CHANGING THE RULES IN THE MIDDLE OF THE GAME: HOW THE PROSPECTIVE APPLICATION OF JUDICIAL...

IMAGE TABLE 1IMAGE TABLE 2

I. INTRODUCTION

A complete understanding of the social and economic environment in the United States must consider

the role of the legal system. It provides the basis for the other elements-the boundaries of the playing field and the rules for the game. To a great extent, our laws express a consensus of our base values and vision of what society should be.1 In this regard, our legal system can be understood to have a substantive role. Additionally, the legal system facilitates the journey to that desired state by providing the enforcement mechanism. In that manner, it has a structural function. If the legislature appears to take on the more substantive aspects by Grafting and revising the rules and playing field with societal goals in mind, the courts are usually viewed as a more structural player: they provide a forum for settling disputes rather than planning society.

However, we have gradually begun to appreciate the substantive role courts can play, particularly in light of the ebb and flow of judicial activism in the later half of the twentieth century.2 While court incursions into the substantive realm may be explicit (typically more controversial3), they can also be subtle. In fact, one of the most important ways that courts substantively impact the social and economic environment is also one of the least discussed: the retroactive revision of the law by applying changes in legal rules to the actions of those who relied on prior rules. By doing so, courts can create a new reality for a wide range of interests, far different from that originally anticipated. And in perhaps no sense is this more significant than when the interests affected are in the nature of personal or real property. This is particularly so due to, not only the economic impact of changes to property value, but also the emotional impact on Western society, which tends to view the sanctity of private property rights as a bedrock principle.4

Thus, when courts render decisions that call into question the enforceability (and, thus, the value) of property rights by retroactively reassessing the propriety of actions undertaken in the past, they are exerting a substantive force that must be considered for the sake of economic efficiency and societal contentment.5 The problem is particularly significant in the field of intellectual property6 due to the magnitude of rule changing that is likely to occur. Because the value of intellectual property rights has become increasingly important to modern businesses-in some cases wholly accountable for the success or failure of a company7-retroactive devaluation of those rights is of high concern. Moreover, if an excess of inefficiency and waste unexpectedly results from objectively rational investment decisions-especially those consuming such a great proportion of society's resources-there is an economic motive to rethink the status quo.

Yet, despite the importance of this issue and its potential to impact the U.S. (and world) economy, very little academic discussion has appeared in the literature of intellectual property or business law that addresses the topic head-on.8 Most of the commentary and case law have been concerned with the retroactive effect of new legal rules in the criminal law context.9 Significantly, economic analysis of these issues, particularly useful in the intellectual property law context,10 is generally lacking. This is a curious fact considering the amount of literature devoted to the economics of the U.S. intellectual property system in general.11

This paper will first lay groundwork for such an analysis by discussing the basic economics of various forms of intellectual property and how retroactive rule changing can affect those economic systems, paying particular attention to patents as a primary example of the phenomenon. Next, the paper will review the current state of the law and how the courts have addressed the issue of retroactive application of judicial determinations. Following that, this paper will recommend alternatives for the way that courts handle such decisions, advocating the incorporation of a modified test for determining when prospective application of new rules is necessary and appropriate. Finally, the paper will analyze three recent, significant cases in the area of patent law and demonstrate how the author's proposed method of analysis would support or diverge from the actual decisions of the courts.

II. THE ECONOMICS OF INTELLECTUAL PROPERTY PROTECTION: DIVERSE GROUNDS WITH ULTIMATE COMMONALITY

Long before the emergence of modern industrial economies, there was great interest in the economic impact and value of intellectual property rights. Scholars, political leaders and business persons alike have understood the utility in allowing individuals or companies to hold property rights in creative output.12 However, the debate over the proper type and degree of protection for a given intellectual creation is ongoing, and calls for change or revision are constant. Out of the debate has come some substantial and useful analysis of the economic rationale for granting intellectual property rights. This perspective is a good starting point for further discussion, especially considering that the inherent utility of modern intellectual property laws in the United States is often regarded from the standpoint of their effect on the economic environment.13 Moreover, the effect a certain act or condition has on economic efficiency is at least one objective measure of whether that act or condition is "good" or "bad" for society.14

A. The "Contract" Between Intellectual Property Owners and Society

There is wide support in the field of law and economics for the proposition that private property ownership tends to encourage efficient use of scarce societal resources.15 Without private ownership, waste or under-use of resources is likely to occur.16 The same is true regarding private ownership rules in the allocation of rights related to intellectual property.17 However, the intangible nature of intellectual property makes more complicated the specific way in which a private property right promotes efficiency.18

Despite the fact that intellectual properties of all types are classified as a legal form of personal property,19 they differ from tangible forms of both real and personal property in that intellectual property is not in and of itself "scarce"; one party's use does not reduce the amount of intellectual property remaining for others to use.20 Additionally, because intellectual property is essentially information, it is extremely easy to transmit, therefore it is difficult to prevent others from accessing or using it without permission.21 It has been proposed that private property rights in information exist primarily to enable the optimum production level of those rights, essentially the same as with tangible personal property.22 Without private rights, intellectual property will tend to be under-produced because of the difficulty in obtaining a reasonable return on the investment in developing the property. However, it has also been argued that the true purpose in protecting intellectual property rights is to aid in the allocation of related resources that are in fact scarce-i.e., those that are used in the development ofthose properties.23 Under this theory, those underlying resources will be underutilized or misallocated unless they can be used to produce privately-owned intellectual property rights.24

Regardless of the general theory to which one subscribes, the interesting questions arise from consideration of the specific ways in which these different types of properties are intended to function in the economic and social environment. The scheme actually differs somewhat for each of the major categories of intellectual property rights.25 It is useful to compare and contrast the categories to gain a better understanding of the factors that can influence the current environment.

Under U.S. law, patent rights, often perceived to be the strongest form of intellectual property right,26 are derived from a specific provision in the U.S. Constitution that empowers Congress to grant inventors exclusive rights to their inventions for a limited time.27 The ostensive purpose for granting patents is written into the Constitution: "To promote the Progress . . . of useful Arts."28 Most have interpreted this to mean the promotion of innovation.29 Thus, the Constitution takes as a given that more innovation is better for society and that the patent grant is one method available to Congress as a means of specifically promoting innovation.30

Certainly, from an economic perspective, it can reasonably be argued that more innovation is better, at least to a point. However, it can also be argued that it is possible to have too much innovation-in other words, a level of innovation that is suboptimal in its use of scarce societal resources as compared to the gain obtained by society.31 But how much is too much? This question may be inherently impossible to answer in any meaningful way (i.e., without resorting to vague pronouncements and circular reasoning). For purposes of furthering the discussion, this paper will join the vast majority of the literature in assuming that innovation is beneficial to society.32

If one accepts the benefits of innovation, the relevant economic question that follows relates to how strong33 the property right should be in order to foster innovation34 without unduly restricting competition by the grant of overly-broad exclusive rights. Theoretically, if patent rights are too strong, a patent could actually dissuade others from innovating or engaging in other business activities.35 Analysis of this "tension" consumes the bulk of the economic literature related to patents.36 Although courts tend to view patent rights as a "reward," appropriate if commensurate in scope to a particular level of invention,37 some modern commentators advocate viewing patents as analogous to a prospector's mineral claim-the so-called "prospect theory of patents"-which suggests that considerations of strength should take into account the fact that the patentee is additionally laying claim to future development of the invention.38 Whatever the theory, the U.S. system can be viewed as creating beneficial output from two perspectives: (1) on an individual level, some return on the investment in innovative research and development and (2) on a societal level, industrial spending that produces innovation that eventually becomes available for public use, and is immediately available for private exploitation by the patent holder.

