input towards resolving the meaning of claim terms. Prosecution history estoppel is a more rigid doctrine that comes into play only when the patentee resorts to the doctrine of equivalents. In practice, however, these two uses of the prosecution history are complementary. Whether employed in the context of the initial task of interpretation, or in a subsequent infringement inquiry under the doctrine of equivalents, resort to the prosecution history ensures that claims may not be construed one way in order to obtain their allowance and in another way against accused infringers. The Supreme Court’s opinion in Warner-Jenkinson explained the procedural aspects of prosecution history estoppel. According to Justice Thomas, when the prosecution history shows that the patentee amended the claims during prosecution, the burden fell to the patentee to explain the reason for the amendment. The court must then “decide whether that reason is sufficient to overcome prosecution history estoppel as a bar to application of the doctrine of equivalents to the element added by that amendment.” 34 If the patentee failed to offer a suitable explanation, “the court should presume that the PTO had a substantial reason related to patentability for including the limited element added by amendment” and that prosecution history estoppel should apply.35 Courts have struggled over what the extent to which prosecution history estoppel impacts the doctrine of equivalents. The following example, using dated technology, illustrates these difficulties. Suppose that, prior to the invention of the transistor, an inventor presents a claim reciting a computer that in part uses an “electric switch.” The PTO examiner rejects the claim based upon prior art. The inventor then narrows the claim by deleting the term “electric switch” and replacing it with the term “vacuum tube.” The PTO examiner then approves the claim. Subsequently, near the end of the patent’s term, the inventor brings suit against a competitor that manufactures computers using a new, state-of-the-art device – the transistor. The case law provides two alternative approaches, the strict bar and the flexible bar, to address the scope of equivalents left to an amended claim limitation. Under the strict bar approach, if a claim limitation has been amended during prosecution, then no range of equivalents exists for that amended limitation. In the above example, because transistors can act as electric switches, the patentee is deemed to have confined his invention to 33 U.S. CLAIM CONSTRUCTION 1997-2002 vacuum tubes and purposefully disclaimed transistors. Prosecution history estoppel would therefore completely defeat the patentee’s charge of infringement. The strict bar approach has the advantage of providing a “bright line” rule, but may hold harsh consequences for patentees. In contrast, applying the flexible bar, a court would assess the reason for the claim amendment to determine the remaining scope of the doctrine of equivalents. Prosecution history estoppel would apply only where the court concluded that a person skilled in the art would reasonably believe that the patentee had surrendered subject matter during prosecution. To continue the previous example, no reasonable competitor would believe that the patentee had surrendered subject ma tter by amending the claims. At the time the patentee made the amendment, the transistor had yet to be invented! As a result, the court would likely hold that prosecution history estoppel did not apply, and proceed to the doctrine of equivalents analysis . The Federal Circuit has traditionally employed a flexible bar approach. In its Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki Co. decision, 36 however, the Federal Circuit abruptly announced its shift to a strict bar approach, only to be reversed by the Supreme Court. Here the plaintiff, Festo, owned the Stoll and Carroll patents. Each patent concerned magnetic rodless cylinders. During prosecution, the claims of both patents were amended to require a pair of sealing rings. The Carroll patent was additionally amended to require a sleeve made of magnetizable material. The accused infringer, SMC, produced a device employing a single, two-way sealing ring and a sleeve made of nonmagnetizable material. Although the Stoll and Carroll patents were not literally infringed, Festo argued that infringement existed under the doctrine of equivalents. SMC in turn contended that prosecution history estoppel barred Festo from resorting to the doctrine. Festo’s position was that, following Warner-Jenkinson, an explanation existed for the two claim amendments. Festo characterized the sealing ring amendment as relating to the examiner’s rejection based on § 112, pertaining to claim definiteness. Festo further described the sleeve material amendment as wholly voluntary. SMC appealed an unfavorable judgment from the trial court, and on appeal the Federal Circuit reversed. The Federal Circuit applied the strict bar rule and held that prosecution history estoppel creates a complete bar to the doctrine of equivalents. “When a claim amendment creates prosecution history estoppel with regard to a claim element, there is no range of equivalents available for the amended claim element,” the court explained.37 34 U.S. CLAIM CONSTRUCTION 1997-2002 The Federal Circuit reasoned that the need for certainty as to the scope of patent protection was paramount. According to the Court of Appeals, amendments should be treated as disclaimers and construed against the inventor, and in favor of the public. A unanimous Supreme Court vacated and remanded the case, effectively restoring the flexible bar approach to the law of prosecution history estoppel. 38 Justice Kennedy explained that prosecution history does not bar the inventor from asserting infringement against every equivalent to the narrowed element. The Court instead mandated a specific determination of the range of subject matter surrendered by a narrowing amendment. The Court followed the approach established in Warner-Jenkinson by confirming the presumption that any territory surrendered through claim amendment is an equivalent of the territory claimed. Patentees can rebut this presumption, however, by showing that at the time of the amendment one skilled in the art could not reasonably be expected to have drafted a claim that would literally encompass the alleged equivalent. In Festo, the Supreme Court did agree with the Federal Circuit that amendments to satisfy with § 112 also could provoke prosecution history estoppel. 39 Most prosecution history estoppel cases concern amendments made to distinguish prior art, and therefore involve §§ 102 or 103 of the Patent Act. But sometimes applicants change claims in order to make them more definite, a requirement under § 112. An examiner might ask the patentee to clarify an ambiguous word, for example, or provide more detail on how the different parts of the invention interact with each other. The Court reasoned that inventors made such amendments in order to procure a patent, and that as a result prosecution history could apply in the context of § 112 as well. Although the Federal Circuit’ss opinion was “vacated and remanded” as a technical matter, the decision largely vindicates the increasingly restrictive practices regarding the doctrine of equivalents of the Court of Appeals. It is also difficult to see how Festo can ultimately prevail based upon the principles set forth in the Supreme Court opinion. A two-way seal was probably foreseeable, and Festo’s amendments were likely more than tangential to a two - way seal. Further, to the extent that the Supreme Court has made foreseeability the central component of the equivalency inquiry, one wonders the extent to which patent trials will hinge upon the testimony of legal and technical experts considering whether the accused infringement could have been predicted or not. It should also be questioned why, to the extent the doctrine of equivalents depends upon foreseeability, why the prosecution history is relevant at all. If an accused infringement was not 35 U.S. CLAIM CONSTRUCTION 1997-2002 foreseeable, then there is little in the prosecution history that will inform the equivalents inquiry. The use of prosecution histories in the task of claim interpretation has suffered criticism. 