The degree to which prosecution history estoppel limits the doctrine of equivalents has long been the subject of much debate among patent owners, competitors, and the courts. In Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki Co.,1 the Supreme Court attempted to provide guidance on the issue by establishing that whenever patent claims are narrowed by amendment, a rebuttable presumption arises that prosecution history estoppel bars all equivalents.2 However, the Court's decision may have done more harm than good by introducing a foreseeability element into the requirements for rebutting the presumption.
The state of the law prior to Festo was unclear at best. In 1997, the Supreme Court held that prosecution history estoppel was a flexible, not an absolute, bar to the doctrine of equivalents in patent infringement cases, meaning that some, but not all, claims of equivalence would be barred.3 In 2000, the Federal Circuit departed from the flexible-bar rule by holding that whenever a patentee amends claims in order to obtain a patent, prosecution history estoppel acts as an absolute bar to the doctrine of equivalents with respect to the amended elements.4 The Federal Circuit justified its harsh, absolute-bar rule on the grounds that the flexible-bar approach had proven "unworkable" and that an absolute bar would
provide desired certainty in determining the scope of an amended patent claim.5
The Federal Circuit's decision was the source of great concern to thousands of patentees because it significantly limited the range of subject matter against which a patentee could successfully assert a claim of infringement by equivalence.6 Patent claims narrowed by amendment during patent prosecution-and many claims are-- would be limited to their literal language for all amended elements, thereby depriving patentees of protection against competitors' devices falling just outside the literal language. Patentees who amended their claims with a flexible-bar rule in mind were understandably concerned that the Federal Circuit's absolute bar devalued their already-issued patents by effectively granting a much narrower monopoly.
Concerned patent holders were relieved when, on May 28, 2002, the Supreme Court vacated the Federal Circuit's decision and reinstituted a flexible-bar rule.7 In re-instituting a flexible-bar rule, the Supreme Court held that a presumption exists that prosecution history estoppel bars all equivalents but that the patent owner may rebut the presumption under certain circumstances.8 The difference, therefore, between the Supreme Court's flexible bar and the Federal Circuit's absolute bar lies in the fact that the presumption is rebuttable. The problem with the Supreme Court's decision is that it sets forth contradictory standards for how rebuttal is to be accomplished and it creates incentives that undercut the goals of the patent system.
This Note will discuss the tension between the doctrine of equivalents and prosecution history estoppel and how the Supreme Court attempted to resolve that tension in its Festo decision. Part II will briefly explain several basic patent law principles and outline the state of the law in 1997 as set forth in Warner-Jenkinson Co. v. Hilton Davis Chemical Co.,9 against which the Federal Circuit's
decision and its subsequent vacation by the Supreme Court were set. Part III will follow with a description of the facts of the Festo case and a brief synopsis of the holdings of the Federal Circuit and the Supreme Court. Part IV will analyze the Supreme Court's Festo decision vacating the Federal Circuit and assert that while the Court correctly reaffirmed estoppel as a flexible, not an absolute, bar to the doctrine of equivalents, the Court undercut some of the most basic policies of the patent system by introducing foreseeability as a component of the flexible bar. Part V will offer a brief conclusion.
A. The Dual Purposes of Patent Claims
The question of how far a patent's monopoly extends has always been a difficult one. In order to obtain a patent on an invention, a patentee is required to particularly point out and distinctly claim the subject matter he regards as his invention.10 Claims serve at least two important functions: (1) claims offer protection to the patent holder by delineating the bounds of the property right granted, thereby defining the subject matter the patentee can legally exclude competitors from making and selling;11 and (2) claims provide notice to the public of what property the patent owner does not own, thereby enabling the public to know what subject matter is freely available.12 There is tension between the protection and notice functions of claims. These two functions can be represented by a spectrum, or sliding-scale, one end of which represents the protection function and the other end the notice function. Every point along the spectrum represents a possible interpretation of the language in a patent claim. Because the spectrum is a sliding scale, it is impossible to maximize both protection and notice at the same time, and either extreme is probably undesirable.