U.S. copyrights have a similar constitutional and economic basis to U.S. patents.39 Again, promoting "progress" of the social sciences such as arts and literature is the impetus for granting copyrights.40 However, the copyright is a more narrow intellectual property right than a patent, and the basis for offering the specific legal protection is somewhat different.41

The most important distinction between copyrights and patents is that the copyright is an attempt to protect a specific expression of an idea made by an author, rather than a broad concept.42 It is meant to protect the work as it exists when finished, not some future development. Because the right is narrower-in contrast to patents, independent creation does not infringe43-there is less likelihood that a copyright will restrain competition.44 Thus, copyright protection is conferred without substantive examination45 and the term is quite long.46 Despite its more limited nature, the economic rationale for conveying exclusive property rights for creative expression is essentially the same as for patents.47 Due to the low cost of copying and using information assets such as copyrighted material, protection (with the possibility of a return) is necessary if one is to invest time and resources in developing them.48 Secondarily, society benefits if the system produces more creative expression than would be produced if no protection existed.

Trademark and trade secret protection have different philosophical bases than patents and copyrights, but economic return as an incentive to produce is still the driving force behind protecting those rights. The trademark right, a creature of both federal and state law, is a property right that indirectly protects investment in a business by protecting the generation of consumer "goodwill."49 Goodwill is the positive consumer reaction to the quality of goods and services produced by a particular source;50 By using a trademark as a source indicator, consumers are able to apply the goodwill generated from products or services back to the originating company. Without trademark protection, the goodwill created by a business based on its products or services can simply be appropriated by competitors. Capitalizing on the goodwill of another business is a disincentive to investment because it allows free riders to unfairly compete.51 Trademarks, thus, promote investment by ensuring that the intangible benefits received by the investor are not misappropriated by competitors, similar in a broad sense to patent and copyright protection.52 This is the primary benefit which is gained by the property owner. Trademarks also serve to prevent consumer confusion as to the source of the goodwill, which is essentially the avoidance of an additional transaction cost.53 This can be viewed as the secondary benefit which is gained by society.54

Trade secret protection is in some respects the opposite of patent and copyright protection, but in many ways it is also a hybrid. Utilizing trade secret protection represents a choice not to disclose information to the public and to primarily guard its value by preventing access. Of course, no law is needed to encourage businesses to keep information secret; the benefit of keeping one's competitors in the dark is obvious. What trade secret laws attempt to accomplish in an economic sense is to discourage those attempting to gain unauthorized access to the information through some improper means, thereby protecting the business investment from low-cost copyists.55 Trade secrets are opposite to patent and copyright protection in that property protections are granted despite the fact that no creative information is conveyed to the public in exchange for the right. But trade secret protection is also similar to a patent right in that it protects information that is necessarily not already in the public domain56 and similar to a copyright in that unauthorized use of the specific trade secret is the gravamen of a cause of action for misappropriation. Again, as with trademarks, the primary benefit is in the protection of the owner's economic investment against free riders while society benefits from the increased innovation as well as the decrease in unfair competition.

Given the somewhat disparate nature of intellectual property economic incentives and benefits, what economic aspects do they share? The common thread that runs through all of the basic intellectual property rights is the concept of encouraging some ultimately beneficial individual (or corporate) investment by dangling an incentive. It is similar to a contract, in a sense, with society and/or government: if an entity invests sufficiently, special protection for that investment may be available if certain conditions are met. Businessmen, lawyers, economists and sociologists debate the terms of these continuing series of "contracts" and whether one side is getting a better deal than the other:57 This constitutes the majority of the academic literature regarding intellectual property.

But there is another issue with these "contracts" that has significant potential to impact the economic scheme of intellectual property protection: the consequences of changing the terms of a party's "contract" after its rights, privileges and responsibilities have vested. In other words, what does it mean to present and future intellectual property investors when government's or society's commitments under the bargain are changed or avoided, and what will any changes in investor perception mean for society as a whole? The real-life application of these questions occurs in those instances in which the law of intellectual property rights is changed, and those changes have a retroactive effect on existing property rights.

B. Judicial Retroactivity and its Impact on Vested Intellectual Property Rights

The main purpose of our judicial system is to settle disputes between litigants in view of the current state of the law. But the broader impact of the courts on property interests is much greater than that suggests. In the common law system, the concept of stare decisis attempts to create certainty in the law by dictating that a decision rendered in a prior case under the same facts will be applied to parties in a case with similar facts by courts of equal or lesser rank.58 While it is true that modern courts view that doctrine as a policy rather than a mandate,59 it is generally adhered to and functions to add additional structure to the existing legal environment by, in essence, creating judge-made rules.60 As with any legal rule, one expects to be able to rely on judge-made law in undertaking or refraining from undertaking some action or investment.

When a court determines that changing an established legal rule is more important than heeding the doctrine of stare decisis, that decision may affect those who acted in accord with the prior rule if the new rule renders those actions inappropriate. While one can see how a wide range of interests can be affected by judicial rule changing, the effect on property interests is especially dramatic.

1. Property Rights Are Especially Sensitive Retroactivity in Judicial Decisions

The basic issue in radically altering the legal rules underlying-property rights is the upsetting of "reliance" interests.61 Investors62 securing property rights do so with the expectation of a particular return at some point in the future.63 Ideally, the probability of obtaining that return is predicted accurately and may be relied upon and balanced against the investment costs in securing said rights.

However, if a revision of legal rules upsets expectations regarding the return on the investment, the reliance placed by the affected investors is demonstrated to be misplaced (though appropriate when made). If the rule change in question is so unforeseeable that an investor could not have taken it into account when calculating his or her investment costs,64 the investment decision is undermined. Additionally, investment in the development of property or contract rights that are later eliminated is wasted to the extent it exceeds the return that could have been obtained from other uses.65

There are also consequences going forward: if it becomes apparent that such unexpected rule changes may occur in the future and upset the legal environment, investors likely to be affected may be less interested in investing at the same level in future property interests unless they are able to fully account for the risk of change. The situation would have even greater economic significance if the fear of unpredictable and non-assessable rule changes were to become widespread within a particular class of contract or property rights. In that case, significant resources may be redirected into other ventures (assuming alternatives exist), resulting in an under-investment in the affected field.66

Those familiar with the study of "eminent domain" law will recognize these concerns as a commonly-cited rationale for compensating investors in property who experience a loss in property value as a result of legislative rule changing.67 As a general rule, legislatures attempt to avoid these issues and ensure that a new legislative act does not effect an appropriation, or "taking," of existing property rights.68 If it does, just compensation is mandated by the Fifth or Fourteenth Amendment.69 The most common method Congress uses to avoid the takings problem is to constrain the application of the law prospectively, such that existing property rights are not affected.70 Additionally, those proficient in international law will recognize the existence of these considerations for companies contemplating investment in a country with a historically unstable legal or governmental structure.71 Simply put, fear of change can dissuade investors when that change has the potential to cause a loss in property value.72

It is possible to draw parallels between these concepts and judicial determinations that have a retroactive effect. If we acknowledge the way in which courts contribute to the legal environment,73 then it follows that changes in judge-made law that retroactively deprive property owners of property value should have the same effects on those owners as described above.74 Furthermore, because protections are not available to offset the effects of judicial retroactivity,75 the case for some relief in the case of property interests is particularly compelling.

2. Intellectual Property Exhibits an Even Deeper Level of Sensitivity to Retroactivity

Although property rights in general merit special consideration when it comes to relroactivity in judicial decision making, the case that can be made for intellectual property is even more convincing. This is due to the unique relationship between intellectual property value and enforceabilily. In order to appreciate why intellectual property rights are particularly sensitive to judicial determinations, one must understand the unique legal environment that creates the heightened impact.

a. Enforceability is a Tenuous Element of Intellectual Property Value as Particularly Exemplified in the Patent Context

Anyone who has been reasonably attentive to the business world in the last twenty years realizes that patents, trademarks, copyrights and trade secrets can constitute extremely valuable corporate assets. They can have the same influence on a business's bottom line as tangible property assets.76 Note that for some time, accountants have accepted the fact that acquired goodwill may be recorded as a business asset on the balance sheet.77 The value of intellectual property assets, especially patents, can even exceed the value of a business' tangible property.78 Over the years, accountants and other valuation specialists have applied a variety of methods to the assessment of intellectual property value, each having some advantages and disadvantages.79 The so-called "income method" is generally considered the most appropriate for valuing intellectual property assets.80

Naturally, with an increased appreciation of intellectual property value has come the better understanding of the appropriate return on investment in intellectual property. For example, businesses have been especially active in the last ten years in reassessing their existing patent portfolios and plans for obtaining patents based on a more realistic appreciation of the value such property is likely to bring to the company.81 Consulting companies have evolved solely to assist businesses in these matters. Similarly, the assets of some companies consist almost entirely of patent property.

