Differential Formalism in Claiming Intellectual Property: A Response to Fromer

Henry Smith - Harvard Law School

Posted in ,

This article is a response to an earlier posted piece by Jeanne C. Fromer: Claiming Intellectual Property.

Claiming intellectual property is an act of communication, but as with all communication not everything can be spelled out with exactitude all the time—far from it. By drawing out an additional distinction between modes of claiming, Jeanne Fromer advances our understanding of claiming in intellectual property.1 To the familiar distinction between central and peripheral claiming she adds the contrasting modes of claiming by exemplar and claiming by characteristic. She very usefully applies these two axes not just to patent law, where the issue of central versus peripheral claiming originates, but also to copyright. The possibility of mixing and combining the four different modes of claiming—peripheral by characteristic, peripheral by exemplar, central by characteristic, and central by exemplar—in different areas opens up not only a new lens for looking at the hybrid nature of existing patent and copyright law but also suggests a framework for thinking about improvements to claiming in these two, and other, areas of intellectual property. In this Response I argue that although these two axes can be regarded as distinct, they do share something in common—a differential reliance on formalism versus contextual interpretation—and that the degree of formalism is relevant to the conclusions we can draw from Fromer’s framework.

In peripheral claiming, which is familiar as the main method in current patent law, the would-be patentee claims the invention by describing its outer boundaries, in a process that is sometimes analogized to a metes and bounds description in the law of real property. If something is infringing it must fall within the claim; otherwise there is no infringement. By contrast, in central claiming, which was more characteristic of early nineteenth-century patent law and which lives on in the doctrine of equivalents and means-plus-function claims, the claim sets forth some central or paradigmatic members of the claimed set and then one relies on some notion of similarity to figure out membership of the full set. In Fromer’s fork example, there are four ways to claim.2 In peripheral claiming one can either describe the set of forks by characteristic or by pointing out (more on this later). In central claiming, one can either set out a description of a central case of a fork or point to a central exemplar of one, and rely on a notion of relevant similarity to extend the set outward to the full set of claimed forks. This latter determination of similarity in central claiming is typically left for ex post decisionmaking, which allows for more flexibility but at the cost (so it is conventionally thought) of less ex ante certainty.

To this distinction between central and peripheral claiming familiar from patent law, Fromer adds claiming by characteristic versus claiming by exemplar. Patent law typically employs claiming by characteristic, paired with peripheral claiming. This leads to the familiar highly regimented style of claiming in which a single sentence is used to delineate the outer bound of the invention space claimed by the inventor. By contrast, claiming by exemplar involves setting out examples. This fits naturally with central claiming, as in copyright, in which central members of the set are set forth and similarity is judged ex post. But peripheral claiming by exemplar is also possible. Fromer assumes that it would involve listing the members of the claimed set, as is done in so-called Markush claims.4 Generally, as we will see, it will not be feasible to literally “point to” every fork that is being claimed (that is, purely by exemplar), and there must be a procedure for inferring from what is pointed out to the entire claimed set. The feasibility of such a procedure and the form it could take would depend in turn on who the relevant interpretive community is; the general public, inventors, and patent attorneys will all bring different background knowledge to bear. 5

Even more generally, what Fromer’s framework (implicitly) raises is the question of what types of information should be made explicit and which are best left implicit under different conditions for claiming. Different audiences, different informational resources, and different dynamics of innovation and age of industries can all influence how easy or difficult it is to leave information implicit and conversely how much reliance on formalism is needed.

