Rules and standards on the forefront of patentability.

• Author: Duffy, John F.
• Position: Boundaries of Intellectual Property Symposium

TABLE OF CONTENTS INTRODUCTION I. CHALLENGES IN DEFINING THE LIMITS OF PATENTABILITY II. THE HISTORY OF FAILED PATENTABILITY RULES A. The Unpatentability of Changes in Form and Proportions B. The Unpatentability of Plants and Animals C. The Unpatentability of New Uses D. The Unpatentability of Methods of Medical Treatment III. THE MODEST SUCCESS OF PATENTABILITY STANDARDS A. The Unpatentability of Natural Principles and Phenomena B. The Unpatentability of Abstract Ideas IV. THE FATE OF THE RULE FROM THE BILSKI EN BANC CONCLUSION: THE FAILURE OF RULES AND THE VALUE OF FAILURE INTRODUCTION

Courts and legislatures face a fundamental dilemma in constructing the law of patents. Patents convey property rights, and a substantial degree of certainty is usually thought to be helpful, or even essential to well functioning property rights. (1) Yet patents also cover invention, and human inventiveness by its nature unsettles certainty, changes the status quo, and breaks through preexisting assumptions.

In legal doctrine, the conflict between certainty and creativity plays out within the familiar jurisprudential debate between rules and standards. (2) Clear rules can provide the certainty that encourages investment both in obtaining and developing the rights, but standards can provide the flexibility to accommodate the new and unpredictable wonders of human ingenuity. The stakes of this traditional debate are highest for the doctrine of patentable subject matter, which governs the fundamental boundaries of the patent law's domain.

The latest controversy in the field of patentable subject matter provides a perfect example. In the en banc decision In re Bilski, (3) rendered in the fall of 2008, the United States Court of Appeals for the Federal Circuit overruled its prior doctrinal test and established a new rule--not a flexible standard--for determining whether a process is patentable subject matter: "[A] claimed process is surely patent-eligible under [section] 101," the Bilski majority confidently announced, "if: (1) it is tied to a particular machine or apparatus, or (2) it transforms a particular article into a different state or thing." (4)

The Federal Circuit identified that its overarching goal was to "clarify the standards applicable in determining whether a claimed method constitutes a statutory 'process' under [section] 101." (5) Because clarification was the goal, it is unsurprising that the Federal Circuit attempted to articulate a definite rule to govern this area of law, and the court plainly understood that it was choosing to attempt a more rule-based approach in this area of law. The court repeatedly referred to its new doctrine as "the machine-or-transformation test," (6) emphasized that its new test "is the only applicable test and must be applied ... when evaluating the patent-eligibility of process claims," (7) and warned that "[n]either the PTO nor the courts may pay short shrift to the machine-or-transformation test by using purported equivalents or shortcuts such as a 'technological arts' requirement." (8) Nothing in the en banc opinion suggested that the court envisioned that future three-judge panels or Patent and Trademark Office (PTO) examiners would have to weigh and balance a variety of competing policy considerations to apply the court's new machine-or-transformation test.

In announcing its new test, the Bilski court expressly disavowed the analysis that had been set forth in its 1994 en banc decision In re Alappat, (9) and had been applied in its 1998 and 1999 panel decisions, State Street Bank & Trust Co. v. Signature Financial Group (10) and AT& T Corp. v. Excel Communications, Inc. (11) That line of cases had, in turn, disavowed an even earlier test, known as the Freeman-Walter-Abele test, which derived from three cases decided between 1978 and 1982. (12) Thus, the Federal Circuit and its predecessor court have changed the rules governing patentable subject matter no less than three times in thirty years.