At the protection end of the spectrum, claims are interpreted broadly to include more than what the literal language encompasses.
As a result, greater protection is afforded a patent holder because he is not limited in his assertions of infringement to those competitors who market devices literally encompassed in the language of the claims. The obvious disadvantage at the protection end of the spectrum is the fact that the public cannot tell what infringes a patent and what does not. The result is a stifling of the economy: competitors decline to market legitimate devices for fear of being sued for patent infringement.
At the notice end of the spectrum, claims are interpreted literally. This enables claims to provide maximum notice to the public of what subject matter is covered by the patent and what subject matter is freely available to the public.13 A patent holder can easily ascertain exactly how much protection his patent affords him and, consequently, his patent's value. His competitors can know with certainty whether their own devices constitute allowable alternatives to the patented invention or whether they infringe. Such certainty for patentees and the public is desirable, but, as the Supreme Court has noted, it comes at a price: "If patents were always interpreted by their literal terms, their value would be greatly diminished."14 For example, if a patent claim recited "four legs" and "a round seat," a competitor's product having four legs and an oval seat would not literally infringe the patent. Literal infringement is generally easy for a competitor to avoid because merely insubstantial differences between the accused product and the claimed invention are sufficient to avoid literal infringement. Such instances can be particularly frustrating to patent owners.
B. The Doctrine of Equivalents
In response to situations in which competitors attempt to avoid literal infringement by making insubstantial changes, courts have created the doctrine of equivalents. The doctrine of equivalents protects a patent holder "against efforts of copyists to evade liability for infringement by making only insubstantial changes to a patented invention."15 Thus, "[t]he scope of a patent is not limited to its
literal terms but instead embraces all equivalents to the claims described."16
Under the doctrine of equivalents, even if a competitor's product does not literally infringe, it can still infringe as an equivalent if it differs from the claimed invention in merely insubstantial ways. For example, suppose again that a patent claim recited "four legs" and "a round seat." A competitor's product having four legs and an oval seat would not literally infringe the patent. However, it could still infringe under the doctrine of equivalents if a court held that the oval seat was equivalent to a round seat." The doctrine of equivalents allows a patent holder to exclude competitors not only from the subject matter literally defined by the claims but also from subject matter constituting an equivalent to the literal subject matter.
The difficulty with the doctrine of equivalents is knowing where to draw the line between what constitutes an infringing equivalent and what does not. The Supreme Court has noted:
[B]y extending protection beyond the literal terms in a patent the doctrine of equivalents can create substantial uncertainty about where the patent monopoly ends. If the range of equivalents is unclear, competitors may be unable to determine what is a permitted alternative to a patented invention and what is an infringing equivalent.18
To illustrate, if a patent claim recites "a round seat," competitors can be certain that if their own devices have round seats, literal infringement will be found. But there may be some uncertainty as to whether an oval seat constitutes an infringing equivalent. There may
be even more uncertainty as to whether a square, triangular, or other shaped seat constitutes an infringing equivalent.19
The doctrine of equivalents provides greater protection to patent holders than does the literal language of the claims. But the increased protection afforded by the doctrine of equivalents comes at the price of a corresponding increase in the uncertainty of the scope of patent claims. As a result, claims provide less notice of what subject matter is freely available to the public.