The substantial value conferred by intellectual property carries with it some special valuation issues. In particular, the need to continually assess and reexamine the validity or enforceability of a particular intellectual property right adds a wrinkle to its valuation that transcends analogous considerations for other intangible or tangible assets. That is because the value of an intellectual property right, which is based on the owner's right to exclude access,82 is directly proportional to its enforceability.83 While the same is true for any property right, enforceability of the right to exclude is a more contentious issue with greater uncertainty in the field of intellectual property. Although cases that redraw or clarify the lines of enforceability of real property rights of exclusion do occur, they are quite rare in comparison to the number of cases concerning the validity or enforceability of intellectual property rights.84 The reasons for this are likely two-fold: (1) the recognition and enforcement of intellectual property rights is a relatively new phenomenon, and our society-through the courts-is still in the process of agreeing on the proper boundary lines and (2) the complexity of intellectual property rights does not easily lend itself to certain identification of its boundaries in the same way that tangible property does.85

Intellectual property assets with doubtful enforceability are less likely to produce a return on the investment, e.g., income through licensing. Thus, an intellectual property valuation must take into account facts such as the likelihood that a court will find the right to be valid86 or find that it covers a particular scope of infringing activity if asserted against another.87 These valuation determinations may be made when the right is first obtained, licensed or subsequently challenged in some manner (or for a variety of other reasons).88

Examples of the link between enforceability issues and value are abundant in the field of intellectual property. Trademark rights are open to attack if their boundaries appear to restrict the use of language or product appearance ("trade dress") that is necessary for competition.89 Copyright interests may be narrowed if courts determine that those interests operate to improperly curtail "fair use" of copyrighted material.90 But of all of the classes of intellectual property assets, patents are arguably most vulnerable to an attack on validity and enforceability grounds.91 A more detailed discussion of enforceability and value in the context of patents is the best way to illustrate the concept.92

With regard to validity, issued patents exit the U.S. Patent and Trademark Office ("PTO") with a statutory presumption that the claims93 are properly granted and that the PTO considered all of the relevant facts and issues.94 That presumption can be rebutted in an Article III court proceeding in which infringement of the patent is alleged.95 Such a rebuttal requires a demonstration, established with "clear and convincing evidence," that a mistake was made during the PTO's examination of the patent application.96 Alternatively, anyone, including the patent owner, can request that the PTO "reexamine" the patent to determine if any of the claims were improperly issued in view of the prior art.97 Significantly, the burden of establishing invalidity during a reexamination is by a mere "preponderance of the evidence."98 If any of the claims of a patent are held invalid, they essentially disappear and cannot be asserted against one who practices the claimed invention in the future."

Due to the complexity of the patent examination process and the temporal and financial limitations of the PTO, errors occur and challenges are often undertaken.100 And once the full value of a patent is understood, potential infringers are able to direct more resources into investigating examination errors or oversights than were available during the original examination process.101 Challenges in these circumstances are often successful.102 Thus, the validity of the claims of an issued patent is an important issue that must be considered in attributing the proper weight to the patent.

The enforceability of a patent that is nonetheless valid can still be called into question in light of the patentee's allegedly inequitable acts during the original examination or after issuance.103 The burden of establishing that a patent is unenforceable is the same as showing invalidity in each of the respective adjudicative forums.104 However, since ajudgment of unenforceability generally affects all of the claims of a patent, it can have more far reaching effects.105 Challenges to a patent's enforceability are based to a great extent on a characterization of the patentees' behavior, such allegations are easier to make but more difficult to ultimately prove. As with issues of validity, facts that may affect the enforceability of a patent must also be considered by anyone interested in accurately evaluating the property.

Because corrective procedures always loom which may end up rendering the patent unenforceable, the strength of a patent cannot be taken for granted. Patent strength must be considered by any owner or licensee planning to depend on the patent for market exclusivity as well as any party interested in purchasing or licensing the property for a fair price. Obviously, a patent with a proverbial skeleton in its prosecution history closet has much less value than one which is relatively solid and not open to attack on any reasonably foreseeable grounds.

Moreover, there is an additional danger in relying on a patent of uncertain validity or enforceability due to some rather curious case law from the late 1960s. In Lear v. Adkins,106 the Supreme Court held that a patent licensee cannot be estopped by contract from challenging the validity of a patent in the future and avoiding the license. In contrast, a so-called "licensee estoppel" clause can be incorporated into trademark107 or copyright108 licenses.109 The population interested in challenging the validity of an issued patent is therefore increased due to the fact that a patent licensee has the ability (as well as strong motive110) to invalidate the patent that is the subject of the license. All of this substantially amplifies the uncertainty in depending on the value of the patent property asset.

b. Enhancement of the Effect of Retroactivily

Intelligent property investment decisions of all types are generally supported by an assessment of the legal environment.111 If a subsequent change in the law affects the enforceability of a property right (e.g., limiting the potential remedies against trespassers in the real property context112), the objective value of the property right is necessarily affected. But with most types of property, changes in the legal environment are slow and predictable, and challenges to their enforceability are infrequent.113 Thus, while unexpected changes in the legal environment can retroactively render various investment decisions unsound, for most property interests, the concern is limited.

On the other hand, due to the various aspects of enforceability that can be called into question in conjunction with the many avenues for doing so, the legal environment for intellectual property is in a greater state of flux.114 As the three cases detailed infra demonstrate, a single decision from a relevant appellate court can devalue hundreds or even thousands of intellectual property rights in a broad swath.

A wise individual interested in ascertaining the value of his or her intellectual property must contemplate the relative chance that the property interest will be declared invalid or rendered unenforceable in the future. This fact underlies a significant part of the "due diligence" analysis that is required when an intellectual property asset is purchased or acquired.115 It follows that any changes in the current legal environment that make existing intellectual property rights more likely to be held invalid or unenforceable necessarily affect their value. Congress has acknowledged that fact in enacting legislation that could be viewed as "taking" a portion of an existing intellectual property value by either limiting the legal effect of the legislation to rights granted only in the future116 or determining that no taking occurred after analyzing the economic effect of the legislation on existing rights.117

In summary, intellectual property owners, as a class, are more likely to suffer negative effects of judicial determinations that retroactively revise the law because (1) there are more opportunities and incentives to attack the enforceability of the property right and (2) the complexity and historical novelty of intellectual property leads to a greater incidence of rule changeing. When those determinations are truly unexpected and investors are not able to take them into account, the impact is enhanced and overall economic efficiency is at risk of being reduced. In that case, some mechanism for relief seems natural and logical.

III. THE CURRENT STATE OF THE LAW REGARDING JUDICIAL RETROACTIVITY

In view of the potential economic consequences of the retroactive application of new legal rules in the context of intellectual property, the question as to whether a mechanism for ameliorating the effects is possible in view of the law as it currently stands is critical. Are there any established legal constraints on the retroactive reach of judicial rulings? Can the judiciary limit the application of new rules prospectively? In fact, a body of law has developed that suggests some limitations on the retroactive effect of judicial determinations in general (i.e., not constrained to property deprivations) are appropriate; that law has applicability to the specific rule changing scenario presented in this paper.

The issue of the retroactive application of judicial determinations received considerable attention in both the criminal and civil context in the latter half of the last century.118 But in the grand scheme of American jurisprudence, the debate is relatively recent, coinciding with the judicial activism that characterized the Supreme Court in the 1950s and 60s.119 Surprisingly, the propriety of temporally limiting the impact of judicial determinations is not a settled issue, and some examination of the general landscape as it developed and exists today is warranted before making conclusions or recommendations about the future conduct of courts in considering issues related to intellectual property enforceability.

