Defining patent scope by the novelty of the idea.

• Author: Chiang, Tun-Jen
• Position: I. Introduction through II. The Incoherence of Disclosure Theory, p. 1211-1239

TABLE OF CONTENTS INTRODUCTION I. THE INVENTION AS IDEA AND EMBODIMENT A. The Embodiment-Centric Assessment of Novelty 1. The Patent Specification Describes an Embodiment 2. Claiming the Inventive Features 3. Comparing the Claimed Features to Prior Embodiments B. Claims and Downstream Coverage of Ideas II. THE INCOHERENCE OF DISCLOSURE THEORY A. Disclosure Theory in Doctrine 1. Section 112 2. The Abstract Idea Doctrine B. The Economic Balancing Alternative and Its Problems C. The Costs of Incoherence D. The Proposals in Existing Literature III. UPSTREAM CAPTURE: A DIFFERENT VIEW OF THE PROBLEM A. Upstream Capture Versus Downstream Capture B. A Proposed Solution: The Patentee's Insight C. Implications of the Framework 1. Abolishing the Enablement Requirement for Prior Art 2. Reconceptualizing the Abstract Idea Doctrine 3. Untying Patent Scope from Disclosure 4. The Role of Non-obviousness IV. RESPONSES TO POSSIBLE CONCERNS A. The Science Fiction Objection 1. Science Fiction Writers Preempting Later Patents 2. Science Fiction Writers Obtaining Their Own Patent B. The Imprecision of Ideas C. The Over- and Under-Inclusiveness of a Rule D. A Rebuttable Presumption as an Extension CONCLUSION INTRODUCTION

The standard theory of patent scope is disclosure. It is axiomatic that the invention "claimed by the patent application must be the same as what is disclosed in the specification." (1) This reflects patent law's fundamental quid pro quo: the monopoly covers the same invention as what the patentee contributes to society through disclosure. (2)

Disclosure theory, however, is incoherent. For example, it is often said that the Wright brothers invented the "airplane." (3) But of course the Wright brothers did not invent the idea of airplanes. They invented only one airplane embodiment: a single barely-flying wooden glider. (4) A strict application of the quid pro quo principle might therefore say that the Wright brothers should be confined to replicas of their wooden glider. But this would eviscerate patent incentives, because a later pirate would change a few nuts and bolts while copying the core aerodynamic concepts. (5) Courts therefore define the invention more abstractly, as an "airplane" and not a "wooden glider airplane." (6)

But once we reject confining patent scope to the precise embodiment that has been disclosed, there is no obvious principled limit to this abstraction process. (7) For example, did the Wright brothers invent:

(1) "A wooden flying machine with wings and rudders"?

(2) "A flying machine with wings and rudders"?

(3) "A flying machine with wings"?

(4) "A flying machine"?

Disclosure theory provides no principle to select among these competing levels of abstraction. Clearly some limit is needed, since otherwise the Wright brothers would claim all flying machines, including a future antigravity spaceship. As a historical matter, courts limited the Wright brothers to airplanes using wings and rudders, and did not give them all flying machines. (8) But any limit (except to the literal embodiment) is arbitrary from the perspective of disclosure theory. After all, the Wrights did not teach how to make every airplane using wings and rudders--they taught nothing except a single wooden glider. And if they could cover some undisclosed flying machines, why not all undisclosed flying machines?

In more practical terms, this incoherence is reflected by the conflict between two lines of case law. (9) One line says that patent scope covers only replication of the embodiments disclosed in the specification. (10) Another line says that patentees must not be confined to the embodiments disclosed in the specification. (11) The result of this conflict is "doctrinal chaos" where outcomes are impossible to predict. (12)

Although the problem has been described in previous work by myself and others, (13) no practical solution has emerged. one proposed solution has been for courts to calibrate patent scope using economic balancing. (14) A patent covering only a wooden glider would eviscerate incentives, while a patent covering all flying machines would create excessive monopoly costs, and so a court should choose something in between. The problem with this economic balancing approach is that courts lack the capability to directly measure such economic variables as social

benefits and monopoly costs with any degree of precision. (15) The purpose of this Article is to provide a more practical solution.