40 Like legislative histories or parol evidence, prosecution histories consist of preparatory documents of uncertain relation to the text ultimately approved. Unlike succinct, integrated patent instruments, these episodic and often lengthy compilations of correspondence can be time- consuming and costly to understand. Commentary has called for the abandonment of the principle of prosecution history estoppel in favor of an objective analysis of prior art and patentability constraints upon the scope of equivalents. As Federal Circuit cases have increasingly relied upon the prosecution history despite such critiques, the patent bar would do well to heed the consequences of the prosecution history both during enforcement proceedings and while corresponding with examiners during prosecution. 36 U.S. CLAIM CONSTRUCTION 1997-2002 D. Prior Art Restrictions The Federal Circuit’s opinion in Streamfeeder LLC v. Sure-Feed Inc.41 provides valuable guidance on the role of the prior art in determining the appropriate scope of the doctrine of equivalents. Long before the Streamfeeder decision, the U.S. courts recognized that a patentee may not obtain, through the doctrine of equivalents, claim coverage that he could not lawfully have obtained from the PTO by literal claims. The prior art therefore limits the scope of permissible equivalents of a claim. Decisions from other patent- issuing states, such as the Formstein opinion from the German Federal Supreme Court,42 recognize parallel doctrines. One technique for determining whether the prior art restricts the range of equivalents was developed in Wilson Sporting Goods Co. v. David Geoffrey & Associates. 43 In that case, Judge Rich proposed that patentees draft a hypothetical claim that is broad enough in scope to literally read on the accused product or process. The court must then consider whether the hypothetical claim could have been allowed by the PTO in view of the prior art. If so, then the prior art does not restrict the application of the doctrine of equivalents. Infringement may be found if the other requirements for equivalent infringement are met. However, if the prior art would anticipate the hypothe tical claim or render it obvious, then a finding of infringement would be improper. The hypothetical claim technique was not immediately popular. Many practitioners found this technique burdensome and confusing, particularly in jury trials. Some trial courts have preferred to consider the effect of the prior art directly, without the sometimes confusing construction of hypothetical claims. In recent years, however, this trend appears to be reversing. As the courts seek to refine its claim interpretation methods, they seem to be turning once again to the hypothetical claim methodology. The Streamfeeder litigation exemplifies this trend, and the guidance the Federal Circuit provided on appeal should prove timely. The plaintiff, Streamfeeder LLC, commenced this litigation in the District Court of the District of Minnesota. Streamfeeder owns U.S. Patent No. 4,991,831, which concerns a "bottom sheet feeder" used to collate paper and other thin materials for mass mailing, printing, binding and other applications. Bottom sheet feeder technology is a crowded art and is the subject of hundreds of other patents. One of these, U.S. Patent No. 5,601,282, is owned by a competitor of Streamfeeder named Sure-Feed Inc. Like Streamfeeder's device, the Sure-Feed system employed a "gate" to allow the 37 U.S. CLAIM CONSTRUCTION 1997-2002 processing of individual sheets. The configuration of the two gates differed somewhat. The gate disclosed in the '831 patent consists of a circular stationary gate with deep grooves at the opening and shallow grooves at the end. However, the gate of the '282 patent is elliptical and contains grooves of uniform depth. In its complaint, Streamfeeder alleged that Sure-Feed’s device, which embodied the invention described in the '282 patent, infringed the '831 patent under the doctrine of equivalents. The district court denied Sure-Feed's summary judgment motion that the prior art prevented Streamfeeder from asserting a scope of equivalents sufficient to ensnare the Sure-Feed device. The case was then tried before a jury, which returned an infringement verdict. The jury’s verdict was based upon a Wilson Sporting Goods-style hypothetical claim submitted by Streamfeeder. The trial judge adopted the jury’s findings of fact and refused Sure-Feed’s motions for judgment as a matter of law and for a new trial. Sure-Feed appealed to the Federal Circuit. Its principal argument was that Streamfeeder’s hypothetical claim was improperly drafted. The result was a unanimous opinion by Judge Lourie , written for a three-judge panel including Judges Schall and Bryson. Judge Lourie agreed with Sure-Feed, concluding that a legally acceptable hypothetical claim that would encompass Sure-Feed's device would also read on the prior art. Judge Lourie began by observing that Streamfeeder's hypothetical claim was nearly identical to actual claim 1 of the ‘831 patent. Streamfeeder introduced two changes, however. First, Streamfeeder narrowed the gate limitation to consist of a cylindrical member containing elastomeric bands. Second, Streamfeeder broadened the limitation requiring two gate surfaces with two different friction coefficients to recite two surfaces with two different frictional force values in relation to the paper. Judge Lourie then explained that Streamfeeder’s hypothetical claim impermissibly narrowed the gate limitation. He noted that a hypothetical claim analysis is not an opportunity to redraft issued patent claims freely. While the use of a hypothetical claim may permit a minor extension of a claim to cover subject matter that is substantially equivalent to that literally claimed, the Federal Circuit stated, the patentee cannot redraft granted claims in litigation by narrowing and broadening a claim at the same time. Thus, although Streamfeeder was allowed to broaden its claims slightly, it could not narrow them. Judge Lourie noted that Streamfeeder should have employed other procedures, such as reissue at the PTO, to edit the claims more 38 U.S. CLAIM CONSTRUCTION 1997-2002 extensively. The court next concluded that without the narrowed limitation, Streamfeeder's hypothetical claim would read upon the prior art. Judge Lourie stated that key prior art patents disclosed each of the limitations of the hypothetical claim and would have rendered it unpatentable. He further explained that these references were sufficiently significant to the bottom sheet feeder art that persons of skill in the art would have been motivated to combine them. The Federal Circuit also took the opportunity to clarify which party bore the burdens of proof in cases involving hypothetical claims. The court explained that the burden of identifying pertinent prior art references fell to the accused infringer. However, the ultimate burden of persuasion that the hypothetical claim does not ensnare this prior art falls to the patentee. In this case, then, Sure-Feed was charged with identifying references that would render the hypothetical claim unpatentable. Sure-Feed met this burden of production, Judge Lourie concluded, while Streamfeeder failed to present convincing evidence that its hypothetical claim was patentable. As a result, the judgment of infringement was reversed. The Streamfeeder opinion clarifies that patentees may not use a hypothetical claim both to expand an actual claim in order to cover substantially equivalent subject matter, yet narrow other portions of the actual claim in order to avoid prior art. It also squarely states how the hypothetical claim methodology works at trial, by clearly assigning burdens of proof to patentee and accused infringer. Streamfeeder therefore presents a welcome clarification to the Wilson Sporting Goods hypothetical claim analysis. E. The All-Limitations Rule In Warner-Jenkinson, the Supreme Court confirmed the principle that “[e]ach element contained in a patent claim is deemed material to defining the scope of the patented invention, and thus the doctrine of equivalents must be applied to individual elements of the claim, not the invention as a whole.”44 This principle is known as the all- limitations rule. In applying it, a finding of infringement may arise only if each element of a claim is expressed in the accused infringement, either literally or equivalently. The doctrine of equivalents is thus not oriented to the patented invention as a whole, but instead is more strictly employed with respect to each element of a patent claim. 