First, consider the role of the modern judiciary in resolving legal issues not specifically addressed by legislative enactments or constitutions. One can view judicial decisions as falling into one of three basic categories: (1) interpretations of statutes; (2) constitutional interpretations and applications; and (3) creation and application of common law doctrine.120 A substantial body of decisions naturally follows from the latter two types of judicial decision making, but even statutory interpretation, which largely consists of attempts to divine the intent of the legislature, can significantly add to the legal environment. All types have the potential to create legal "regimes," although the common law is arguably more susceptible to the whim of the judiciary (reined in by stare decisis).121 Interestingly, there is some disagreement as to the source from which courts draw the authority to create such legal regimes. Significantly, the theory one applies can directly influence one's conclusion as to whether retroactive application of "new" judge-made rules is legally appropriate (or required).

For years, courts adhered to Blackstone's model of jurisprudence, which is based on the philosophy that judges do not make law; rather, they "find" the law and articulate it in their opinions.122 True lawmaking is left to the legislature.123 Under this theory, there is no notion of retroactivity. If a latter judicial opinion conflicts with an earlier one, a Blackstonian theorist would say that the previous opinion was merely an incorrect understanding.124 In other words, the law doesn't change, only the articulations do at different points in time. In general, reliance on the incorrect articulation is deemed misplaced-the law is the law, and it is a constant.125

Eventually, American jurists began to agree that courts do, in fact, create their own body of law.126 This recognition eventually led to a concern regarding the impact of changing that law (or establishing new law) on those whose actions were based on assumptions made in view of the old legal regime. In essence, there was recognition that judicial decisions can act retroactively, and this retroactivity can have objectively unfair consequences. A parallel was drawn between the judiciary and the legislature: logically, if judge-made law has the ability to effectively change the past, many of the same issues encountered by the legislature in drafting laws are implicated.127

The new aversion to judicial retroactivity first impacted the criminal cases heard by the Warren Court.128 Faced with the possibility of thousands of convicted criminals demanding new trials and release based on a new application of the exclusionary rule,129 the Court in Linkletter v. Walker130 determined that prospective application of the new rule was appropriate. In doing so, the Court found that the retrospective application of judicial decisions is neither required nor prohibited by the Constitution.131 The Linkletter decision was a milestone for the school of jurisprudence that advocated the consideration and, if necessary, avoidance of the overly burdensome retroactive effects of some adjudications.

The Linkletter decision eventually led to the extension of the prospective application scheme in the civil context in Chevron Oil Co. v. Huson.132 In that case, the Supreme Court addressed the proper statute of limitations for personal injuries sustained in an offshore oil rig. The Court found that the shorter limitations period mandated by state law was proper under another recent decision, but refused to apply it to the plaintiff in the case.133 Noting that the plaintiff had no way to know that a shorter limitation period would become the new rule and render his case time-barred, the Court acknowledged this as a problem inherent in judicial law making, and clearly rejected the Blackstonian point of view.134 The Chevron Court articulated their retroactivity analysis as a three-factored test:

First, the decision to be applied nonretroactively must establish a new principle of law, either by overruling clear past precedent on which litigants may have relied . . . or by deciding an issue of first impression whose resolution was not clearly foreshadowed . . . . Second, it has been stressed that "we must . . . weigh the merits and demerits in each case by looking to the prior history of the rule in question, its purpose and effect, and whether retrospective operation will further or retard its operation." . . . Finally, we have weighed the inequity imposed by retroactive application, for "where a decision of this Court could produce substantial inequitable results if applied retroactively, there is ample basis in our cases for avoiding the 'injustice or hardship' by a holding of nonretroactivity."135

Significantly, the Chevron case is an example of "selectively prospective" adjudication, as the new rule regarding the controlling law was applied to the parties in the earlier case, but not the Chevron litigants.136

The question of retroactivity is not necessarily limited to litigants in cases already filed at the time a new rule is annunciated; it can theoretically be applied to any actions undertaken in reliance of the old law that will in the future have some new and different legal significance.137 Thus, under the Chevron analysis, litigants have been able to argue that contracts made or property acquired under a prior legal regime that would be substantially burdened or forfeited under a new legal regime (counter to the original expectations of the property owners) should not be subject to retroactive application of the new regime.138 The same logic and consideration of reliance would seem applicable for intellectual property interests secured in reliance on a prior rule, but the concept has apparently not been applied in that manner.139

Through the 1970s, the Chevron test became a staple for civil actions and remained an important consideration until the tide began to turn in retroactivity analysis. In the in early 1980s, the Supreme Court began to intimate that a reversal of its precedent regarding the propriety of applying some rules prospectively was coming in the criminal context.140 The case that finally heralded the new era was Griffith v. Kentucky,141 and it remains the law in federal criminal jurisprudence to this day. In that case, the Court held that the failure to apply a newly-declared constitutional interpretation to criminal cases pending on direct review was unconstitutional.142 In rejecting that application of prospectivity, the Griffith Court squarely rejected the so-called "clear break" rule for criminal cases.143 To do otherwise, the majority stated, would fly in the face of the principle that the Court does not disregard current law when adjudicating a case pending on direct review.144 Additionally, it would often work to treat similarly situated defendants differently in violation of the Equal Protection Clause.145 However, the Supreme Court's broad reversal of its earlier precedent was limited to the criminal context in Griffith;146 a clear pronouncement on civil retroactivity was reserved for a subsequent case.

It appeared for a short period of time following Griffith that the law in the civil context might remain unchanged from the Chevron era.147 However, the Court eventually adopted the reasoning of Griffith and reversed its stance on a major part of the ability of federal courts to prospectively apply new civil rules. The strongest articulation of the reversal came in Harper v. Virginia Dept. of Taxation.148 Although the basic retroactivity issue in the case, which involved the proper application of another decision declaring a particular type of employee taxation unconstitutional, had been recently addressed by the Court, the Harper majority took the opportunity to reevaluate the situation and clearly state the relevant law. The majority opinion, relying on positions taken by that majority in a previous decision,149 boiled down the Court's shift from the Chevron era to a specific new rule:

When this Court applies a rule of federal law to the parties before it, that rule is the controlling interpretation of federal law and must be given full retroactive effect in all cases still open on direct review and as to all events, regardless of whether such events predate or postdate our announcement of the rule.150

Thus, Harper effectively eliminated the opportunity for prospective application of new rule in a civil case if that rule has been applied to parties in another case.151 This is the current state of the law in the federal civil forum.152 The Harper rule has been applied by courts in intellectual property cases, for example, to justify the reinterpretation of patent claims in cases on review in order to incorporate a new rule of claim construction.153

Significantly, the Harper decision leaves open an important issue regarding a court's ability to limit the application of a decision prospectively: the instance in which a rule is articulated but not applied to the parties in the case-in other words, a fully prospective decision. Such decisions have occurred, but they are relatively rare,154 especially in comparison to fully retrospective decision making.155 In fact, there is a real question as to the continued viability of purely prospective decision making. Some have suggested that an Article III court cannot render a purely prospective decision without violating the case and controversy requirement of the Constitution.156 Others have declared judicial prospectivity to be a clear infringement on the province of the legislature.157 Regardless, purely prospective decision making is an avenue that remains open for the time being and may provide an important option for judicial decision making.

As the brief history presented above demonstrates, all indications are that the trend is moving against the practice of constraining new legal regimes to prospective application. Should this trend be encouraged to its logical end, in which complete retroactivity is mandated in all circumstances? Or is there a reason for permitting an exception for some rules in some situations? In view of the economic effects of judicial retroactivity on intellectual property interests, it seems that a compelling case can be presented for such an exception.

The courts have not been entirely hostile to this proposition in a more general sense. The case law generally reflects the notion that established property and contract rights present themselves as more susceptible to changing legal regimes in both the legislative and judicial forums.158 With respect to retroactivity analysis, courts have acknowledged that the case for limiting retroactivily in view of Chevron is stronger when vested property or contract rights are in question.159

Unfortunately (and somewhat surprisingly), there are not a great number of cases concerning property or contract rights generally-let alone intellectual property rights, specifically-in which courts have actually constrained the application of new rules. Although many courts have addressed the issue, few have found situations worthy of a Chevron limitation.160 Thus, it is not empirically clear that courts have actually found the Chevron test satisfied more frequently in cases of property or contract, despite the fact that they have indicated that such a result should ideally occur.161

Moreover, most of the federal cases that suggest prospective rule making is appropriate when property or contract rights are concerned were decided before the Supreme Court began to partially dismantle the Chevron structure in James B. Beam Distilling Co. v. Georgia162 and Harper. While some decisions suggest an impetus for greater care with regard to vested contract/property interests, it appears that the federal courts are unable (or unwilling) to find a case that actually merits relief from retroactivity.