While it is true that the en banc Bilski court was vigorously pursuing the goal of clarifying the law of patentable subject matter, that objective does not distinguish the present from the past. In the now disavowed AT&T decision, for example, the Federal Circuit had reassured the practicing bar that the "concern, that the rules [of patentable subject matter] are not sufficiently clear to enable reasonable prediction of outcomes, should be less of a concern today [1999] in light of the refocusing of the [section] 101 issue that Alappat and State Street have provided." (13) Not only did that reassurance fail to remain valid for the statutory life of a patent (approximately twenty years); it failed to remain good even for the time of prosecuting some patent applications. The very application at issue in Bilski had been already pending for two years when the AT&T court was touting the clarity that Alappat and State Street had brought to the law of patentable subject matter.

Since the announcement of the en banc court's opinion, it has become evident that the Federal Circuit will not have the final word as to whether the machine-or-transformation test will be embraced as a means for clarifying the law of patentable subject matter. On June 1, 2009, the Supreme Court granted certiorari in the Bilski case. (14) This development dramatically increases the importance of the case. In light of the Federal Circuit's past failures to bring clarity to this area, the practicing patent bar--and more importantly their clients, the inventors of our society--might reasonably wonder whether the Supreme Court will also endorse the machine-or-transformation test or some other new rule in an attempt to bring clarity to the area. More importantly, the inventors and their attorneys will want to know (and indeed, to make investments, they may need to know) how durable any new test promulgated by the Supreme Court is likely to be.

That question is vitally important because subject matter doctrine governs the fundamental scope of the patent system. Changes to the doctrine have the potential to exclude from the system whole fields of endeavor, such as financial innovations, data processing methods, operations engineering, and diagnostic techniques. By contrast, changes to other patent doctrines such as obviousness law, enablement requirements, or the doctrine of equivalents might change the fate of some claims to invention or infringement. But at least in most instances, those claims would have been recognized as having marginal quality even under a somewhat different set of doctrinal rules.

Moreover, clarity without durability has limited value for a system in which long term investment in tomorrow's innovations is supposed to be fostered through property rights lasting for two decades. (15) For such a system, a clear but transient rule may be inferior to a standard that is less clear and less predictable in application, but more durable.

Evaluating the longer historical record, this Article makes two points about rules and standards in this field. First, for patentable subject matter, there is a clear winner in the long run. Eventually, rules always fail. This should surprise no one who studies innovation. The unruly process of creative destruction has the power to undermine today's legal rules every bit as much as it renders obsolete today's industrial products, processes, and institutions. Moreover, the long term failure of rules in defining patentability is also consistent with general models that predict standards to be more durable than rules when conditions are changing, and innovation presents a quintessential circumstance of change. (16) Thus, patent lawyers and inventors need to take into account that any apparent clarity provided by a newly promulgated rule of patentable subject matter may not survive any longer than the rules provided by Alappat, State Street, AT&T, and their predecessors.

A second point, which should also be familiar to students of innovation policy, provides some consolation to the proponents of rules. Just as in innovation generally, failures have value. The short term certainty associated with rules may provide necessary, if temporary, safe harbors that allow property rights to thrive. And even in the long run, the repeated failures of patentable subject matter rules provide crucial insights into the meaning and process of invention in our society.

This history of failures should not necessarily dissuade courts from ever attempting to create relatively clear rules of patentable subject matter, but the history provides some practical lessons into how courts should go about making and interpreting rules when they do decide to create them. Though knowing their rules will ultimately fail, courts might reasonably decide to fashion rules if the rules have a good chance at meaningful durability, but "durability" in the context of a patentable subject matter rule must, at the very least, exceed the statutory lifetime of a patent. A rule of patentability expected to endure for ten or fifteen years should not be considered a success. For such a short-lived rule, rational inventors would appropriately discount the apparent certainty of today's rule and attempt, to the extent possible, to average today's rule with the possible permutations of rules that might follow in future years. Of course, inventors may have very limited ability to forecast the precise contours of future new rules, but their limited ability to forecast the future simply makes the apparent certainty of today's rule all the more useless.

The need for durability in patentability rules also has a corollary: during its lifetime, the rule must be such that judges of differing policy perspectives can agree to the rule and be willing to tolerate its inevitably formalistic boundaries. Such a rule does not need to approximate the policy preferences of the "median judge," with an approximate balance between those judges...