39 U.S. CLAIM CONSTRUCTION 1997-2002 The all- limitations rule is of great practical significance to claims drafters. Consider the example of the following, simplified claims: 1. A fork comprising: a cylindrical handle; and four tines attached to said handle. 2. A fork comprising: a cylindrical handle; a first tine attached to said handle; a second tine attached to said handle; a third tine attached to said handle; and a fourth tine attached to said handle.45 These claims appear to provide the same scope of protection in terms of literal infringement. But suppose a competitor markets a fork with three tines. The all- limitations rule would not bar a finding of equivalent infringement with regard to claim 1. But with respect to claim 2, the absence of the final recited element – “a fourth tine attached to said handle” – would violate the all-limitations rule and prove fatal to the case of equivalency. Although this example is straightforward enough, application of the all- limitations rule has proven rather more difficult in the litigated cases. Consider two cases decided in 2002, Cooper Cameron Corp. v. Kvaerner Oilfield Products, Inc. , 46 where the all- limitations Rule barred a finding of equivalent infringement, and Eagle Comtronics, Inc. v. Arrow Communication Laboratories, Inc.47, where it did not. In Cooper Cameron, the defendant was charged with infringement of U.S. Patent 5,544,707, relating to subsea wellheads. The patented wellhead included a so-called “workover port” that protected the integrity of the well during “workover” activities for repair and maintenance. The asserted claim of the ‘707 patent in part recited “a workover port extending laterally through the wall of the spool tree from between the two plugs.” The accused infringer sold a device with a workover port that was placed above the plugs, rather than between them. Writing for the Federal Circuit, Judge Lourie affirmed the district court’s finding that the accused devi ce did not infringe the ‘707 patent under the doctrine of equivalents. The court explained: Were we to ignore Cooper’s decision to claim in the ‘707 patent a workover port that connects to the assembly only “between” 40 U.S. CLAIM CONSTRUCTION 1997-2002 the plugs, we would vitiate that limitation and run afoul of the all- limitations rule. Moreover, we are not persuaded by Cooper’s argument that the court should have defined the relevant limitation for purposes of the all- limitations rule as the “workover port,” which is in fact present in Kvaerner’s device, though admittedly not at the claimed location between the two plugs. Cooper cannot escape application of the all-limitations rule by recharacterizing its claim so as to ignore a material limitation. 48 This holding should be compared with that of Eagle Comtronics, authored several months later by Judge Linn. The plaintiff there held U.S. Patent No. 5,662,494, relating to electrical signal filters. The claimed invention called for an improved cable filter structure that provided a sealed collet assembly to prevent moisture or other contaminants from entering the filter. The ‘494 patent claims required that the collet assembly comprise a front cap, a rear insert body, and a seal located between the front cap and rear insert body. In contrast, the accused products consisted of a one -piece or “integrated” insulation block. According to the accused infringer, because the patent’s claims required the collet assembly to consist of three separate structural elements, a one-piece collet assembly could not infringe under the all-limitations rule. Judge Linn disagreed that the all-limitations rule barred a finding of equivalent infringement in that case. The Federal Circuit reasoned that “whether or not a limitation is deemed to be vitiated must take into account that when two elements of the accused device perform a single function of the patented invention, or when separate claim limitations are combined into a single element of the accused device, a claim limitation is not necessarily vitiated, and the doctrine of equivalents may apply if the differences are insubstantial.” Noting that “one-to-one correspondence of components is not required,” the Court of Appeals remanded the case to the district court for further proceedings. It seems difficult to reconcile the holdings of Cooper Cameron and Eagle Comtronics. It would have just as easy for the Cooper Cameron court to agree with the patentee and hold that the limitation at issue was a “workover port,” rather than a “workover port between the two plugs.” As well, the Eagle Comtronics court could have held that the level of detail in the patentee’s claims required a differentiated, rather than an integrated collet assembly. The Cooper Cameron opinion says only that the claim’s identification of the location of the workover port was a “material limitation” – 41 U.S. CLAIM CONSTRUCTION 1997-2002 a determination that is not further quantified and seems, at present, a quite subjective inquiry. At the end of the day, the cases essentially leave open the question of what claim language constitutes a limitation for purpose of the all- limitations rule. As a result, claims drafters are well advised to draft as sparse and abstract claims as possible, thereby decreasing the number of limitations that may restrict the scope of patent protection. Whatever the merits of the all- limitations rule, the Supreme Court opinion in Warner-Jenkinson broadly upheld it. 49 Interestingly, the facts of Warner-Jenkinson did not provide an apt vehicle for the approving statement of Justice Thomas. Recall that the crucial claim limitation there identified the pH at which a dye purification process should be performed. No one doubted that the accused process occurred at some acidity -- it could hardly be otherwise. Warner-Jenkinson simply wasn’t a case of a missing claim limitation. Nor did the Court see fit to discuss or even cite the crucial Pennwalt and Corning Glass opinions. The lower courts have always found the dicta of the Supreme Court extremely persuasive, however, and the all- limitations Rule remains with us as a significant constraint upon the doctrine of equivalents. F. The Public Dedication Doctrine In its 2002 decision in Johnson & Johnston Associates, Inc. v. R.E. Service Co., Inc. , 50 the en banc Federal Circuit recognized an additiona l restraint upon the doctrine of equivalents. Under the Johnson & Johnston “public dedication doctrine,” subject matter that is disclosed in a patent, but not claimed, may not be appropriated through the doctrine of equivalents. Such unclaimed subject matter is considered to have been deliberately disclaimed and therefore dedicated to the public. The Johnson & Johnston case involved a recurring situation: sometimes the claims of a patent are not as broad as the technical disclosure contained in that patent’s written description. The patent at issue in that case, owned by Johnson & Johnston, concerned a printed circuit board. A printed circuit board is a thin plate on which computer chips or other electronic components are placed. The patent’s specification explained that the substrate of the circuit board could be manufactured of a number of materials, including aluminum, steel and nickel. However, the patent’s claims were limited to a substrate made of aluminum. Accused infringer R.E. Service manufactured printed circuit boards with a steel substrate. Prior to Johnson & Johnston, competing views arose as to the impact 42 U.S. CLAIM CONSTRUCTION 1997-2002 of described, yet unclaimed subject matter upon the Doctrine of Equivalents. One possibility is that disclosure within the patent instrument acts in favor of a finding of equivalency. Under this perspective, the patent as a whole provides notice to interested competitors what the patentee regarded as an equivalent at the time the patent was granted. At the other extreme, as represented by the 1996 Federal Circuit decision in Maxwell v. J. Baker, Inc., 51 