With respect to state courts-the more likely venue for contract and most property cases-analyzing post-Harper decisions is complicated. This is mainly due to the fact that Harper is generally construed as applying only to decisions regarding federal law. Thus, state courts may adopt or ignore Harper at their discretion.163 In fact, courts have explicitly refused to follow Harper, and have instead chosen to continue to apply the standard Chevron analysis.164 Obviously, these decisions have little bearing on the treatment of property interests for which validity and enforceability issues are controlled by federal law.165 At best, they provide evidence that the rationale behind the Supreme Court's shift from Chevron is not uniformly accepted by the judiciary at large.

IV. "ENHANCED CHEVRON." SUGGESTIONS FOR A MORE AGGRESSIVE APPLICATION OF THE CHEVRON TEST IN THE CASE OF INTELLECTUAL PROPERTY RIGHTS

Because of the special sensitivity of intellectual property to retroactive adjudication and the potential for economically inefficient consequences, the Chevron test is a key limitation that should not be eliminated. In fact, intellectual property presents the perfect platform on which to build back part of the structure of a doctrine that has perhaps been too hastily dismantled. To do so, we must once and for all recognize that courts do make "law" in the sense that they create binding rules for society. When applied retroactively, they have the same effects as legislative determinations vis-a-vis devaluing established property interests. There is no reason that courts should not assess those effects and determine if their rules would be more fairly applied prospectively.

The Chevron decision continues, in the civil context, to give courts the opportunity to assess the effects of their decisions on property interests and plan accordingly. In many respects, it is even more important in the current legal environment to embrace a rational application of the Chevron test for property interests most prominently affected by retroactivity in order to make the case for the continued viability of prospective decision making. Failure to do so may result in Chevron being completely eliminated before the impact of such a decision is fully appreciated.

In light of the sparse utilization of the Chevron limitations, it appears that applying judicial determinations in a purely prospective fashion is not an easy thing for courts to do. The underlying basis for that reticence is easy to understand. First, a court must be willing to openly acknowledge that it is changing the rules. In the case where past precedent exists, it has traditionally been easier to skirt the conflict by distinguishing the prior decisions or implying a conflict of precedent that must be resolved. Second, a court must be willing to buck the clear trend in retroactive adjudication. As explained above, the case for the utilization of prospective adjudication in decisions related to intellectual property is especially compelling and may allow a court to overcome any trepidation.166 If a court were further willing to add some meat to the basic Chevron analysis, which can be fairly criticized as somewhat vague and arbitrary, the doctrine could have an even better chance of a sustained existence in the more limited context of property interests.

Based on an understanding of the above issues and concerns regarding the application of the Chevron doctrine, it is possible to propose a set of recommendations for enhancing the Chevron analysis that can make it more useful and palatable. These recommendations focus primarily on invigorating the first and third Chevron factors by attempting to add more specificity to the test for use with intellectual property interests. Additionally, this paper recommends de-emphasizing the second Chevron factor in recognition of the fact that it merely functions as a means to elevate judicial policy making authority over the interests of property owners and economic efficiency. If recommendations such as these are adopted, they will have the effect of not only protecting the incentive to invest that is central to intellectual property protection, but also of allowing a party involved in a matter ripe for prospective adjudication to make a cogent argument for or against such an application.167

A. Make a Realistic Assessment of When Rules are "New"

Even a strict reading of the first Chevron factor168 allows some ambiguity as to what constitutes "past precedent" or an issue of "first impression."169 It is a relatively simple matter for a court to interpret the prior law such that the new rule can be argued to have always been present170 or not squarely contradict existing precedent.171 To fairly assess the effect of judicial determinations, courts must move past self-serving, artificial interpretations and attempt to truly appreciate the law that was generally understood to control prior to the decision.

In essence, this understanding of the law is the quality that other commentators have recognized as "reliance."172 But, of course, it is impossible for a single court to accurately divine which rules were truly relied upon by the country's intellectual property developers and which ones were viewed with skepticism and concern. In a paper written from the perspective of the "legal process" school, Professor Jill Fisch analyzed the propriety of reliance on existing legal rules by whether those rules were objectively in a state of "stable equilibrium" or "unstable equilibrium."173 Changes to rules in stable equilibrium reasonably upset reliance interests and tend to create unfair and inefficient results.174 Alternatively, changes in rules in unstable equilibrium are not reasonably relied upon and can be made with less concern.175 Professor Fisch's paper thus attempts to separate punctuated changes in legal regimes from those that are part of the natural progression of a developing legal regime, the former affecting existing vested interests much more dramatically.

The equilibrium analysis actually translates quite well to an economic approach. It is reasonable to presume that investors would not rely on rules in a state of unstable equilibrium; the converse is also reasonable. However, a bit more certainty regarding when a legal regime is stable or unstable is required to make the analysis immediately useful. Thankfully, objective rules can be fairly intuitive. A tri-leveled categorization system can be employed to characterize the legal environment with corresponding recommendations for retroactive or prospective application (see Table 1):

IMAGE TABLE 3

Table 1

Although the precise definition of the categories used to assess the degree of reliance can differ, the effect of clearly delineating the test promotes fairness and efficiency in and of itself.

B. Consider That Certain Types of Decisions Naturally Have A Greater Effect on Existing Intellectual Property Rights

Not all decisions related to a given intellectual property asset's enforceability have an equal consequence on all intellectual property assets of a given type.176 The courts should strive to determine the scope of the effect of a particular decision on existing intellectual property rights. This is an enhancement of the third Chevron prong.177 Although a quantitative analysis would be most useful, putting such an analysis together for even a small fraction of the cases related to intellectual property that are heard by U.S. courts each year is probably a practical impossibility.178 One option might be to invite briefing on the subject by counsel, but the adversarial nature of the process might render the information substantially useless. A better alternative might be to set out broad categories of decisions that suggest greater scrutiny in their outcome, i.e., creating a presumption that the effects would be broad (see Table 2).179

IMAGE TABLE 4

Table 2

Of course, more specificity can be derived based on empirical analysis.

C. De-Emphasize the Second Chevron Factor: Curb the Inclination Toward Judicial Policy-Making By Acknowledging That the Legislature is Often a More Appropnate Venue

In considering the philosophical attacks on retroactivity (or adherence to stare decisis, for that matter), a natural counterargument is that courts should not be prevented from revising ineffective rules; indeed, they should be encouraged to do so. Eliminating retroactivity can blunt the corrective effect of new legal rules. This is, in essence, the purpose of the second Chevron factor180 and de-emphasizing it naturally raises some important questions. Should we be restraining courts from righting past wrongs? Would that lead to the conclusion that once a court of sufficient authority rules on an issue affecting certain property rights-rational or not-society is stuck with it?

Yes and no. This paper does argue that, when legal regimes have become sufficiently established and affect broad enough categories of property interests, the courts should leave well enough alone and refrain from rearranging the system. Society is, of course, not stuck with those rules because they can be legislatively overruled. In that case, property that is devalued is ostensibly protected by the provisions of the Fifth or Fourteenth Amendment. More than anything, this recognizes that courts make law that is relied upon by investors in property, and it makes sense to apply the same protections that one expects when legislative enactments change the rules of the game.181

The second Chevron factor merely provides an avenue for the courts to give special weight to the policy goals of the new rule. It suggests that if a new rule's objectives are furthered by application to existing interests, that fact can counter the negative effects on property owners. However, this view tends to encourage encroachment of the judiciary into the powers of the legislature with none of the Constitution's protections. Courts should acknowledge that the legislature is the better forum for making broad, sweeping changes to the established legal environment, at least when property interests are concerned.182 This maximizes predictability in the rules, which promotes efficiency and fairness overall. Only when a clear choice cannot be derived from an analysis of the first and third factors should courts consider the second Chevron factor.

