Patent at Your Own Risk: Linguistic Fences and Abbott Labs v. Sandoz

John Cordani

Posted in , , , , , , , , , , , , ,

Patenting is the procedure by which novel scientific inventions become protected by an intellectual property right.  The United States Constitution provides the foundation for patent law and vests the power and duty of its implementation in Congress. 1  Congress fulfilled its constitutional duty by granting inventors an intellectual property right in return for public disclosure of their innovation.  Today, an inventor secures this right by obtaining an issued claim from the United States Patent and Trademark Office.  Among other requirements, an inventor must “particularly point out and distinctly [claim] the subject matter which the [inventor] regards as his invention”2 to obtain a claim from the Patent Office.  One can understand this statutorily mandated feature of claims by analogy to a fence.  Instead of poles and wire, an inventor utilizes language to fence off the “metes and bounds” of what the inventor takes to be his or her invention.  This linguistic fence allows the public to know what intellectual “land” they are not allowed to use.  As a result of modern patent statutes, no one may obtain a patent claim without creating this linguistic fence.

One particular method of claiming is known as “product-by-process.”  Here, the inventor will define his or her invention in terms of the process by which it is made.  For example, a traditional product claim would refer directly to the composition of matter: “I claim the substance comprising gaseous water.”  Conversely, the equivalent product-by-process claim would read: “I claim the substance obtainable by the process comprising the steps of heating pure water to 100 degrees Celsius  at 1 atmosphere of pressure.”  Both of these claims refer to precisely the same substance existing in the physical world (water vapor), but product-by-process claims “fence-off” their physical referents in an obviously different and more indirect manner.

The recent Federal Circuit en banc decision, Abbott Laboratories v. Sandoz, Inc.,3 drastically limited the interpretation of product-by-process claims for the purpose of determining patent infringement.4  The court held that “the recited process steps limit the product-by-process [claim]” during its interpretation for determining patent infringement.5  Using the product-by-process claim example above, the court held that gaseous water resulting from heating water to 85 degrees Celsius at 0.5 atmospheres of pressure would not infringe that product-by-process claim.  Gaseous water made at 85 degrees Celsius, however, would infringe the traditional product claim because the product is still gaseous water.  Although the majority in the case nominally bases its decision on 35 U.S.C. § 112 ¶ 2’s requirement that claims “particularly point out and distinctly” claim their subject matter,6 the court also relies on case law, policy, and the all-elements rule that the Supreme Court described in Warner-Jeninson.7

The Abbott Labs court incorrectly analyzed product-by-process claims for three reasons. First, a scientifically proper theory of language, which Gottlob Frege first advocated, proves that the process steps a product-by-process claim enumerates are not claim elements for the purposes of Warner-Jenkinson.  Second, the Federal Circuit’s reliance on 35 U.S.C. § 112 ¶ 2 is misplaced because product-by-process claims clearly succeed in particularly pointing out and distinctly claiming the subject matter of the invention.  Third, the policy concerns on which  the Abbott Labs court relied can be rebutted by considering the scientific realities and civil procedure of patent infringement suits.

A. Gottlob Frege’s Analysis of Product-by-Process Claiming

In evaluating product-by-process claiming, I rely on the long-established tradition in analytical philosophy, which Gottlob Frege founded, distinguishing an expression’s linguistically expressed “sense” from the object in the world that languages identify (its tangible “referent”).  Frege states, “[i]t is natural, now, to think of there being connected with a sign (name, combination of words, letter), besides that to which [in the physical world] the sign refers, which may be called the referent of the sign, also what I would like to call the sense of the sign, wherein the mode of presentation is contained.”8  As an example, consider “water” and “H2O.”  Both have the same referent but use different senses, different modes of expressing the designation of the substance.  “Water” has the more colloquial sense, and “H2O” is generally a scientific sense.  Thus, for a given sense there is a defined referent, but for a given referent there can be multiple senses. 9

The various methods of claiming differ only in sense, not in reference.  Consider, first, the easy example of “I claim the substance comprising gaseous water” compared to “I claim the substance comprising gaseous H2O.”  These claims refer to precisely the same substance in the physical world, and therefore have the same referent.  Under Frege’s theory, these claims differ only in the sense of expression.  Importantly, however, unless the reader of the two claims possesses rudimentary chemical knowledge, the fact that the two claims refer to the same physical substance might not be obvious.10  Likewise, the only difference between the preceding two claims and “I claim the substance obtainable by the process comprising the steps of heating pure water to 100 degrees Celsisu at 1 atmosphere of pressure” is one of sense.  Furthermore, the experimentation required to determine that the product-by-process substance is the same as gaseous water or gaseous H2O is the same as the experimentation required to determine that water is H2O.  An experimenter must obtain the referenced objects and compare them by using some form of analysis.  Thus, the only difference between product-by-process claims and conventional product claims is one of sense.