is the position that subject matter disclosed but not claimed in a patent application is, as a matter of law, dedicated to the public. Under this view, by failing to claim the full extent of the disclosed subject matter, an applicant deprives the PTO of the opportunity to consider whether this subject matter is patentable. Allowing an applicant to obtain narrow claims from the PTO, and then assert broader protection for unclaimed alternatives described in the specification, would defeat the fundamental principle that a patent’s claims define its scope of proprietary rights. A third position, adopted by the 1998 Federal Circuit opinion in YBM Magnex, Inc. v. International Trade Commission, 52 was that no per se rule should dictate whether subject matter included in the written description but not claimed is equivalent to the claimed invention. Proponents of this view emphasized that the doctrine of equivalents seeks to establish a just balance between the purpose of claims to provide notice of what is patented, and the judicial responsibility to avoid a "fraud on the patent" based on insubstantial changes from the patented invention. Whether the accused infringement was disclosed, but not claimed in the asserted patent simply formed one of many factors to consider in the equivalency determination. The en banc Federal Circuit resolved its inconsistent precedent in Johnson & Johnston by adopting Maxwell and expressly overruling YBM Magnex. The Federal Circuit reasoned that allowing the doctrine of equivalents to extent to disclosed, but unclaimed subject matter would conflict with the primacy of the claims in defining the scope of the patentee’s exclusive right. Otherwise patent applicants would be encouraged to present a broad disclosure in the specification of the application and file narrow claims, avoiding examination of broader claims that the applicant could have filed consistent with the specification. Applying this principle to the facts of the case, the Federal Circuit concluded that R.E. Service did not infringe the Johnson & Johnston patent under the doctrine of equivalents. In the wake of the Johnson & Johnston decision, patent applicants have been encouraged to draft claims with increasing care. Commentators have 43 U.S. CLAIM CONSTRUCTION 1997-2002 suggested that applicants draft a broad generic claim as well as more specific claims to each embodiment that can be envisioned. If a patent issued prior to the Johnson & Johnston case, and failed to claim disclosed subject matter, the Federal Circuit further suggested that the patentee pursue a reissue or continuation application as the circumstances allow. 53 Most apparent is the Federal Circuit has introduced a significant restraint upon the doctrine of equivalents that will require patent applicants to craft claims with increasing precision. III. Construction of Means-Plus -Function Claims Functional claims define an invention in terms of what it does rather than in terms of its structure. Patent practitioners generally term such claims as “means-plus- function” claims, a reference to the final paragraph of § 112. Section 112 ¶ 6 provides that an element in a combination claim may be expressed as a means or step for performing a specified function. It further directs that such a claim shall be construed to cover the corresponding structure, material or acts described in the specification and equivalents thereof. In the following simplified claim, elements (a) and (b) are expressed structurally, while element (c) is drafted in means-plus-function form. 1. A hammer, comprising: (a) a head; (b) a handle; and (c) means for attaching said head and said handle. To determine the literal scope of element (b), § 112 ¶ 6 directs that the reader turn to the patent’s specification. Assume that the patent’s specification explained that the “handle and head may be attached through the use of a nail or bolt.” In this case claim 1 should be literally read to cover the following combination: 1. A hammer, comprising: (a) a head; (b) a handle; and (c) [a nail, bolt and equivalents thereof] for attaching said head and said handle. Because § 112 ¶ 6 refers to “a claim for a combination,” a claim that consists of a single means for accomplishing a particular task is improper.54 44 U.S. CLAIM CONSTRUCTION 1997-2002 Section 112 ¶ 6 owes its origin to increasing judicial hostility to functional claiming in the first half of the Twentieth Century. This trend culminated in the 1946 Supreme Court decision Halliburton Oil Well Cementing Co. v. Walker.55 Walker’s patent claims, directed towards an apparatus for measuring the depth of oil wells, included a number of limitations cast in means -plus- function format. These included “means communicating with said well for creating a pressure impulse in said well” and “echo receiving means.” The Court struck down Walker’s claims, reasoning that they because they were not tied to the specific structures Walker had invented they were overbroad and indefinite. According to the Court, “unless frightened from the course of experimentation by broad functional claims like these, inventive genius may evolve many more devices to accomplish the same purpose.”56 The patent bar protested mightily against Halliburton Oil, which cast great doubt upon a large number of issued patents that had employed a functional claiming style. In response Congress amended the patent statute by adding the provision now codified at § 112 ¶ 6. 57 The intent of Congress was to establish clear parameters within which functional claims could be written and interpreted.58 Two principal issues concern the operation of § 112 ¶ 6. The first concerns which claims should be read as invoking § 112 ¶ 6 and its interpretational protocol. Second, the precise parameters of the closing phrase of § 112 ¶ 6, “and equivalents thereof,” is of paramount concern during the interpretation of means plus function claims. This report address these subjects in turn. 45 U.S. CLAIM CONSTRUCTION 1997-2002 A. Triggering the § 112 ¶ 6 Interpretational Protocol The Federal Circuit has held that the use of the word “means” triggers a presumption that the inventor used this term to invoke the statutory mandates for means-plus- function clauses. 59 However, not every use of the term “means” indicates that the claim is a means-plus- function claim. The presumption of invoking § 112 ¶ 6 can be rebutted if the claim language does not link the term “means” to a recited function. Cole v. Kimberly-Clark Corp. 60 demonstrates how courts determine whether a particular claims is drafted in means-plus- function format or not. There the Federal Circuit construed a claim involving disposable diapers with sides that easily tear open to facilitate removal of a soiled brief. Among the claim elements was “perforation means extending from the leg band means to the waist band means through the outer impermeable layer means for tearing the outer impermeable layer means for removing the training brief in the case of an accident by the user.” The district court determined on summary judgment that a “perforation means” is merely a “perforation” and that the bonded tearable side seams on the accused briefs were not perforations. The Federal Circuit affirmed on appeal, also declining to apply §112 ¶ 6. According to the court, the “perforation means . . . for tearing” element of Cole’s claim failed to satisfy that statute because it recited not only perforations, the structure supporting the tearing function, but also their location (extending from the leg band to the waist band) and range (extending through the impermeable layer). The court reasoned that a claim that cited such detailed structure should not qualify as functional. The term “means” is not the only one which will invoke §112 ¶ 6. In Mas-Hamilton Group v. LaGard, Inc., 61 the Federal Circuit considered a claim that recited a “lever moving element.” The court held that §112 ¶ 6 applied to this claim element as well, despite its failure to use the usual catch word “means.” The court recognized that many devices take their name from the functions they perform, such as a screwdriver, brake or lock, and that the use of these terms in a claim would count as a structural, rather than functional definition. However, the phrase “lever moving element” lacked