C. The Doctrine of Prosecution History Estoppel
The doctrine of prosecution history estoppel helps reduce the uncertainty associated with the doctrine of equivalents in certain situations. "Competitors may rely on [prosecution history] estoppel to ensure that their own devices will not be found to infringe by equivalence."20 Prosecution history estoppel dictates that when a patentee narrows a claim in order to obtain a patent, the subject matter between the broad claim and the narrower claim is surrendered and cannot later constitute an equivalent.21 Therefore, a competitor's product falling within the scope of the original claim but outside the scope of the narrowed claim will not infringe, either literally or by equivalence. For example, suppose that a patent claim originally recited "four legs" and "a seat" and that the claim was rejected by the Patent and Trademark Office as being overly broad. If the patentee then amends his claim to recite "four legs" and "a round seat" in order to obtain the patent, prosecution history estoppel will apply. In that case, the difference in subject matter between the broad, "seat" claim and the narrower, "round seat" claim is surrendered and cannot constitute an equivalent. This is significant because if the patentee had originally claimed "a round seat," he could possibly assert a claim of infringement by equivalence against a competitor's device having a square seat. But because he narrowed his claim by amendment from "a seat" to "a round seat," prosecution history estoppel says he cannot assert that a square seat is equivalent to a round seat. Subject matter encompassed by the broad claim and excluded by the narrow claim (like square seats) cannot
constitute an equivalent. Therefore, examining a patent's prosecution history is important because whether an accused product infringes by equivalence may depend on whether the patent was amended.
D. The State of the Law Prior to the Federal Circuit's Decision in 2000, as Set Forth by the Supreme Court in Warner-Jenkinson22
Courts have struggled with at least two difficult questions regarding prosecution history estoppel and its effect on the doctrine of equivalents.23 First, should a claim amended for any reason give rise to prosecution history estoppel?24 Second, when an amendment does give rise to prosecution history estoppel, are all equivalents barred or only some of them?
In 1997, the Warner-jenkinson Court attempted to provide answers to these questions. The Court held that not all claims narrowed by amendment necessarily give rise to prosecution history estoppel: "Where the reason for the change was not related to avoiding the prior art, the change may introduce a new element, but it does not necessarily preclude infringement by equivalents of that element."25 Therefore, "certain reasons for a claim amendment may avoid the application of prosecution history estoppel."26 The Court noted, however, that its holding did not mean that "the absence of a reason for an amendment may similarly avoid such an estoppel."27
The Warner-Jenkinson Court further instructed that when amendments are made, there is a presumption that prosecution
history estoppel bars application of the doctrine of equivalents but that the patentee has the opportunity to overcome the presumption:
[W]e think the better rule is to place the burden on the patent holder to establish the reason for -an amendment required during patent prosecution. The court then would 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. Where no explanation is established, however, the court should presume that the patent applicant had a substantial reason related to patentability for including the limiting element added by amendment. In those circumstances, prosecution history estoppel would bar the application of the doctrine of equivalents as to that element.28
Under this regime, if a patentee amended her claims during prosecution, she would have the opportunity to explain her reasons for amending. The court would then decide whether the reasons were sufficient to avoid application of prosecution history estoppel. In the absence of an explanation, the court would presume the amendment was made for patentability reasons (i.e., made in order to obtain the patent) and all subject matter surrendered by the amendment would be barred from later constituting an infringing equivalent. The Warner-Jenkinson Court characterized this rule as a "presumption ... subject to rebuttal."29 However, the Court provided little guidance as to what sort of reasons for amending were "sufficient to overcome" the presumption.
Therefore, the law as set forth by Warner-Jenkinson may be summarized by stating that whenever a patentee amends claims during prosecution, a rebuttable presumption arises that prosecution history estoppel bars application of the doctrine of equivalents as to the amended elements. The patentee may rebut the presumption, thereby preserving her right to the doctrine of equivalents, by establishing "an appropriate reason for [amending]."30 However, because the Court failed to define what constitutes an "appropriate reason" for amending, Warner-Jenkinson left unclear what the standard for rebuttal is. Establishing the standard for rebuttal is
precisely what the Supreme Court attempted in its late Festo decision.