For many purposes, a simple definition of formalism is helpful. An instance of communication is formal to the extent that its interpretation is invariant to context.6 The language of first-order logic, for example, is formal because its semantic rules are highly worked out and thus do not rely much on context, compared with more informal everyday mathematical notation. Natural languages (of which English and Chinese are examples) on this definition are less formal than the language of mathematics, but language in different situations can be more or less formal. Legal language tends to be formal because drafters are wary of reliance on context. For example, legal language in contracts and statutes is notorious for the repetition of noun phrases (as in “the party of the first part agrees with the party of the second part that the party of the first part,” and so on) as opposed to the use of pronouns which do require context for their interpretation. The more that the speaker and hearer are socially close, can rely on common knowledge, and are trying to cooperate, the less formal language needs to be. Thus, saying, “It’s cold in here” is a shorter way of saying, “Because you are closer to the window, please close the window.” But the looser formulation that relies on pragmatic inference relies on a lot of background knowledge and coincidence of communicative interest in order to succeed. In the absence of these factors, the more formal approach would be warranted—possibly including a description of what a window is and how it works. Sociolinguistic studies have shown that individual speakers employ more informal variants with socially closer audiences.7

No useful system of communication is perfectly formal in the sense of being absolutely invariant to context, but some communications and systems of communication are more invariant to context than others. I have argued elsewhere that formalism is a matter of degree, and that differential formalism depends largely on the nature of the audience for the communication. Communication is subject to an informational tradeoff. At the same cost one can communicate intensively with a small, socially close, knowledgeable audience or one can communicate in a more stripped-down, formal, and explicit way with a larger, more heterogeneous, and more anonymous audience.8 Formalism, which is a matter of degree, should correlate with the breadth and heterogeneity of the audience. The basic distinction between the use of language in contracts and property reflects this tradeoff: communicating an in rem right to “the world” requires a more formal and more regimented style of language and forms, including the standardization supplied by the numerus clausus.10—but in the examples given here, there is a lot of implicit context-based reasoning going on under the table, which makes claiming by exemplar fall towards the informal end of the spectrum.

To the extent that the costs and benefits of formality and contextuality along different dimensions go together, we might expect to find the formal and informal poles of these spectrums to line up: peripheral claiming will tend to be done by characteristic, and central claiming will tend to rely on exemplars. And the broad contours of patent and copyright do line up this way.11 In this sense the two axes share a lot more in common than Fromer lets on. But Fromer is right to point out that each area uses a hybrid of the types of claiming, and we should ask what factors at this more detailed level push toward or away from the different modes of claiming.

What I suggest here is that the formality versus contextuality of claims—their differential formalism—will emerge as an important lens when trying to apply Fromer’s categories to the problems of claiming in patent, copyright, and other areas of intellectual property. Formalism implicates the costs and benefits of delineation depending on the audience’s background knowledge, the degree of cooperativeness of the communication, and the nature of the informational resource.

The audience for the delineation of rights matters when it comes to the costs and benefits of delineation, as Fromer recognizes. In general, patent law is directed more to expert audiences and much, though not all, of copyright can implicate the activities of nonexperts.12 Thus, more expert knowledge can generally be presumed in patent law. Delineation in patent law can use technical terms and rely on knowledge in the field that copyright law cannot. But which area of law can rely more on implicit (less formal) means of communication? The role of audiences suggests that different types of implicitness can be employed in patent and copyright. Patent attorneys or agents (including those working for industrial competitors) versed in the relevant art can be expected to fill in certain kinds of information, and potential copyright violators different kinds of information. Possibly the latter kind of background information is closer to everyday knowledge. There are areas of copyright that do involve expertise of an artistic sort, but with a few exceptions expert knowledge (other than possibly legal knowledge) does not seem to feature as much in filling in informal delineation in copyright law as in patent law. Thus the nature of the audience points to the possibility of leaving different information implicit in the two areas. Correspondingly, patent and copyright rely differently on formalism when this reliance on background information is not reasonable.

Claiming intellectual property is an example of noncooperative communication which generally cuts in favor of formal communication. In a cooperative setting, one can assume that the audience will fill in information and draw inferences in a charitable fashion. At the center of the study of the “logic of conversation” is Grice’s Cooperative Principle: “Make your conversational contribution such as is required, at the stage at which it occurs, by the accepted purpose or direction of the talk exchange in which you are engaged.”
13 That is, the interpreter will interpret in accordance with the assumption that the speaker is trying to be helpful. Thus, in the Gricean example where Person A says “I’m out of gas” and Person B says there is a gas station around the corner, we and Person A can safely conclude that it is open, because otherwise Person B would be violating the Cooperative Principle and the maxims of conversation.