a well- understood structural meaning in the art. As a result, the court held that the claim defined this element functionally rather than structurally. Its literal scope was therefore restricted to structures disclosed in the specification and equivalents thereof that performed the identical function, rather than to any conceivable device for moving a lever. Similarly, a PTO Board decision has found claim language reciting “a jet driving device” to invoke §112 ¶ 6.62 46 U.S. CLAIM CONSTRUCTION 1997-2002 B. The Scope of Structural Equivalency A second issue concerning § 112 ¶ 6 is the operation of its equivalency provision. The statute calls for functional claim language “to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.” The Federal Circuit has interpreted this provision to hold that for a means-plus-function limitation to read on an accused device, the accused device must employ means identical or equivalent to the structures, material, or acts described in the patent specification. The accused device must also perform the identical function as specified in the claims.63 The Federal Circuit has provided that, for purposes of § 112 ¶ 6, an equivalent results from an insubstantial change which adds nothing of significance to the structure, material, or acts disclosed in the patent specification. 64 The court often expresses this concept through the succinct phrase “structural equivalency.” 65 For example, recall that in the hammer example above, the specification explained that “the handle and head may be attached through the use of a nail or bolt.” A screw would likely be considered a structural equivalent to a nail or bolt, but the use of glue to secure the head and the handle would probably not be judged structurally equivalent. The court has stated often stressed that structural equivalency under § 112 ¶ 6 differs from equivalency under the doctrine of equivalents. For present purposes, it is enough to recognize that despite their demands for clear claiming, the courts have been willing to find liability even where an accused infringer diverges slightly from a strict reading of the claims. An equivalent under the doctrine results from an insubstantial change which, from the perspective of persons of ordinary skill in the art, adds nothing of significance to the claimed invention. An equivalent under the doctrine, though not literally meeting the claims, still infringes the patent. 66 As can be imagined, the contrast between structural equivalency and equivalency under the doctrine has proven a rather subtle affair. These distinctions ma y largely be academic given that claims written in means-plus- function format are as subject to the doctrine of equivalents as any other sort of claim. The extent of the “equivalents thereof” under § 112 ¶ 6 continues to confound all who confront this issue. Not only have decades of experience failed to provide a sharp definition of the doctrine of equivalents, the patent law does not readily yield alternative conceptions of equivalency. Like the doctrine of equivalents, § 112 ¶ 6 equivalence must focus upon the substantiality of the differences between the claimed element and a component 47 U.S. CLAIM CONSTRUCTION 1997-2002 of the accused product. Still, the Supreme Court’s treatment of § 112 ¶ 6 in Warner-Jenkinson requires that some distinction exist between these two notions of equivalency.67 A Federal Circuit opinion subsequent to Warner-Jenkinson, Al-Site Corp. v. VSI International, Inc., 68 presents the most thoughtful attempt to demarcate the scope of § 112 ¶ 6 equivalents. Judge Rader identified three differences between the doctrine of equivalents and § 112 ¶ 6 equivalents. The first is a matter of timing. Under Warner-Jenkinson, the proper time for evaluating insubstantial differences under the doctrine of equivalents is at the time of infringement. However, an equivalent under § 112 ¶ 6 cannot embrace technology developed after the time the patent issues because the PTO fixes the literal meaning of the claims at that time. Therefore, an “after-arising” technology could infringe under the doctrine of equivalents without infringing literally as a § 112 ¶ 6 equivalent. Second, under the literal language of § 112 ¶ 6, the equivalent must perform the identical function recited in the claim element. Under the function-way-result test, the doctrine of equivalents may be satisfied when the function performed by the accused device is only substantially the same as that of the claimed invention. Finally, the Al-Site court employed the term “structural equivalents” in connection with § 112 ¶ 6. This wording appears to demand an equivalency comparison that exclusively concerns physical structure. Under this view, § 112 ¶ 6 provides a scope of equivalents far more narrow than that of the doctrine of equivalents. Suppose, for example, that a patent claimed a “means for securing said engine to said frame.” The corresponding portion of the specification describes the use of bolts to attach the frame to the engine. Although nails and screws have the same structure as bolts, such well-known securing mechanisms as adhesives or magnets do not. Professor Mark Janis has observed that such a cramped claim coverage is little more than a subterfuge for no equivalency. 69 As the term “structural equivalents” does not appear in § 112 ¶ 6, the legitimacy of this interpretation is also suspect as a textual matter. 48 U.S. CLAIM CONSTRUCTION 1997-2002 C. Functional Claims for Software Inventions In Globetrotter Software, Inc. v. Elan Computer Group, Inc ., 70 the Federal Circuit considered section 112, paragraph 6 in the context of a software -related invention. U.S. Patent No. 5,390,297 pertains to a license management system for controlling the number of concurrent copies of a program in use on a computer network. The invention allows software vendors to restrict the number of individuals using a computer program at one time, in keeping with the license obtained by a customer. The specification of the ‘297 patent discloses two embodiments. The first is called the “license transfer” technique. In this embodiment, each individual license is associated with a separate file. When a computer user wishes to use an application, the application software will operate only if the copy the user is trying to run corresponds to a license file. If that license file is not available, the computer system will search for another available license file. The other embodiment is called the “license pool” arrangement. Here, several licenses may be contained in a single license file. A user’s request to use a copy of the application software is granted only if the pool of licenses in the license file has not been exhausted by other similar requests. Each time a request is successfully made to run an application, a counter is decreased to indicate that one less license is available. Globetrotter Software Inc., the owner of the ‘297 patent, sued Elan Computer Group, Inc. for patent infringement. Early during this litigation, Globetrotter requested that the trial court issue a preliminary injunction against Elan. This request prompted the trial judge to interpret claim 55 of the ‘297 patent. With emphasis added, claim 55 of the ‘297 patent read as follows. 