Additionally, if a court were to give full weight to the second Chevron factor, in the case of intellectual property rules, it is not likely to provide information of much value. This is because the explicit "purpose and effect"183 of a rule affecting the enforcement of intellectual property rights is almost always "furthered" by applying it to existing interests, but if the implicit goals of intellectual property rules are included in the analysis, the net effect probably works out to be very small. To make a complete assessment of the "purpose and effect" of most new rules related to intellectual property interests, one must take into account the impact on the incentive scheme discussed earlier. A rule that simplifies the conduct of litigation but ends up discouraging or invalidating investment in the system may have no net positive effect on economic efficiency. Additionally, attempting to actually quantify the net purpose and effect would likely be a very difficult proposition, and when the impact of that analysis is likely to be small, it is a less than productive but overly inflated part of the overall Chevron test.

The analysis in the above three sections can be combined to produce a matrix (Figure 1)184 that better allows one to assess the propriety of prospective rule application. In this manner, prongs one and three of the Chevron test are enhanced to better address the issues related to intellectual property interests.

IMAGE ILLUSTRATION 5

Figure 1

V. A REVIEW AND REINTERPRETATION OF THREE REGENT CASES: A DEMONSTRATION OF HOW AN "ENHANCED CHEVRON" TEST MIGHT WORK IN PRACTICE

Three recent decisions by the U.S. Court of Appeals for the Federal Circuit185 crystallize the issues discussed in this paper and provide a basis for demonstrating how Chevron may be more effectively employed. In Festo Corp. v. Shoketsu Kinzoki Kogyo Kabushiki Co.,186 Symbol Technologies, Inc. v. Lemelson Medical, Education & Research Foundation187 and Johnson & Johnston Assoc. Inc. v. R.E. Service Co.,188 we see demonstrations of the potential court decisions have to retroactively diminish intellectual property value by placing the enforceability of affected patents in doubt. The exact diminution individual property owners will incur as a result of these cases is difficult to quantify until the occurrence of a critical event (like litigation), and it will certainly vary depending on the particular patent in question. But the three cases discussed herein take the retroactive weakening of existing intellectual property rights to an extreme by calling into question broad conduct and practices in the prosecution189 of patents. The broad reach of these decisions has created a major stir in the patent community and will force many to reevaluate patent property rights once thought secure. Thus, they provide excellent material for analysis and discussion. Furthermore, the Supreme Court's recent reversal of the Festo decision-based in part on the same reasoning used in this paper to support prospective decision making-provides an excellent example of the only fair alternative to making use of Chevron limitations: refraining from significantly changing the rules at all.190

A. The Extension of File History Estoppel in Festo: Retroactive Revision of the Common Law

In the fall of 2000, an en banc panel of the Federal Circuit handed down its controversial decision in Festo Corp. v. Shoketsu Kinzoki Kogyo Kabushiki Co.191 Few cases in recent memory have elicited so much commentary and general gnashing of teeth in the patent bar.192 Of course, the Supreme Court's recent reversal substantially eviscerates the Federal Circuit's dramatic decision and its critics are largely placated.193 However, as a stand-alone decision, the Federal Circuit ruling is worth discussing as a powerful exhibition of judicial retroactivity issues.

1. Factual Background and Decision

The case involved Shoketsu Kinzoki Kogyo Kabushiki Co.'s (referred to in the decision and heretofore as "SMC") alleged infringement of Festo's patents for "magnetically coupled rodless cylinders."194 The wrinkle in the case began with the fact that Festo did not claim that SMC infringed by practicing the literal language of the claims.195 Rather, Festo accused SMC of infringing under the "doctrine of equivalents." The doctrine of equivalents is a judicially-created set of guidelines that prevents one from escaping infringement merely by making insubstantial changes to the invention as described in the claims.196 It ensures that a patentee is not tied to a particular phrasing of the invention in the claim language when an alleged infringer is clearly practicing its "equivalent."197

SMG defended against Festo's infringement claims on the grounds that Festo had waived its right to equivalents for certain aspects of the claimed invention by entering narrowing "amendments"198 during the prosecution of the patents-the so-called "prosecution history estoppel" defense.199 According to SMC, by entering the amendments, Festo rejected any claim to subject matter specifically excluded by the amendments. Festo countered that it had not intended to narrow its claims and that it was entitled to a range of equivalents that included SMC's allegedly infringing rodless cylinders. Festo prevailed on a motion for summary judgment of infringement in the U.S. District Court for the District of Massachusetts and, following a jury trial, judgment was entered against SMC for infringement under the doctrine of equivalents.200

On appeal of the summary judgment motion, the Federal Circuit addressed the case en banc,201 primarily to clarify the law relating to the application of the doctrine of equivalents to amended claims.202 The most significant question addressed by the Federal Circuit, was whether an applicant who narrows a claim element210 by amendment is barred from asserting infringement over allegedly equivalent acts or devices that fall outside of the narrowed element.204 After reviewing its past precedent and concluding that the answer depended on which of two lines of cases was followed, the Federal Circuit reversed the line that had previously been established as controlling precedent and held that claims narrowed by amendment for purposes of patentability may be accorded no coverage of equivalents for the amended element.205 This is known as the "complete bar" rule.206

In reversing its earlier precedent, the majority stated that the court's experience since its inception207 led it to conclude that the previously established rule, known as the "flexible bar" rule, was unworkable due to its inherent ambiguities.208 The majority favored the complete bar rule because of its tendency to support the principle that a patent must give notice to the world as to what the patentee claims to be his or her invention.209

2. Appropriate Retroactive Effect on Existing Patents

From the moment the decision appeared in print, a swell of controversy ensued. Many patent owners as well as members of the patent bar complained that the decision unfairly called into question the scope of enforceability of many patents with amendments in their prosecution history.210 The problem according to those concerned is that many patent prosecutors believed (supported by previous Federal Circuit decisions) that claim amendments could be entered without necessarily surrendering all equivalents. In relying on that belief, patent practitioners adopted a claim drafting practice that entailed the use of broad initial claims followed by amendments in order to define the invention. In contrast-and often cited as the proof of the inequity of the Federal Circuit's Festo-is the fact that, if an identically-worded claim had been submitted in the original application without amendment, that claim would be entitled to a range of equivalents. Patent practitioners can reasonably claim that, if they had known of the potential effects of amending, they would have made different decisions. Of course, the file history for existing patents is set in proverbial stone and cannot be undone.

As a result of the Federal Circuit's Festo decision, some patent claims could have been limited to their literal terms (at least for certain elements). Previously broad and powerful patent claims might have only been enforceable over an objectively narrow range of potentially infringing acts. Moreover, competitors would know exactly what aspects of the invention to change in order to practice just outside of the claim language, but still make use of the essence of the invention. That would make any patent so affected potentially much less valuable-even worthless in some cases.211 In his statements on behalf of Festo, Robert H. Bork noted, " [I]f what happened to Festo here and what's going to happen to other patent holders continues, this case, were it not done by a court, were it done by any other agency, would be a taking under the Constitution."212

Turning to the analysis in Part IV of this paper, we can see that the Federal Circuit's Festo decision would have been ripe for prospective treatment. The court acknowledged that prior decisions constituting binding precedent were being overruled.213 Under prong one, that strongly suggests that retroactive treatment would undermine patent owner reliance. Additionally, the Festo decision would have affected a large number of patents because it changes the rules regarding a common prosecution practice. However, the effect is tempered by the fact that only a part of the patent right, equivalents to the literal claim scope, were being eliminated. Putting it all together in the matrix (see Figure 2) shows that prospective treatment of the decision (which necessarily includes the parties in the case) would have been appropriate.