III. FESTO CORP. V. SHOKETSU KINZOKU KOGYO KABUSHIKI CO.
A. The Facts and the Federal Circuit's Holding
The Festo Corporation owned two patents relating to an improved piston-cylinder assembly, both of which were amended during prosecution.31 The application for the first patent was initially rejected because "the exact method of operation was unclear and some claims were made in an impermissible way."32 The second patent was amended during a reexamination proceeding.33 Both patents were amended to include a new limitation: two one-way sealing rings.34 The first of the two patents was additionally amended to include a second new limitation: that the outer shell of the device be magnetizable.35 After Festo's patents issued, Shoketsu ("SMC") began selling a device similar to the ones disclosed in Festo's patents, but with two notable differences: (1) SMC's device employed a single two-way sealing ring instead of two one-way sealing rings; and (2) SMC's device had a shell made of a nonmagnetizable material.36
Festo sued SMC for patent infringement. At trial, Festo argued that while SMC's device did not literally infringe Festo's patent, it infringed by equivalence. SMC argued that Festo narrowed its claims by amendment to exclude the very subject matter it now sought to claim as an equivalent.37 In other words, Festo's claims as originally drafted were broad enough to cover the elements of SMC's device, but were subsequently narrowed by amendment to claim two oneway sealing rings and a magnetizable sleeve. The amended claim's language therefore excluded a single two-way sealing ring and a nonmagnetizable sleeve from the claimed subject matter. SMC argued that prosecution history estoppel barred Festo from trying to
recapture as an equivalent the very subject matter it had surrendered by amendment.38
The Federal Circuit agreed, holding that Festo was barred by prosecution history estoppel from asserting that SMC's device infringed under the doctrine of equivalents.39 Several elements of the Federal Circuit's holding were especially controversial. First, the court held that amending a patent claim for any reason related to patentability would give rise to prosecution history estoppel.40 Second, and more controversial still, the court held that when prosecution history estoppel applies, it acts as an absolute bar to all alleged equivalents of the amended claim, effectively meaning that once a claim is amended, there can be no such thing as an equivalent to the amended elements of that claim.41 The Federal Circuit justified its departure from the flexible bar established in Warner-- Jenkinson on the grounds that the flexible bar had proven "unworkable" and that an absolute bar was needed to provide certainty to the scope of an amended patent claim.42
B. The Supreme Court's Holding
The Supreme Court granted certiorari to "address once again the relation between two patent law concepts, the doctrine of equivalents and the rule of prosecution history estoppel."43 The Court's opinion may be subdivided into three important sections. First, the Court stated its approval of both the doctrine of equivalents and the rule of prosecution history estoppel. Second, the Court addressed the question of whether all types of amendments give rise to prosecution history estoppel. Third, the Court addressed the question of whether prosecution history estoppel, once invoked, bars all equivalents to an amended claim.
1. The doctrine of equivalents and the rule of prosecution history estoppel are both good law
The Court began by noting its general approval of both the doctrine of equivalents and the rule of prosecution history estoppel, stating that "[flf patents were always interpreted by their literal terms, their value would be greatly diminished.... The scope of a patent is not limited to its literal terms but instead embraces all equivalents to the claims described."44 Further,
"[p]rosecution history estoppel ... precludes a patentee from regaining, through litigation, coverage of subject matter relinquished during prosecution of the application for the patent." Were it otherwise, the inventor might avoid the [Patent and Trademark Office's] gatekeeping role and seek to recapture in an infringement action the very subject matter surrendered as a condition of receiving the patent.45
Therefore, according to the Supreme Court, both "the doctrine of equivalents and the rule of prosecution history estoppel are settled law."46 However, there is tension between the two doctrines: the doctrine of equivalents broadens the scope of protection afforded patent holders beyond that provided by the literal claim language, while the doctrine of prosecution history estoppel acts to contract the scope of protection back towards the literal claim language. The difficult questions the Court was required to address in Festo involved the interplay between the two competing doctrines-- specifically, when and how they affect each other.
The Supreme Court's Festo decision was more instructive than the Court's prior decision in Warner-Jenkinson because in Festo the Court cleanly separated the amorphous issues surrounding the relationship between estoppel and the doctrine of equivalents into two questions: When does claim amendment give rise to prosecution history estoppel? And, when prosecution history estoppel does arise, does it bar all equivalents to the amended element, or only some of them? The Festo Court addressed each of the two questions in turn.