55. A license management system for limiting the number of copies of a given computer program that are permitted to run simultaneously on one or more nodes of a network in which said nodes are connected, said limiting being according to the number of licenses for said given computer program that are authorized for said network; said system comprising: license file means on at least one of said nodes for storing at least one and up to a selectable authorized number of said licenses; 49 U.S. CLAIM CONSTRUCTION 1997-2002 program storage means for storing a copy of said given computer program on at least one of said nodes without limiting the running of said given computer program to running on said one node; and license management means responsive to a request to run, said request being from one of said copies at a particular one of said nodes, for searching as many nodes as are necessary to locate one of said license file means that has a license that is available for authorizing a copy of said given computer program to run at said particular one of said nodes; said license management means being responsive to said search not locating any of said license file means having an available license for returning to said requesting copy a message preventing said copy from running in response to said request. The specification of the ‘297 patent described the structure necessary to perform the recited functions of the license management system of claim 55. In particular, the specification described the use of a unique identification (“UID”) to perform the function of “storing at least one and up to a selectable authorized number of said licenses.” Elan argued that because the products that were accused of infringing the ‘297 patent did not contain a UID, then they could not infringe. The district court agreed with Elan. The trial judge first identified the phrase “license file means . . . for storing” was drafted in means-plus-function format. The court also concluded that in keeping with section 112 paragraph 6, the ‘297 patent required that “license file means for storing” must contain a UID. Because Elan’s devices did not include a UID, they could not infringe. As a result, the trial judge decided not to issue a preliminary injunction. Globetrotter appealed to the Federal Circuit. Judge Bryson, Senior Judge Plager and Judge Dyk heard the appeal. The majority decision, joined by all three judges, was written by Judge Dyk. The Federal Circuit affirmed the claim interpretation provided by the trial judge. According to Globetrotter, the trial judge’s claim interpretation was improper. Globetrotter stated that claim 55 recited only the function of storing a license, and that a UID was not necessary to perform this function. Globetrotter further argued that “storing a license” and “assigning a UID” were separate functions performed by different means. Moreover, 50 U.S. CLAIM CONSTRUCTION 1997-2002 Globetrotter asserted that the district court focused on the license transfer embodiment and not the license pool embodiment. The Federal Circuit disagreed with each of Globetrotter’s contentions. According to Judge Dyk, the ‘297 patent specification clearly identified that a UID for the license file is an essential part of the structure required to perform the function of “storing at least one and up to a selectable authorized number of said licenses” performed by the license file means. Each time a license was stored, Judge Dyk observed, that storage location was assigned a UID. That UID was then compared with the UID written into a directory to determine whether a particular license is in use. This arrangement took place whether the license transfer or the license pool embodiment was in use. Judge Dyk’s review of the drawings of the ‘297 patent confirmed this result. Each of the relevant figures in the patent indicated that a UID is required to store licenses. As well, none of the drawings showed any alternate method of storing a license with a license file means that would not require a UID. The Federal Circuit therefore concluded that a UID was necessary to perform the function of storing licenses by the license file means. Because Globetrotter had not shown that Elan’s accused software used a UID, Elan could not infringe the ‘297 patent. As a result, the Federal Circuit affirmed the district court’s denial of Globetrotter’s request for a preliminary injunction. Globetrotter Software, Inc. v. Elan Computer Group, Inc . demonstrates that drafters of means-plus-function claims should be wary. Many drafters of software patent claims have assumed that means-plus- function claims provide broader coverage than structural claims. This assumption is not necessarily a sound one. Here, the Federal Circuit gave a very narrow interpretation to a means-plus- function claim. The court effectively made a feature that was not expressly claimed, but was discussed within the specification, a limitation of the claims. Software patent claims drafters should be sure to choose how they write their means-plus- function claims very carefully. They should avoid characterizations of the claimed function that might encourage a court to add limitations based on embodiments discussed in the specification. In addition, within the specification, patent drafters should avoid characterizing as essential any elements of the invention. 51 U.S. CLAIM CONSTRUCTION 1997-2002 D. Step-Plus -Function Claims Section 112 ¶ 6 also refers to process or method claims, providing that “an element in a claim for a combination may be expressed as a . . .step for performing a specified function without the recital of . . . acts in support thereof.” The Federal Circuit has acknowledged the propriety of step-plus- function claims,71 but scant case law has thus far been devoted towards interpreting them. That court has been quick to point out that not every claimed process step invokes the interpretational method of § 112 ¶ 6. Step-plus- function claim elements provide a step, or generic description of a portion of a process, without the recital of more specific acts to perform that abstract function. Such claims should be construed as covering the corresponding “acts described in the specification and equivalents thereof.” For example, suppose that one of the elements of a process claim was “the step of raising the temperature to 300 degrees.” If the patent’s specification described the use of an oven to raise the temperature to 300 degrees, the literal coverage of the claim would extend to the use of an oven and equivalent acts. In Seal -Flex Inc. v. Athletic Track and Court Construction, 72 the Federal Circuit considered the possibility of interpreting a method as a “step-plus- function” claim, analogous to means-plus- function claims under 35 U.S.C. § 112 ¶ 6. The plaintiff in this case was Seal-Flex Inc., owner of U.S. Patent No. 4,529,622. The ‘622 patent relates to a method of making an all- weather athletic running track. In 1991, Seal-Flex brought a patent infringement suit against Athletic Track and Court Construction (AT&CC) in the District Court for the Eastern District of Michigan. In 1994, the district court had ruled that the patent was invalid under 35 U.S.C. § 102(b) due to the on-sale bar. 73 However, the Federal Circuit reversed and remanded. The Federal Circuit held that the district court had improperly applied a "sufficiently complete" standard for the on-sale bar.74 On remand, a jury determined that AT&CC had literally infringed the '622 patent. Following the disposition of various post- verdict motions, the district court entered judgment for Seal- Flex. AT&CC appealed to the Federal Circuit. A panel consisting of Judges Newman, Rader and Bryson heard the appeal. The appeal resulted in a so-called “per curiam” opinion; that is, an opinion written by the court as opposed to an individual judge. Both Judges Rader and Bryson filed concurring opinions, with Judge Newman joining the latter. The per curiam opinion by noting claim 1 of the ‘622 patent, which recited in part: "A method for constructing an activity mat over a foundation 52 U.S. CLAIM CONSTRUCTION 1997-2002 comprising the steps of ... spreading an adhesive tack coating for adhering the mat to the foundation over the foundation surface." According to agreement between the parties, the district court interpreted this "spreading" step as invoking § 112 ¶ 6. That provision requires each functional element in a combination claim to be interpreted by reference to the corresponding structure, material, or acts described in the specification and their equivalents. At trial, the district court limited the term "adhesive tack coating" to the "suitable tack coating materials" disclosed in the specification and to those which "result[] from an insubstantial change which adds nothing of significance to the materials disclosed.” Based on this instruction, the jury determined that latex qualified as an equivalent tack coating material to those disclosed in the specification of the '622 patent. On appeal, AT&CC contended that it did not literally infringe the '622 patent because it used latex to perform the function recited in the spreading step, not the tack coating materials that were disclosed in the specification. The Federal Circuit did not side with AT&CC. According to the per curiam court, the testimony of both parties