3. The Supreme Court's Post-Script: Reliance Interests Take Precedence

In a move that was predicted by most of those interested in this somewhat technical case, the Supreme Court soundly rejected the Federal Circuit's "absolute bar" rule, adopting instead a much more flexible rule creating a rebuttable presumption against equivalents.214 The decision, issued in May 2002, is particularly interesting because it bases its reversal on precisely the same reliance interests discussed Part III above as the primary rationale for utilizing prospective decision making.215 Essentially, the Supreme Court explicitly recognized the retroactivity problem, but addressed the issue by refraining from significantly altering the law, leaving any such change to the legislature.216

Justice Kennedy, writing for a unanimous Court, expressed the opinion that the Federal Circuit's rule was inconsistent with the purposes of the doctrine of file history estoppel, as well as the mandate in the Supreme Court's decision in Warner-Jenkinson.217 In doing so, he explained why the Federal Circuit decision was inappropriate in terms of its affect on investment-backed expectations:

The Court of Appeals ignored the guidance of Warner-Jenkinson, which instructed that courts must be cautious before adopting changes that disrupt the settled expectations of the inventing community. . . . In that case we made it clear that the doctrine of equivalents and the rule of prosecution history estoppel are settled law. The responsibility for changing them rests with Congress . . . . Fundamental alterations in these rules risk destroying the legitimate expectations of inventors in their property. The petitioner in Warner-Jenkinson requested another bright-line rule that would have provided more certainty in determining when estoppel applies but at the cost of disrupting the expectations of countless existing patent holders. We rejected that approach: "To change so substantially the rules of the game now could very well subvert the various balances the PTO sought to strike when issuing the numerous patents which have not yet expired and which would be affected by our deasion."218

The Court emphasized the harm occurring from retroactive application of the Federal Circuit's rule as a reason for finding the judiciary impotent to make such a significant change.219 Instead, the Court adopted a much more flexible approach in which the patentee bears the burden of showing that the amendment in question does not surrender equivalents, and if that burden is satisfied, equivalents are available for the claimed element.220 The Court did not discuss the alternative of prospective application, probably because it was not raised by the parties.221 Interestingly, one should note that if the Federal Circuit had chosen to limit their decision to prospective application only, a great deal of the Supreme Court's rationale for reversing would not exist.

Although the Court held that it was doing nothing more than applying the law of its previous decision in Warner-Jenkinson, neither that case nor any subsequent decision explicitly places the burden on the patentee to establish a right to equivalents.222 Therefore, it can be said that some minor change in the law has resulted, though far smaller than in the Federal Circuit decision. If one adjusts the placement of this decision on the "Change in Legal Environment" side of the matrix (see Figure 2), one can see that the Supreme Court revision of Festo is properly retroactive.223

IMAGE ILLUSTRATION 6

Figure 2

B. Opening the Door for the "Prosecution Laches" Defense in Symbol: Retroactive Revision of Statutory Interpretation?

In the Federal Circuit's recent decision in Symbol Technologies, Inc. v. Lemelson Medical, Education & Research Foundation,224 the court awakened a defense based on the use of a common patent prosecution practice that most in the patent bar thought was long put to rest. Following so quickly on the heels of Festo, it demonstrates the retroactivity phenomenon is not merely a single-case issue, but an ongoing concern that must be continually assessed by patent owners.

1. Factual Background and Decision

This case involved several patents resulting from applications originally filed by the famous (infamous to some) late inventor, Jerome H. Lemelson in 1954 and 1956.225 Lemelson is well-known to patent aficionados as the most prominent practitioner of the "submarine patent"-a patent based on an application filed many years before and kept alive by filing "continuation applications"226 in the PTO until finally issuing with claims drawn to cover the current state of the art. The "submarine" label derives from the fact that U.S. patent applications have historically been kept confidential unless and until a patent is issued227; thus, no one is aware of the content of a submarine patent until it "surfaces" when issued.228 Although Lemelson died in 1997, his patent empire is kept alive by a foundation that maintains and enforces them.229 The Lemelson patents involved in the Symbol case230 related generally to machine vision and automatic identification technology.

The plaintiffs, a group of companies collectively referred to in the decision as "Symbol" and another company known as "Cognex," designed, manufactured and sold bar code scanners and related products. After receiving letters from the Lemelson Foundation accusing Symbol and Cognex of infringing various Lemelson patents, the plaintiffs filed independent actions (later consolidated) against the Lemelson Foundation for declaratory judgment that the patents in question were invalid, unenforceable and not infringed. Among the bases for judgment detailed in the Symbol and Cognex's Complaints was the contention that Lemelson committed prosecution "laches"231 by unduly delaying the issuance of claims based on the 1950s applications.232 The Lemelson Medical, Education & Research Foundation ("Lemelson") requested dismissal of the case, in part because the prosecution laches defense failed to state a claim upon which relief could be granted.233 Although the trial court refused to dismiss the case entirely, it did dismiss the prosecution laches defense.234

The only issue appealed to the Federal Circuit as a result of the district court's ruling was the dismissal of the prosecution laches defense. In reviewing the plaintiffs' arguments, the majority of the three-judge panel noted that the prosecution laches defense had a basis in two Supreme Court decisions from the 1920s, Woodbridge v. United States235 and Webster Electric Co. v. Splitdorf,236 which found that delays of nine and eight years, respectively, were sufficient to establish laches on the part of the patent applicant.237 The majority affirmed the continued existence of the defense and dismissed Lemelson's arguments as to why the court might be constrained from doing so, including that the Federal Circuit was bound by earlier, "nonprecedential" Federal Circuit decisions rejecting the prosecution laches defense238 and that the defense was legislatively foreclosed by the revisions of the Patent Act of 1952.239 As to the latter argument, the majority determined that the defense of laches was implicitly incorporated into the infringement defenses detailed into the act.240 Under the majority's view, this included not only laches with respect to filing the litigation, but also laches in prosecuting the patent.241 Thus, the majority concluded that even if a patent applicant followed the statutory procedures and administrative rules for obtaining a patent (including meeting the many timing requirements already in the system, such as filing deadlines), the applicant could nonetheless face an equitable claim of delay in prosecuting the claims.242

2. Appropriate Retroactive Effect on Existing Patents

As with Festo, the Symbol decision quickly raised eyebrows in the patent bar. The complaint is similar: Symbol retroactively calls into question a common practice in patent prosecution long believed to be entirely legitimate and proper.243 As a result, a patent may be rendered unenforceable for "delay" that was explicitly provided for by the patent statute and PTO rules. Many patents now in force could be affected because they were obtained following one or more continuations from an original application. Patent owners who placed value on their patent rights in partial reliance upon the generally-accepted notion that the defense was eliminated with the 1952 revision to the Patent Act were taken by surprise. To add salt to the wound, many patent owners likely shored-up that reliance following two non-precedential Federal Circuit decisions that at least implied that the court did not view the defense with favor; the court's recent pivot was therefore even more of a shock to those diligent in keeping abreast of the case law in order to accurately valuate patent portfolios.244

The potential inequity behind the retroactive effect was highlighted by Judge Pauline Newman in her dissent. Judge Newman focused on the statutory foreclosure argument, noting that the drafters of the 1952 Act specifically provided for the procedure of keeping a patent application alive over several years through the continuation process under 35 U.S.C. 120 and 121.245 Echoing the essence of the retroactivity arguments in Festo, Judge Newman stated that

It simply adds to the uncertainties of the patent grant, to create a new cause of action whereby routine actions, in full statutory compliance, can years later be challenged as having been done too late or having taken too long.246

Those added uncertainties can have a real effect on the value of issued patents. As a result of Symbol, patent owners must take into account the prosecution laches defense in evaluating the strength of their patent properties, and any potential reduction in value.

Referring to the temporal application analysis in Part IV, we see that the Symbol decision is strongly indicated for prospective treatment. The court changed a rule that was not only generally thought to be settled by the enactment of the 1952 Patent Act, but also contravened recent non-precedential opinions by previous Federal Circuit panels. Additionally, if the doctrine of "prosecution laches" becomes a widely used defense, it has the potential to render many patents wholly unenforceable. Using the matrix, we see that Symbol falls squarely within the prospective adjudication field (see Figure 3).