2. When prosecution history estoppel arises
The Court held simply that "[e]stoppel arises when an amendment is made to secure the patent and the amendment narrows the patent's scope."47 The Court stated that the inquiry should not focus on the applicant's purpose for amending, but should focus rather on the impact the amendment has on the subject matter-i.e., whether or not the subject matter is in fact narrowed.48 Amendments that do not narrow the scope of the claimed subject matter do not give rise to prosecution history estoppel 49 Therefore, the threshold question for determining whether prosecution history estoppel applies is, "does the amendment narrow the claim?" not "what was the purpose of the amendment?"
3. Prosecution history estoppel constitutes a flexible bar to alleged equivalents
The Supreme Court rejected the Federal Circuit's holding that when prosecution history estoppel applies, it stands as an absolute bar to all equivalents. Instead, the Court held that when prosecution history estoppel applies, a presumption is raised: when a patentee narrows a claim by amendment, it is presumed that "the patentee surrendered all subject matter between the broader and the narrower language."50 However, the patentee may rebut the presumption.51 This is the critical difference between the Supreme Court's holding and that of the Federal Circuit. The opportunity to rebut is the reason estoppel constitutes a flexible, not an absolute, bar to the doctrine of equivalents.
The critical inquiry focuses on what must be done to rebut the presumption that prosecution history estoppel bars all equivalents to the amended claim. In other words, how does a patent holder rebut the presumption? Practically speaking, the answer to this question is the most important part of the Supreme Court's Festo decision, as well as that which distinguishes Festo from Warner-Jenkinson.
The Court stated that to rebut the presumption that estoppel bars all equivalents, the patentee "must show that at the time of the amendment one skilled in the art could not reasonably be expected to have drafted a claim that would have literally encompassed the alleged equivalent."52 Had the Court stopped there, it might have avoided considerable confusion. But the Court went on to state that rebuttal could be accomplished by showing that "[t]he equivalent [was] unforeseeable at the time of the application."53 It appears that the test the Court is aiming at with this language is one of foreseeability: a patentee who amends his claims can rebut the presumption that estoppel bars all equivalents by showing that the alleged equivalent was unforeseeable. However, the Court was unclear about whether foreseeability is to be determined "at the time of the amendment," or "at the time of the application." This is significant because several years can pass between the time of application and the time of amendment, and with rapidly-changing technology, that which is completely unforeseeable today may be easily foreseen tomorrow. Moreover, if foreseeability really is the test, additional complications arise because a foreseeability test creates incentives running counter to the main goals of the patent system. These incentives will be discussed in further detail in Part IV of this Note.
In summary, the rules of law handed down by the Supreme Court in its 2002 Festo decision may be stated thus: whenever a patent claim is narrowed by amendment to secure the patent, prosecution history estoppel applies, regardless of why the claim was amended.54 When prosecution history estoppel applies, it is presumed to bar all equivalents to the amended claim.55 However, the patentee can rebut the presumption by showing that the alleged equivalent was unforeseeable to one skilled in the art.56 It is unclear at which point in time foreseeability is to be determined.
IV. ANALYSIS: WHY THE SUPREME COURT'S FLEXIBLE BAR WAS GOOD POLICY UNTIL THE INTRODUCTION OF A FORESEEABILITY ELEMENT
The analysis section of this Note consists of three parts. Part A discusses an initial policy reason supporting the Supreme Court's flexible-bar rule: thousands of patents currently in force were drafted and issued with a flexible bar in mind, and instituting an absolute bar now would upset the settled expectations of the inventing community. Part B discusses the shortcomings of language as a second independent policy reason supporting estoppel as a flexible, not an absolute, bar to the doctrine of equivalents. Part C agrees with the Court's holding that estoppel ought to act as a rebuttable presumption barring equivalents but objects to the Court's introduction of foreseeability into the standard for rebuttal.