as well as the prosecution history of the '622 patent fully supported the jury's conclusion that latex is the equivalent of the "suitable tack coating materials" disclosed in the specification and is therefore literally an "adhesive tack coating" as required by the claim. Although the Federal Circuit admitted that the specification did not expressly list latex as a suitable tack coating material, the record showed that latex is an equivalent of the adhesive tack coatings described in the specification. Both latex and one of the materials described in the specification, emulsified asphalt, were water based. As well, latex and emulsified asphalt had similar viscosities and both could be applied at ambient temperatures with the same spray equipment, the Federal Circuit explained. The per curiam court also pointed to the admission of AT&CC's president, who explained that he used latex when a project called for asphalt emulsion. The Federal Circuit reasoned that this testimony showed that he considered latex to be equivalent to the materials disclosed in the specification. The court also reasoned that applicant's addition of the word "adhesive" before the words "tack coating" clarified its intention to cover broadly any material, including latex, that would serve as an adhesive tack coating. The per curiam court therefore concluded that the jury's finding that the ‘622 patent was literally infringed was supported by substantial evidence and affirmed the judgment. Judge Rader issued a separate opinion which concurred in the result of 53 U.S. CLAIM CONSTRUCTION 1997-2002 the per curiam court. According to Judge Rader, however, the court should have determined whether § 112 ¶ 6 applied to the claim element at issue. Judge Rader believed that a proper determination of this case required the determination whether the “spreading” step was a means-plus-function element, a step-plus- function element, or neither. Judge Rader explained that the district court apparently believed the element was in step-plus- function form, but gave further instructions implicating means-plus- function considerations. Although similar, Judge Rader continued, these two sots of elements are not identical and require distinct analyses. Analyzing the text of § 112 ¶ 6, Judge Rader maintained that the terms "structure" and "material" are associated with "means," while the term "acts" is associated with the term "step.” Thus, a claim element deserves means-plus- function treatment when expressed as a means for performing a specified function without the recital of structure or material. On the other hand, a claim element deserves step-plus- function treatment when expressed as a step for performing a specified function without the recital of supporting acts. The difficulty of distinguishing acts from functions in step-plus- function claim elements makes the identification of such claims very difficult, Judge Rader observed. Thus, the phrase "step for" in a method claim raises a presumption that § 112 ¶ 6 applies. However, even when a claim element uses the step-plus- function format, § 112 ¶ 6 does not apply when the claim limitation itself recites sufficient acts for performing the specified function. Judge Rader explained that the "underlying function" of a method claim element corresponds to what that element ultimately accomplishes in relationship to what the other elements of the claim and the clai m as a whole accomplish. "Acts," on the other hand, correspond to how the function is accomplished. Therefore, claim interpretation focuses upon the underlying function of the claim limitation in relation to what is accomplished by the other limitations and the claim as a whole. If a claim element recites only an underlying function without acts for performing it, then § 112 ¶ 6 applies even without express step-plus function language. Judge Rader summarized the Federal Circuit’s approach towards step-plus-function claims as follows: In sum, similar to means-plus- function claims, this court employs a straightforward analysis for identifying a step-plus- function claim. If the claim element uses the phrase "step for," then § 112 ¶ 6 is presumed to apply. Because the 54 U.S. CLAIM CONSTRUCTION 1997-2002 phrasing "step for" would appear to claim every possible act for performing the recited function, in keeping with § 112 ¶ 6, such a claim covers only the specific acts recited in the specification for performing that function, and equivalent acts. On the other hand, the term "step" alone and the phrase "steps of" tend to show that § 112 ¶ 6 does not govern that limitation. Accordingly, this court has similarly denied step-plus- function treatment to method claims which use the conventional "steps of" language.75 Judge Rader then applied this methodology to the case at hand. Claim 1 of the '622 patent used the phrase "steps of" in the preamble to introduce several several "steps." Among these steps was the contested step of "spreading.” Because the phrase "step for" was not used, Judge Rader reasoned, the verb "spreading" recites an act rather than a function. Although claim 1 did recite several "steps for" constructing an activity mat, Judge Rader continued, the recitation of the overall function of the claim in the preamble did not convert each element into an act so as to preclude application of § 112 ¶ 6. Rather, according to the statute, each element must be examined individually to determine whether it merely recites that element's function or a specific act for performing it. Judge Rader then explained that the function of the element at issue, namely "adhering the mat to the foundation," appears explicitly in the claim language. The preposition "for" introduces this underlying function and links the "adhering" function to the act of "spreading." Accordingly, he concluded, because this claim limitation did not invoke § 112 ¶ 6. Not only was that claim element not drafted in step-plus-function form, according to Judge Rader, it was not a means-plus- function element either. Judge Rader observed that the claim element did not use the traditional term "means," and that the term "adhesive tack coating" disclosed material for performing the claimed function. Although Judge Rader concluded that the spreading step of claim 1 did not invoke § 112 ¶ 6, he also found that the district court’s erroneous use of this statute was harmless. Judge Rader reasoned that without the limiting strictures of § 112 ¶ 6, the claim term "adhesive tack coating" might have a broader scope. Indeed, applying § 112 ¶ 6, the district court limited the term to the "suitable tack coating materials" disclosed in the specification and those which "result[] from an insubstantial change which adds nothing of significance to the materials disclosed." Based on this instruction, the jury determined that latex qualified as an equivalent structure to those disclosed in 55 U.S. CLAIM CONSTRUCTION 1997-2002 the specification of the '622 patent. This determination demonstrates that the jury found latex to be an "adhesive tack coating" as required by the claim. Under a correct claim construction, therefore, the jury would also have found literal infringement. As a result, Judge Rader concurred in the result reached by the per curiam court. Judge Bryson w rote a separate concurring opinion that was joined by Judge Newman. He criticized Judge Rader for addressing the step-plus- function issue in his concurring opinion. Neither litigant objected at any point to the district court's characterization of claim 1 as subject to § 112 ¶ 6, Judge Bryson observed, nor was it necessary to reach this issue to resolve this appeal. The usual rule that appellate courts limit themselves to issues that both parties have preserved below and raised on appeal serves important interests of both trial and appellate processes, according to Judge Bryson. Judge Rader’s concurring opinion presents the view of at least one Federal Circuit jurist towards step-plus- function claims under § 112 ¶ 6. As such, claims drafters may wish to review the often confusing distinction between steps and acts under that statute. As the invocation of § 112 ¶ 6 can result in a significant narrowing of claim coverage, the drafter of a method claim should carefully consider whether he wishes to employ that statute or