IMAGE ILLUSTRATION 7

Figure 3

C. Johnson's Rules for Claiming What is Detailed in the Specification: Clarifying a Nominally Simple Rule

In another recent and heavily-watched case, Johnson & Johnston Assoc. Inc. v. R.E. Service Co.247 the Federal Circuit reviewed the legal rules regarding another patent prosecution-related issue that can ultimately affect claim breadth. Hailed by many as the latest "nail in the coffin" of the Doctrine of Equivalents,248 the case caused as much concern for what it is perceived to say about the court's attitude toward the Doctrine than its retroactive effect. However, the case may force many patent owners to reevaluate the scope of their patents which have long since passed out of prosecution.

1. Factual Background and Decision

The case concerned Johnson & Johnston Associates, Inc.'s ("Johnson") patent related to the manufacture of printed circuit boards.249 Johnson's invention-at-issue was a circuit board assembly that included a metal substrate which attaches to etched copper circuits to prevent damage that can be caused by worker handling. Johnson brought an action against R.E. Service Co., Inc ("RES") for infringement of the patent by manufacturing copper circuits with a stainless steel substrate and claimed damages in excess of $1 million. Following a trial, a jury found RES guilty of infringement under the Doctrine of Equivalents250 and awarded Johnson $1,138,764 in damages.251 The judge increased the award to $2,957,342 as a consequence of the jury's finding that the infringement was willful.252

RES appealed the judgment on a question of claim scope and interpretation. The main issue was Johnson's ability to claim certain materials as equivalents based on the wording of its claim language in view of the specification. Although the specification253 in Johnson's patent referred to several materials as options for the substrate, including aluminum as the preferred metal as well as stainless steel and nickel alloys, the claims of the patent mentioned only the use of aluminum. RES claimed that Johnson's mention of several substrate options in the specification without including them in the literal scope of the claims worked Lo dedicate that material to the public under long-standing patent law doctrine.254 Johnson would therefore be prevented from including the dedicated material within scope of the invention under the Doctrine of Equivalents.255 If RES's argument were adopted, Johnson's patent could not cover RES's use of a stainless steel substrate in place of the claimed aluminum.

The Federal Circuit took the case en banc, in part, because it recognized that the resolution to the issue on appeal depended on the adoption of one of two lines of arguably divergent Federal Circuit precedent. The position propounded by RES was reflected in the 1996 case, Maxwell v. J. Baker, Inc., wherein the Federal Circuit refused to find as equivalent an iteration of a fastening system for shoes that was stated in the specification but not claimed.256 Johnson's position was reflected by the 1998 case, YBM Magnex, Inc. v. Int'l Trade Comm'n,257 in which the court held that the Maxwell decision did not create an absolute rule against finding equivalents in a disclosed-but-not-claimed scenario,258 and proceeded to reinstate a finding of infringement under the doctrine of equivalents in exactly that instance.259

The Johnson court essentially adopted the Maxwell precedent and overruled the YBM case law.260 In doing so, the court focused on the importance of being able to rely on the claims of a patent as defining the invention and giving notice to the world. The court held that allowing a patentee to claim through the doctrine of equivalents that which he or she "deliberately left unclaimed" would "conflict with the primacy of the claims in defining the scope of the patentee's exclusive right."261 In the court's view, it could allow a patentee to narrowly claim the invention to avoid scrutiny during prosecution in the PTO, only to recapture it in litigation.262 The judgment in favor of Johnson was wholly reversed.263

2. Appropriate Retroactive Effect on Existing Patents

Most dramatic about the Johnson decision is its rather strict rule prohibiting the capture of subject matter disclosed in the specification but not claimed. According to a singular and vociferous dissenting opinion written by Judge Newman,264 the majority's "new rule" was an unnecessary exercise of judicial power.265 Judge Newman believed that the case could have been resolved within the framework of both YBM and Maxwell, and she voiced concern that patent owners would now have a significantly lowered incentive to disclose helpful information in the specification:

The public interest in fostering innovation and technological advance is not served by a judicial decision that imposes legal obstacles to the disclosure of scientific and technological information. Information dissemination is a critical purpose of the patent system. By penalizing the inclusion of information in the specification the patent becomes less useful as a source of knowledge, and a more guarded legal contract.266

Although Judge Newman cited several sources to support her argument that the majority's decision was economically unsound, she did not specifically note the retroactive economic effects on issued patents. It is, however, possible to consider at least one. One could argue that some patent applicants drafted their disclosures with an eye toward the YBM decision, believing that a more inclusive disclosure would better support the enablement and written description requirements. Those applicants, one might contend, expected that the disclosure would not disenfranchise them of disclosed equivalents as long as those equivalents fit within the constraints of the YBM decision.

Turning to the analysis of future application in Part IV, we see that the Johnson decision is probably best applied retroactively. The prior rule supporting the appellant's contention was less than clear, depending merely on one case that stood in the face of prior precedent. Moreover, although the case could affect a broad class of patents, it only regards claim equivalents. Therefore, using the matrix (Figure 4), we see that Johnson falls within a field indicating retroactive application.

IMAGE ILLUSTRATION 8

Figure 4

VI. CONCLUSION

The concern over the relroactivity of judicial determination has waned in recent years in view of a significant shift in the attitude of the Supreme Court. However, from an economic and social perspective, the issue is as important today as ever. Particular categories of property rights may provide the incentive to test the continued viability of prospective judicial decision making and preserve the useful remnants of the doctrine articulated in Chevron. Intellectual property interests present an exceptional opportunity for prospective decision making because they present special retroactivity problems due to the connection between enforceability and value and the fact that the case law makes this form of property more susceptible to changes in judicially-created rules. The courts should give increased consideration to the prospective application of its rules in certain circumstances by creating a more workable and specific version of the Chevron test for intellectual property interests.

IMAGE FORMULA 9IMAGE FORMULA 10IMAGE FORMULA 11IMAGE FORMULA 12IMAGE FORMULA 13IMAGE FORMULA 14IMAGE FORMULA 15IMAGE FORMULA 16IMAGE FORMULA 17IMAGE FORMULA 18IMAGE FORMULA 19IMAGE FORMULA 20IMAGE FORMULA 21IMAGE FORMULA 22IMAGE FORMULA 23IMAGE FORMULA 24IMAGE FORMULA 25IMAGE FORMULA 26IMAGE FORMULA 27IMAGE FORMULA 28IMAGE FORMULA 29IMAGE FORMULA 30IMAGE FORMULA 31IMAGE FORMULA 32IMAGE FORMULA 33IMAGE FORMULA 34IMAGE FORMULA 35IMAGE FORMULA 36IMAGE FORMULA 37IMAGE FORMULA 38IMAGE FORMULA 39IMAGE FORMULA 40IMAGE FORMULA 41IMAGE FORMULA 42IMAGE FORMULA 43IMAGE FORMULA 44IMAGE FORMULA 45IMAGE FORMULA 46IMAGE FORMULA 47IMAGE FORMULA 48IMAGE FORMULA 49IMAGE FORMULA 50IMAGE FORMULA 51IMAGE FORMULA 52IMAGE FORMULA 53IMAGE FORMULA 54IMAGE FORMULA 55IMAGE FORMULA 56IMAGE FORMULA 57IMAGE FORMULA 58IMAGE FORMULA 59IMAGE FORMULA 60IMAGE FORMULA 61IMAGE FORMULA 62IMAGE FORMULA 63IMAGE FORMULA 64IMAGE FORMULA 65IMAGE FORMULA 66IMAGE FORMULA 67IMAGE FORMULA 68IMAGE FORMULA 69IMAGE FORMULA 70AUTHOR_AFFILIATION

Daniel R. Cahoy*

AUTHOR_AFFILIATION

* Assistant Professor of Business Law, The Pennsylvania State University, Smeal College of Business. Thanks to Austin Jaffe and Jeff Sharp for helpful comments and advice. Earlier versions were presented at the Mid-Atlantic Academy of Legal Studies in Business Annual Meeting in Philadelphia, Pennsylvania, April 2002, and the Academy of Legal Studies in Business ("ALSB") Annual Meeting in Las Vegas, Nevada, August 2002. This paper was awarded the 2002 Homes-Cardozo Award for Outstanding Conference Paper at the ALSB's Las Vegas conference.