But legal interpretation is not like this. Disingenuousness of speaker and audience is almost to be presumed. The “logic of legal conversation” diverges from the ordinary conversation studied by Grice. Reliance on central claiming and claiming by exemplar requires context to be filled in and inferences to be drawn. In an adversarial context, it is difficult to converge on the relevant context and these pragmatic inferences, because each party—patentee and alleged infringer, say—has different incentives as to how to fill these items. The noncooperative nature of much of legal communication probably accounts for some of its formality. And legal language in general often relies on a more rigid analogue of pragmatic-style inference. The linguistic study of pragmatics focuses on language meaning in context, as distinct from the study of more formal meaning in semantics. In legal language a lot of what would be loose pragmatics is regimented into the semantics.

Whether to rely on formal or context-dependent modes of communication should depend on the nature of the resource. Do certain kinds of information require greater formality in delineation? One difference between patent and copyright is in how “close” a claimed contribution is to what another might wish to use or develop. In copyright, delineation can be left somewhat unclear when two copyrighted works are not likely to be close. Thus, copyright can allow a defense of independent creation without as much conflict as patent law would have if it provided the same defense (which it does not).14 Apart from low-authorship works like maps and charts, it is unlikely that two works will be identical without one creator having had access to the other work.15 But, as Fromer points out, derivative works present a special situation in copyright, where the notion of fortuitous resemblance is a greater possibility.16 I would say that the right to prepare derivative works is different from the rest of copyright and, for better or worse, comes closer to patent law or even the law of ideas.17here is a point in this series of abstractions where they are no longer protected, since otherwise the playwright could prevent the use of his ‘ideas,’ to which, apart from their expression, his property is never extended. Nobody has ever been able to fix that boundary, and nobody ever can.”) (citations omitted).] In patent law, particularly in well-established areas of technology, there is a high likelihood of two claims being close or even contiguous in invention space. (Indeed claims can overlap.) Take for example the field of tennis rackets. Many rackets have different “splayed” patterns of strings, with the ends of the strings attached alternately above and below the central plane of the racket frame at different distances from each other, such that in a patent claim, infringement can turn on whether “varies between” requires three or more distances between the strings.18
In a less crowded field, the potential ambiguity in “varies between” would not matter so much.

Finally, the question of the best degree of formalism in claiming potentially interacts with the psychological evidence that Fromer adduces.19 This evidence suggests that people use characteristics when learning broad categories and early on in the process of learning categories of all kinds and that they use exemplars for narrower categories and later in the general learning process. This might suggest that claiming by characteristic would be more suited to exclusion strategies and exemplars would be used more in governance strategies, when picking out special uses. As Fromer acknowledges, this evidence is to be taken cautiously, because the conditions in the experimental studies may differ relevantly from the situation of those faced with intellectual property claims. It would be a worthwhile subject for further study whether leaving information implicit (as in central claiming and claiming by exemplar) inhibits learning more or less depending on the level of expertise of the audience and on the nature of the category to be learned. Whether the two kinds of implicitness show the same or different learning effects would be relevant to the question of how formal claiming should be with different audiences in view.

In sum, Fromer’s addition of the distinction between claiming by characteristic and claiming by exemplar to the more familiar central versus peripheral claiming dichotomy is a real advance, which provides a lens for more detailed evaluation of hybrid claiming methods and their variation within areas of intellectual property law. But like central versus peripheral claiming, the exemplar-characteristic claiming distinction also crucially relies on a spectrum of how much information is spelled out formally and how much is left implicitly to be filled in by reference to context. This differential formalism will be important to further development of Fromer’s very fruitful framework.


Copyright © 2010 University of Chicago Law Review.