Given this theory of language, the result obtained in Abbott Labs leads to absurd results. The only difference between “gaseous water” and “gaseous H2O” is one of Fregean sense.  If a difference in sense can constitute a difference in claim elements, then, under the all-elements test, someone using what he calls gaseous H2O would not infringe a valid patent that claims “gaseous water.”11  This result is absolutely absurd; it would be like saying that letters comprising the word “gaseous water” constitute elements of the claim.  Product-by-process claims, however, also differ from conventional claims only in terms of Fregean sense.  Thus, the absurdity of thinking that gaseous water and gaseous H2O have different elements logically must be imported into cases of product-by-process claims. Warner-Jenkinson is irrelevant to the holding of Abbott because, whatever a claim “element” is, product claims and product-by-process claims have the same elements if they refer to the same physical entity.

B. The Definiteness of Product-by-Process Claiming

Section 112 ¶ 2 requires applicants to “particularly [point] out and distinctly [claim] the subject matter which the applicant regards as his invention.”12  In Abbott Labs, the court held that if product-by-process claims were not limited to their process steps then the claim would not particularly point out and distinctly claim the invention.13  However, product-by-process claims, assuming the inventor adequately defines process steps, perfectly demarcate the metes and bounds of what the claim includes and what the claim excludes.  A product-by-process claim should be understood as identifying whatever exists in the world as a result of performing the process steps.  Assuming the described process controls for all scientifically relevant variables in the process itself, a product-by-process claim fences off precisely one substance and excludes everything that does not result from the process.

Before Abbott Labs, courts had always recognized that precisely worded product-by-process claims are definite for any construction. The Court of Customs and Patent Appeals explained in In re Hughes:

We cannot agree with the solicitor that defining a product in terms of process makes the language of the claims imprecise or indefinite. Their scope, if anything, is more definite in reciting a novel product made by a specific process, assuming, of course, that the process is clearly defined. It does not create a definiteness problem under § 112.14

The same court made similar holdings in In re Steppan15 and In re Brown16.  The Abbott Labs court failed to recognize that in 35 U.S.C. § 112 ¶ 2, “Congress has placed no limitations on how an applicant claims his invention, so long as the specification concludes with claims which particularly point out and distinctly claim that invention.”17

Patent claims are analogous to land deeds in many ways with respect to the requirement to particularly point out and distinctly claim the real or intellectual property.  In real estate, however, two valid manners by which one may claim property exist: the Jeffersoninan grid system or metes and bounds.  In the Jeffersonian grid system, a plot of land is given a name much like the vapor rising out of hot water is given the name “steam.”  On the other hand, a typical metes and bounds description might start: “Begin at the southernmost end of the old rock wall. Head due north to the covered bridge. Head east 100 feet to the stream.”18  Metes and bounds is the real estate analog of product-by-process.  Metes and bounds is considered a definite and valid method of claiming in real estate, and no one would argue that the metes and bounds deed is limited to the process steps enumerated therein.  Intellectual property in a product-by-process claim need not be treated any differently.

C. Public Policy, Civil Procedure, and Scientific Realities

The Abbott Labs court worries that if product-by-process claims are not limited to their processes, “how would the courts ascertain that the alleged infringer’s compound is really the same as the patented compound?”19  However, in our adversary system, courts will never have to worry about this concern.  A patent holder must prove patent infringement by a preponderance of the evidence.20  The same worry would be present for a court adjudicating trespassing rights with a metes and bounds deed.  The plaintiff will prove that the trespasser encroached onto the plot of land that the deed demarcates by performing the steps enumerated in the deed and comparing the result with the trespasser’s position.  Likewise, a patent infringement plaintiff will need to analyze the defendant’s product and compare it with the product that results from performing the  steps in the product-by-process claim.

Secondly, the realities of physical science require the availability of robust product-by-process claims.  Science abounds with examples of products that cannot be precisely described by name.  A first example is a physical phenomenon known as “glow discharge,” the plasma produced by running electrical currents through low-pressure air and one commonly sees in neon signs.  The characteristics of the produced plasma depend on the gas (which is the matter that transitions into the plasma), the gas pressure, the geometry of the electrodes, and the electric voltage.21  Thus, an adequate description of the precise nature of the glow discharge depends on the process that produced it.  Case law also exhibits a myriad of examples of product-by-process claims being used out of necessity in a wide variety of scientific fields other than glow discharge technology.  Three such examples are (1) complex organic extracts,22 (2) activated catalysts,23 and (3) products created with heat or pressure treatments.24  If science and language are to blame for the necessity of product-by-process claims, courts should not force inventors to accept inferior patent rights in return for the same useful and enabled disclosure provided by all patentees.