A. Changing the Rules Affecting the Scope of Patents Now Would Upset the Settled Expectations of the Inventing Community
The Supreme Court has warned that courts must use caution when they adopt "changes that disrupt the settled expectations of the inventing community."57 This caution is especially necessary in dealing with prosecution history estoppel and the doctrine of equivalents because both "are settled law."58 Changing the rules that affect the scope of protection afforded by patents risks "destroying the legitimate expectations of inventors in their property."59
Patents are prosecuted with the relevant case law in mind.60 Both applicants and examiners rely on case law throughout the prosecution process. Applicants choose their claim language in light of case law and examiners issue notices of acceptance and rejection with the case law in mind. For example, suppose that a patent claim recites "four legs" and "a seat" and that the examiner initially rejects the claim. At this point the applicant must decide whether to amend his claim or leave it as it is and appeal the rejection. Further suppose the applicant knows that the Supreme Court has recently held that prosecution history estoppel acts as a flexible, not absolute, bar to
equivalents. Knowing of the flexible bar, the applicant decides to amend his claim to recite "four legs" and "a round wooden seat," and the patent issues. Soon after, a competitor begins selling devices having four legs and round seats comprised of a previously-unknown fiberglass material. In this situation, the patent owner "had no reason to believe [he was] conceding all equivalents" when he amended his claim, because of the controlling case law.61 As a result, the patent owner, thinking fiberglass equivalent to wood, brings suit against the competitor for infringement by equivalence. Now suppose that while the patent owner is in the initial stages of discovery, the Supreme Court adopts an absolute bar such that once a claim is amended, estoppel bars all claims of equivalence. The patent owner now would have no remedy against his competitor. He cannot now appeal the initial rejection occurring during prosecution, nor can he successfully assert infringement. This is just the sort of situation the Festo Court had in mind when it said that if the inventor had known the flexible bar would be replaced by an absolute bar, he "might have appealed the rejection instead. There is no justification for applying a new and more robust estoppel to those who relied on prior doctrine."62
Adopting an absolute bar and applying it to patents currently in force would "disrupt the expectations of countless existing patent holders."63 However, there is another option that the Supreme Court did not expressly consider in its Festo decision. The Court could have adopted an absolute bar which would only apply to patents yet to be prosecuted. This would eliminate the unfairness and the disruption of settled expectations the Court sought to avoid. At the same time, such a holding would also have all the benefits of certainty and judicial economy that are normally associated with a "bright-line rule."64 However, an absolute bar, even if applied prospectively only, should still be avoided because it cuts against the policy upon which the doctrine of equivalents is based in the first
place: that language is a poor fit for invention and can never perfectly capture the essence of an invention.65 If the doctrine of equivalents is still good law-and the Supreme Court has reaffirmed time and again that it is66 its very existence seems to preclude a rule that makes prosecution history estoppel an absolute bar to equivalents.