Henry E. Smith is a Professor of Law at Harvard Law School.
For helpful comments I would like to thank John Golden and Oskar Liivak. All errors are mine.

  1. Jeanne C. Fromer, Claiming Intellectual Property, 76 U Chi L Rev 719, 727 (2009).
  2. See id at 728 table 1.
  3. In re Schechter, 205 F2d 185, 189 (CCPA 1953); Fromer, 76 U Chi L Rev at 741 (cited in note 1).
  4. One might imagine claiming by exemplar through a sequence such as one-pronged fork, two-pronged fork, three-pronged fork, and so on. This requires the interpreter to understand the nature of the sequence being invoked.
  5. See John M. Golden, Construing Patent Claims According to Their “Interpretive Community”: A Call for an Attorney-Plus-Artisan Perspective, 21 Harv J L & Tech 321, 330–31 (2008) (discussing background knowledge assumed for hypothetical interpretive audiences).The Federal Circuit is quite split on the role in claim interpretation of the person having ordinary skill in the art. Phillips v. AWH Corp., 415 F3d 1303, 1303 (Fed Cir 2005) (en banc) (dissents by Judges Lourie and Mayer).
  6. See, for example, Henry E. Smith, The Language of Property: Form, Context, and Audience, 55 Stan L Rev 1105, 1112 (2003); Francis Heylighen, Advantages and Limitations of Formal Expression, 4 Found Sci 25, 49–53 (1999).
  7. For a summary, see Smith, 55 Stan L Rev at 1133–39 (cited in note 5).
  8. See Smith, 55 Stan L Rev at 1125–57 (cited in note 5).
  9. Henry E. Smith, Modularity in Contracts: Boilerplate and Information Flow, 104 Mich L Rev 1175, 1196–97, 1210 (2006); Smith, 55 Stan L Rev at 1157–61 (cited in note 5); Thomas W. Merrill and Henry E. Smith, Optimal Standardization in the Law of Property: The Numerus Clausus Principle, 110 Yale L J 1, 24–42 (2000).
  10. See generally, for example, Sun-Joo Shin, The Logical Status of Diagrams (Cambridge 1994).
  11. Henry E. Smith, Intellectual Property as Property: Delineating Entitlements in Information, 116 Yale L J 1742, 1814–17 (2007).
  12. Smith, 55 Stan L Rev at 1173–77 (cited in note 5). See also Golden, 21 Harv J L & Tech at 340 (cited in note 4); Clarisa Long, Information Costs in Patent and Copyright, 90 Va L Rev 465, 487–89 (2004).
  13. Paul Grice, Logic and Conversation, in Paul Grice, ed, Studies in the Ways of Words 22, 26 (Harvard 1989).
  14. Norman Siebrasse, A Property Rights Theory of the Limits of Copyright, 51 U Toronto L J 1, 38–42 (2001). See also Mark A. Lemley, Should Patent Infringement Require Proof of Copying?, 105 Mich L Rev 1525, 1531–32 (2007) (pointing out that an independent invention defense would lessen the marketability of patent rights); Samson Vermont, Independent Invention as a Defense to Patent Infringement, 105 Mich L Rev 475, 493–500 (2006) (arguing for an independent invention defense).
  15. See David D. Friedman, Law’s Order: What Economics Has to Do with Law and Why It Matters 130 (Princeton 2000).
  16. Fromer, 76 U Chi L Rev at 745 (cited in note 1).
  17. Consider Nichols v Universal Pictures Corp, 45 F2d 119, 121 (2d Cir 1930) (Hand) (“Upon any work, and especially upon a play, a great number of patterns of increasing generality will fit equally well, as more and more of the incident is left out. . . . [T
  18. Athletic Alternatives, Inc v Prince Manufacturing, Inc, 73 F3d 1573, 1573–77 (Fed Cir 1996).
  19. Fromer, 76 U Chi L Rev at 763–77 (cited in note 1).

Post a Comment (all fields are required)

You must be logged in to post a comment.