not. IV. Concluding Observations The past five years have led to many significant developments in the construction of U.S. patent claims. Regarding basic tenets of claim construction, the Federal Circuit has continued to develop the basic Markman framework established in the mid-1990's. The most notable development is the relaxation of the strict categorization of dictionaries and learned treatises as extrinsic evidence. The Federal Circuit has also added more precision to its interpretation of the preamble and transition phrase, and continued its in adamant refusal to apply the reverse doctrine of equivalents. One question the Markman interpretational protocol did not answer remains an issue that still troubles the Federal Circuit: the appropriate role of the specification as a mechanism for interpreting, or perhaps limiting, the scope of the claims. The U.S. courts been most active in the area of the doctrine of equivalents. The 1997 Supreme Court decision in Warner-Jenkinson resolved many issues concerning the doctrine, including its continued viability, the timing of the equivalency inquiry, and the existence of additional principles that restrain the scope of the doctrine. Following 56 U.S. CLAIM CONSTRUCTION 1997-2002 Warner-Jenkinson, the Federal Circuit became notably more conservative in its use of the doctrine of equivalency. Stressing precision claim drafting and rejecting patentee attempts to employ the doctrine as a mechanism for correcting drafting errors, the Federal Circuit increasingly denied the These developments culminated in the Federal Circuit’s decision in Festo, in which the en banc court held that prosecution history estoppel resulting in a “strict bar” denying the patentee any scope of equivalents. The Supreme Court later overturned the Federal Circuit on that point, but its decision largely vindicates the increasingly restrictive practices regarding the doctrine of equivalents of the Court of Appeals. The Supreme Court identified three scenarios that would avoid prosecution history estoppel: (1) unforeseeable equivalents for which the applicant could not have fashioned literal claim coverage, (2) amendments with only tangential relations to the equivalent in question, and (3) some other reason frustrating accurate description in claims. One wonders whether these standards will soon apply not just to avoiding prosecution history estoppel, but to the equivalency inquiry more generally. The Federal Circuit has done much to clarify the interpretation of means-plus- function claims. Patent applicants are much better informed about how to trigger the interpretational protocol of § 112 ¶ 6. The entire patent community is much better informed about the scope of equivalency such claims will enjoy as well. In addition, the Federal Circuit has spoken about step-plus- functional claims and applied § 112 ¶ 6 to the crucial area of software technologies. Interestingly, in a time of sweeping changes, the U.S. patent law is in many ways becoming more certain. Few patents will be awarded a range of protection beyond the literal scope of their claims. On the other hand, during the same time period, the Federal Circuit has held that virtually all innovations, no matter what the discipline from which they originate, are patentable.76 Nonobviousness is becoming increasingly vitiated, with the more objective novelty requirement left as the chief gatekeeper to patentability. 77 In many ways the U.S. has become more like the traditional, rather stereotyped view of the Japanese system. In such a regime where many patents issue, but few are awarded a scope beyond their literal wording. One difficulty with this regime is that experience teaches us that firms can obtain many small patents in the place of a few big ones -- at a considerable administrative to their local patent office. As its patent system continues to evolve, only the future will reveal the wisdom of these choices upon U.S. innovation policy. 57 U.S. CLAIM CONSTRUCTION 1997-2002 * 1 1 Professor of Law, Georgetown University, Washington, DC, USA. E.g., 35 U.S.C. § 271(a) (2000). 52 F.3d 967 (Fed. Cir. 1996), aff’d, 517 U.S. 370 (1996). 182 F.3d 1298 (Fed. Cir. 1999). 90 F.3d 1576 (Fed. Cir. 1996). 517 U.S. 370 (1996), aff’g 52 F.3d 967 (Fed. Cir. 1995). 526 U.S. 137 (1999). 191 F.3d 1356 (Fed. Cir. 1999). 197 F.3d 1377 (Fed. Cir. 1999). 261 F.3d 1329 (Fed. Cir. 2001). 179 F.3d 1350 (Fed. Cir. 1999). 30 F.3d 1475, 1479 (Fed. Cir. 1994). 517 U.S. 370 (1996), aff’g 52 F.3d 967 (Fed. Cir. 1995). 239 F.3d 1239 (Fed. Cir. 2001). 285 F.3d 1362, 62 USPQ2d 1349 (Fed. Cir. 2002). 287 F.3d 1108 (Fed. Cir. 2002). 339 U.S. at 608-09. See SRI Int’l v. Matsushita Elec. Corp., 775 F.2d 1107, 227 USPQ 577 (Fed. Cir. 1985). 2 3 4 5 6 1 1 2 3 4 5 6 7 8 9 10 See Robert P. Merges & Richard R. Nelson, On the Complex Economics of Patent Scope, 90 COLUM . L. REV. 839 (1990). 11 12 279 F.3d 1357 (Fed. Cir. 2002). See Toro Corp. v. White Consol. Industries, 266 F.3d 1367, 1370 (Fed. Cir. 2001). 13 Martin J. Adelman & Gary L. Francione, The Doctrine of Equivalents in Patent Law: Questions that Pennwalt Did Not Answer, 137 U. PA. L. R EV. 673 (1989). 14 520 U.S. 17 (1997). 520 U.S. at 25-27. 15 58 U.S. CLAIM CONSTRUCTION 1997-2002 16 17 18 520 U.S. at 27-28. 520 U.S. at 37. 520 U.S. at 34-36. 339 U.S. 605 (1950). 62 F.3d 1512, 1518 (Fed. Cir. 1995). 520 U.S. at 39-40. 520 U.S. at 36. 19 20 21 22 23 Roton Barrier Inc. v. The Stanley Works, 79 F.3d 1112, 1128, 37 USPQ2d 1816, 1828 (Fed. Cir. 1996) (Nies, J., concurring). 24 25 26 27 229 F.3d 1091 (Fed. Cir. 2000). 229 F.3d at 1106. 126 F.3d 1420 (Fed. Cir. 1997). 126 F.3d at 1425. 35 U.S.C. § 251 (2000). Ibid. 35 U.S.C. § 252 (2000). E.g., Loral Fairchild Corp. v. Sony Corp., 181 F.3d 1313 (Fed. Cir. 1999). 28 29 30 31 32 This example is loosely based upon Intervet America, Inc. v. Kee-Vet Laboratories, Inc., 887 F.2d 1050 (Fed. Cir. 1989). 33 34 35 See Cybor Corp. v. FAS Technologies, Inc., 138 F.3d 1448 (Fed. Cir. 1998) (in banc). Warner-Jenkinson, 520 U.S. at 33. Id. 234 F.3d 558 (Fed. Cir. 2000) (en banc). 234 F.3d at 569. 535 U.S. 722, 122 S.Ct. 1831 (2002). 122 S.Ct. at 1839-40. 36 37 38 39 59 U.S. CLAIM CONSTRUCTION 1997-2002 40 See John R. Thomas, On Preparatory Texts and Proprietary Technologies: The Place of Prosecution Histories in Patent Claim Interpretation, 47 UCLA L. REV. 183 (1999). 41 175 F.3d 974 (Fed. Cir. 1999). [1991] R.P.C. 597. 904 F.2d 677 (Fed. Cir. 1990). 520 U.S. at 29. See M ARTIN J. ADELMAN ET AL., PATENT LAW : CASES AND MATERIALS 947 (1998). 291 F.3d 1317 (Fed. Cir. 2002). 305 F.3d 1303 (Fed. Cir. 2002). 291 F.3d at 1322. 520 U.S. at 28-30. 285 F.3d 1046 (Fed. Cir. 2002). 86 F.3d 1098 (Fed. Cir. 1996). 145 F.3d 1317 (Fed. Cir. 1998). 285 F.3d at 1055. See In re Hyatt, 708 F.2d 712 (Fed. Cir. 1983). 329 U.S. 1 (1946). 329 U.S. at 12. Warner-Jenkinson Co. v. Hilton Davis Chem. Co., 520 U.S. 17 (1997). 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 Dawn Equip. Co. v. Kentucky Farms, Inc., 140 F.3d 1009 (Fed. Cir. 1998) (Plager, J., additional views). 59 York Products Inc. v. Central Tractor Farm & Family Center , 99 F.3d 1568 (Fed. Cir. 1996); Greenberg v. Ethicon Endo-surgery, Inc., 91 F.3d 1580, 1584 (Fed. Cir. 1996). 60 61 102 F.3d 524, 531 (Fed. Cir. 1996). 156 F.3d 1206 (Fed. Cir. 1998). Ex parte Stanley, 121 USPQ 621 (PTO Bd. 1958). King Instruments Corp. v. Perego, 65 F.3d 941 (Fed. Cir. 1995), cert. denied, 517 U.S. 1188 62 63 60 U.S. CLAIM CONSTRUCTION 1997-2002 (1996). 64 65 Valmont Indus., Inc. v. Reinke Mfg. Co., 983 F.2d 1039 (Fed. Cir. 1993). Laitram Corp. v. Rexnord, Inc., 939 F.2d 1533 (Fed. Cir. 1991). See Warner-Jenkinson Co. v. Hilton Davis Chem. Co., 520 U.S. 17 (1997). See 520 U.S. at 27. 174 F.3d 1308 (Fed. Cir. 1999). 66 67 68 69 See Mark Janis, Who’s Afraid of Functional Claims? Reforming the Patent Law’s § 112 ¶ 6 Jurisprudence, 15 SANTA CLARA COMPUTER & HIGH TECH. L.J. 231 (1999). 70 236 F.3d 1363 (Fed. Cir. 2001). See O.I. Corp. v. Tekmar Co., 115 F.3d 1576 (Fed. Cir. 1997). 172 F.3d 836 (Fed. Cir. 1999). 870 F. Supp. 753 (E.D. Mich. 1994). 98 F.3d 1318 (Fed. Cir. 1996). 172 F.3d at 850. 71 72 73 74 75 76 State Street Bank & Trust Co. v. Signature Financial Group, Inc., 149 F.3d 1368 (Fed. Cir. 1998). 77 In re Lee, 277 F.3d 1338 (Fed. Cir. 2002). 61