The starting point is to note that, conceptually speaking, patent scope is directed to an idea. (16) Although patent claims are often phrased in terms of an embodiment--e.g., "a flying machine"--the legal effect of the claim is to cover the underlying idea. A claim to "a flying machine" is really a claim to the idea of artificial flight, since it will cover every flying machine. I mean this as a linguistic point about the mechanics of the patent document, not about whether such a patent would issue. That is, a claim to "a flying machine" might not be granted, but if a patentee submits such a claim to the patent office, every patent lawyer would understand the claim to cover all flying machines notwithstanding the use of the word "a." This a-to-every linguistic trick is basic patent law. (17)

Although this linguistic point is simple, once stated explicitly, it becomes clear that patent law's fundamental quid pro quo is comparing apples and oranges. While patent protection is enforced in terms of an idea against later accused infringers downstream, the patentee's contribution over prior inventors is not assessed in those terms. Patent law says that the airplane was novel because the Wright brothers were the first to make one airplane embodiment. (18) Patent law does not ask whether the Wright brothers invented the idea of airplanes: the idea of airplanes has existed since ancient times.

But in order to be fairly credited as the inventor of the idea and obtain a monopoly covering all future implementations of that idea, a patentee should be required to be the first to think of the idea in addition to merely creating one embodiment of it. My proposal is thus that a patentee should obtain a monopoly covering the new and non-obvious idea that he is the first to articulate and that is reflected in his embodiment. Importantly, this does not mean that the Wright brothers would receive no patent because the idea of airplanes had already been articulated. (19) Rather, the Wrights would receive a patent covering the idea at a lower level of abstraction. The Wrights did not invent the idea of airplanes, because that idea had already been articulated. (20) But the Wrights did invent the idea of using rudders to control flight (or, more technically, linking yaw to roll via a rudder). (21) Not fortuitously, as a matter of historical fact they received patent scope closely corresponding to that contribution. (22)

The beauty of my proposal is that it provides a coherent principle to select among levels of abstraction. It is important to note that any new and non-obvious (23) embodiment will always contain a new idea among its many levels of abstraction--the new idea is what makes the embodiment non-obvious. My claim is simply that this novel insight, rather than the embodiment itself, represents the patentee's contribution to downstream improvers and should define patent scope. A later aluminum airplane using rudders for flight control should still infringe the patent, even though the Wright brothers could not make aluminum airplanes, because the aluminum airplane is still using the Wright brothers' idea of rudders for flight control. But in order to claim rights to the idea, the patentee must really invent the idea as an idea (i.e., be the first to articulate it).

The argument that patent scope should be defined by the idea that was actually invented by the patentee is primarily a moral claim. But it can also be cast in economic terms so that it fits comfortably within the dominant economics-based paradigm of patent law. (24) Economically speaking, the role of patent protection is to protect against misappropriation of information goods that, in the absence of protection, would be too easily copied. (25) Because the misappropriation risk applies to the idea and not the embodiment, the patentee's insight represents a much better way to conceptualize the invention and what patent law should protect. But unless the idea was original to the patentee, there is nothing to misappropriate.

This Article proceeds in four Parts. Part I provides a brief background on novelty doctrine and the claiming of inventions. A key point here is the dichotomy between ideas and embodiments, where the patentee contributes a novel embodiment, but claims an idea. Part II then describes how this dichotomy between ideas and embodiments renders conventional disclosure theory incoherent. Part III describes my proposed alternative, where the focus is on the patentee's novel insight rather than the embodiment being disclosed. Part IV then considers some potential objections regarding my proposal, and discusses why these concerns are misplaced. A conclusion follows.

1. THE INVENTION AS IDEA AND EMBODIMENT

   This Part provides some doctrinal background to patents. In particular, it is important to note two fundamental features of current patent law. First, novelty is assessed on an embodiment-centric basis, in terms of asking whether prior inventors had made the invention before. Second, patent scope is defined in terms of an idea. These doctrinal features are explained in more detail below.

   1. The Embodiment-Centric Assessment of Novelty

      As a matter of first principles, the contribution of an inventor to society can be conceptualized in two distinct ways: in economic terms as the amount of social welfare benefit that is created by investments in research, and in more philosophical terms as the "invention," i.e., the product of that research. Although patent law is...