B. The Policy Behind the Doctrine of Equivalents Supports Application to Amended Claims as Well as to Original Claims
The Federal Circuit's Festo decision held that "prosecution history estoppel acts as a complete bar to the application of the doctrine of equivalents when an amendment has narrowed the scope of a claim."67 According to the Federal Circuit, an original claim is entitled to some range of equivalents, but once amended, the claim must be interpreted to include only subject matter falling within the literal language of the claim. The Federal Circuit's holding results in an interesting paradox: prior to amendment, the court will recognize that an inventor cannot perfectly describe his invention with language, but if he amends his claims, he is then treated as if he were able to perfectly describe his invention. The Supreme Court's Festo decision recognized this inconsistency when it stated:
By amending the application, the inventor is deemed to concede that the patent does not extend as far as the original claim. It does not follow, however, that the amended claim becomes so perfect in its description that no one could devise an equivalent. After amendment, as before, language remains an imperfect fit for invention. The narrowing amendment may demonstrate what the claim is not; but it may still fail to capture precisely what the claim is.... As a result, there is no more reason for holding the patentee to the literal terms of an amended claim than there is for abolishing
the doctrine of equivalents altogether and holding every patentee to the literal terms of the patent.68
Whether or not the doctrine of equivalents should be abolished altogether is another interesting question, but the Supreme Court "has acknowledged [the uncertainty the doctrine of equivalents produces] as the price of ensuring the appropriate incentives for innovation, and it has affirmed the doctrine [of equivalents] over dissents that urged a more certain rule."69 The Court has further stated that "if the doctrine is to be discarded, it is Congress and not the Court that should do so. . . . 'Congress can legislate the doctrine of equivalents out of existence any time it chooses.'"70
The Supreme Court's argument against the absolute bar imposed by the Federal equivalent is mainly one of consistency. Legislating the doctrine of equivalents out of existence and holding inventors to the literal language of their claims, whether or not amended, may be bad doctrine of equivalents out of existence and holding inventors to the literal language of their claims, whether or not amended, may be bad policy, but at least it is internally consistent policy. So is allowing application of the doctrine of equivalents to original and amended claims (because in both cases inventors cannot perfectly describe their inventions using language). However, it is inconsistent to hold, as the Federal Circuit did, that original claims are entitled to the doctrine of equivalents while amended claims are not. In other words, original and amended claims should both be entitled to equivalents, or else neither should. The doctrine of equivalents exists in the first place because language is an imperfect means of describing an invention, and that problem exists wheter or not a claim is amended. Therefore, the same justification for the existence of the doctrine of equivalents in the first place also justifies its application to amended claims.
Because of language's imperfection, the Supreme Court had little difficulty in reaching its decision that estoppel should constitute a flexible, not absolute, bar to equivalents.71 Similarly, the Court easily reached its holding that claim amendment creates a rebuttable presumption that estoppel bars all equivalents.72 However, the real difficulties arose when the Court established that the standard for rebuttal hinges on foreseeability.
C. The Supreme Court's Standard of Foreseeability Is Ambiguous and Creates Incentives that Run Counter to the
Goals of the Patent System
The Festo Court held that in order for a patent holder to rebut the presumption that estoppel bars all equivalents, she "must show that at the time of the amendment one skilled in the art could not reasonably be expected to have drafted a claim that would have literally encompassed the alleged equivalent."73 The Court also stated that a patent holder can rebut the presumption by showing the alleged equivalent was "unforeseeable at the time of the application."74 As stated previously in this Note, the Court was unclear about whether foreseeability is to be determined at the time of amendment or at the time of the application. This ambiguity is problematic because several years can pass between the "time of the application" and the "time of the amendment," and during this time the scope of the subject matter considered "foreseeable" may change greatly. This ambiguity will almost certainly require the Supreme Court to address the issue again to clarify the point in time at which foreseeability is to be determined.
Additionally, because foreseeability is a mixed question of law and fact,75 basing the standard of rebuttal upon foreseeability would raise new factual issues, making infringement cases less likely to be resolved on summary judgment." Courts would often need to resort
to expert testimony in order to determine what was foreseeable at a given point in time.77 As a result, introducing the concept of foreseeability seems to complicate, rather than simplify, an "already ... amorphous and vague area of the law."78
Perhaps the most serious problem with introducing foreseeability into the rebuttal analysis lies in the resulting incentives that are created. Judge Lourie of the Federal Circuit notes that "the concept of foreseeability seems akin to obviousness,"79 and he advances a persuasive argument based on that assumption of similarity. An example serves to illustrate: Assume that foreseeability and obviousness are similar80 and that for an amended claim, foreseeability precludes equivalence consistent with the Festo Court's holding. Suppose again that a patent claim recites "four legs" and "a seat" originally, and is initially rejected. Suppose the applicant then amends his claim to recite "four legs" and "a fiberglass seat," and that he is unworried because fiberglass happens to be the only known material suited to the device's purpose anyway. Further suppose that his patent issues and that some time later a competitor
discovers that a new carbon-fiber material is a perfect substitute for fiberglass in any application. The competitor begins marketing a device exactly like the patented device, except that carbon-fiber is substituted for fiberglass. The patentee now desires to sue his competitor for infringement by equivalence.