In conclusion, patents represent a busy intersection of science, language, and the law, and product-by-process claiming is the inevitable collision.  The Federal Circuit has limited the protection of these claims to their process steps.  In doing so, I believe the courts have aggrieved inventors in many scientific fields to the eventual detriment of the public, whom patent law was created to serve.  I have shown that Abbott Labs’s holding was not required by Warner-Jenkinson, the patent statutes, public policy, or judicial efficiency.  Yet, the holding in Abbott Labs is reminiscent of Samuel Morse’s famous criticism of patent law: “It is not the way to encourage the Arts, . . . to drive the Artists into exile or to the insane hospital or to the grave.”25


John Cordani is a 2011 J.D. Candidate at Cornell Law School.

This Legal Workshop Editorial is based on Mr. Cordani’s Note: John Cordani, Note, Patent at Your Own Risk: Linguistic Fences and Abbot Labs v. Sandoz, 95 CORNELL L. REV. __ (forthcoming 2010).

Copyright © 2010 Cornell Law Review.

  1. U.S. Const. art. I, § 8, cl. 8 (granting Congress the power to pass laws “{t}o promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries”).
  2. 35 U.S.C. § 112 ¶ 2 (2006).
  3. 566 F.3d 1282 (Fed. Cir. 2009) (en banc).
  4. In doing so, the en banc Federal Circuit was resolving an intra-circuit split on the issue. See Scripps Clinic & Research Foundation v. Genentech, Inc. 927 F.2d 1565 (Fed. Cir. 1991) (holding that product-by-process claims are not limited to their process); Atl. Thermoplastics Co. v. Faytex Corp., 970 F.2d 834 (Fed. Cir. 1992) (holding that product-by-process claims are limited to the recited process).
  5. Id. at 1295.
  6. Id. at 1295 (citing 35 U.S.C. § 112).
  7. Warner-Jenkinson Co. v. Hilton Davis Chemical Co., 520 U.S. 17 (1997).  In relying on the all-elements rule, the Abbott Labs court reasoned that courts must give effect to the language employed by the patentee, including the process steps of a product-by-process claim. See Abbott Laboratories, 566 F.3d at 1293.
  8. Gottlob Frege, On Sense and Reference, 57, 3 THE PHIL. REV. 209, 210 (1948).
  9. Id. at 211.
  10. See Paul Needham, The Discovery that Water Is H2O, 16, 3 Int’ Stud. in the Phil. of Sci., 205, 211 (2002) (noting that the consensus that water is H2O did not come about until the 1861 Karlsruhe conference and required the skills of some of history’s most notable chemists and physicists including Cannizzaro, Lussac, Avogadro, and Berzelius).
  11. See Warner-Jenkinson, 520 U.S. at 29 (holding that every element in a claim must be found in an allegedly infringing product to establish patent infringement liability).
  12. 35 U.S.C. § 112 ¶ 2 (2006).
  13. Abbott Laboratories, 566 F.3d at 1294–95.
  14. 496 F.2d 1216, 1218 (C.C.P.A. 1974) (emphasis added).
  15. 349 F.2d 1013, 1019 (C.C.P.A. 1967).
  16. 459 F.2d 531, 535 (C.C.P.A. 1972).
  17. In re Pilkington, 411 F.2d 1345, 1349–50 (C.C.P.A. 1969).
  18. William Hubbard, Note, Communicating Entitlements: Property and the Internet, 22 YALE L. & POL’Y REV. 401, 406 (2004).
  19. Abbott Laboratories, 566 F.3d at 1294.
  20. SmithKline Diagnostics, Inc. v. Helena Laboratories Corp., 859 F.2d 878, 889 (Fed. Cir. 1988).
  21. See R. A. Dugdale, The Application of the Glow Discharge to Material Processing, 1 JOURNAL OF MATERIALS SCIENCE 160, 161 (1966).
  22. See In re Fessmann, 489 F.2d 742, 743 (C.C.P.A. 1974) (stating that the applicants called their product “liquid smoke”).
  23. See In re Brown, 459 F.2d 531, 532 (C.C.P.A. 1972).
  24. See In re Pilkington, 411 F.2d 1345, 1346 (C.C.P.A. 1969) (describing the finish as a “fire finish”).

Post a Comment (all fields are required)

You must be logged in to post a comment.