The facts in the above example serve to illustrate the awkward positions from which the respective parties would be forced to argue in light of the Supreme Court's Festo holding. As previously discussed, since the amendment narrowed the claim (regardless of why the amendment was made), prosecution history estoppel applies and raises a rebuttable presumption that there are no equivalents to "a fiberglass seat" (the amended claim element). The patent holder must rebut the presumption by arguing that the carbon-fiber seat was not foreseeable, i.e., not obvious. But if the carbon-fiber seat was not foreseeable or obvious, it should be patentable.81 Therefore, the patent holder is forced to make the awkward argument that while his competitor's device represents a sufficient level of innovation to merit its own patent (because it was not obvious), it nevertheless infringes the patent holder's patent by equivalence (because it was not foreseeable).82 It is axiomatic that a device meriting its own patent does not infringe any other patents.
The competitor's argument under the same facts would be no less awkward. In order to escape liability, the competitor must argue that his carbon-fiber seat was foreseeable and therefore not an equivalent. But if the carbon-fiber seat was foreseeable, then it was obvious, and the competitor is forced to argue that his device does not infringe by equivalence because it represents such a low level of innovation as to be foreseeable and obvious.83 A party accused of infringement generally wishes to show that his device represents a substantial innovation, not that it was obvious.
In addition to forcing parties to make counterintuitive arguments, basing the standard for rebuttal on foreseeability also creates undesirable incentives. One of the goals of the patent system is to encourage inventors to publicly disclose their innovations via the filing of patents.84 The Festo Court's foreseeability standard
actually discourages the filing of patents in some instances. For example, suppose an inventor wishes to market a new device and that the device is legitimate but close to the amended claims of an existing patent. If the inventor seeks a patent, he must show that his device is non-obvious (i.e., not foreseeable) in order to obtain the patent.85 But if for some reason the patent is not granted, the inventor will be left in a difficult position: if the inventor is sued for infringement by equivalence, he will have to reverse direction and argue exactly the opposite of what he argued before the Patent Office: he will now need to argue that his invention was obvious and foreseeable and that it therefore cannot constitute an equivalent. In such an instance, the inventor actually has an incentive at the outset to forego filing a patent application in order to preserve his ability to effectively argue obviousness and foreseeability in infringement actions.86 Giving the inventors of legitimate devices an incentive to avoid filing a patent runs directly counter to the goal of the patent system of encouraging disclosure through the filing of patents.
In summary, the Court erred by setting forth contradictory times at which foreseeability is determined, and it erred at a more critical level by adding an element of foreseeability to the rebuttable presumption. Doing so forces patentees and competitors to make arguments that are counterintuitive to the patent system and also creates inappropriate incentives.
The Supreme Court essentially took one step forward and two steps back in its Pesto decision. The Court stepped forward by holding that when a claim is narrowed by amendment, a rebuttable presumption arises that prosecution history estoppel bars all assertions of equivalence. The rebuttability of the presumption makes estoppel a flexible, not an absolute, bar. But the Court took two steps backward by holding that successful rebuttal requires a showing that the alleged equivalents would have been unforeseeable to one skilled in the art. Not only did the Court confuse the rebuttal analysis by contradicting itself as to the time at which foreseeability is to be determined, but the Court also created incentives that run
counter to the very incentives the patent system exists to provide in the first place.
Because of the confusion and undesirable incentives created by its Festo decision, the Supreme Court will need to address the relationship between the doctrine of equivalents and prosecution history estoppel again in the future. It is possible that the Court will merely clarify the point in time at which foreseeability is to be determined. Such a decision would certainly represent an improvement from where the Court left things in Festo, but the better approach would be to revise the flexible bar to exclude the element of foreseeability altogether. Introducing an element of foreseeability, or any other element that creates incentives which undercut the goals of the patent system, is not the answer.…