Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Proprietary Rights and the Norms of Science in Biotechnology Research

Proprietary Rights and the Norms of Science in Biotechnology Research The Yale Law Journal Volume 97, Number 2, December 1987 Articles Norms of Proprietary Rights and the Science in Biotechnology Research Rebecca S. Eisenberg As basic research in biotechnology yields increasing commercial applica- tions, scientists and their research sponsors have become more eager to protect the commercial value of research discoveries through intellectual property law. Some scientists fear that these commercial incentives will weaken or even undermine the norms that have traditionally governed scientific research. In this Article, Professor Eisenberg examines the interaction of proprietary rights in inventions with these traditional scientific norms. Trade secrecy, she ar- research discoveries gues, is an undesirable strategy for protection of basic because it impedes dissemination of new knowledge to the scientific community. She finds that patent law is in many respects more congruent with scientific norms than trade secrecy because it is premised on disclosure rather than Eisenberg demonstrates, however, that the fit between the secrecy. Professor and incentives of the scientific community is patent system and the norms hardly perfect. Patent law may operate to delay the dissemination of knowl- edge to other researchers. Moreover, by granting rights to exclude others from using patented inventions for a term of years, the patent system threatens the interest of the scientific community in the free use and extension of new discov- eries. Professor Eisenberg concludes that greater sensitivity to the impact of © 1987 Rebecca S. Eisenberg of Law, University of Michigan Law School. I wish to thank Marty Add- t Associate Professor Gage, Judah Garber, Paul Goldstein, Don Herzog, Dick man, Ed Cooper, Mel Eisenberg, Julie Kaba, Jim Krier, Rick Lempert, Jessica Litman, Fred Schauer, Jonathan Weinberg, Peter Westen, and Virginia Wise for helpful comments on earlier drafts of this Article. The Yale Law Journal [Vol. 97: 177 patent law doctrine on scientific norms will help to reconcile the norms and incentives of these two systems. INTRODUCTION The commercial potential of recent advances in biotechnology has sub- stantially increased private investment in basic research in the biomedical sciences.' Some academic scientists have viewed this influx of private funding as a mixed blessing. Although it comes at a time when the con- tinuity of public funding for biomedical research is uncertain, it raises concerns about the potential impact of commercial incentives to protect intellectual property on traditions of open communication and free flow of information within the scientific community.' Indeed, some scientists have 1. See S. OLSON, BIOTECHNOLOGY: AN INDUSTRY COMES OF AGE (1986); Blumenthal, Gluck, Louis & Wise, Industrial Support of University Research in Biotechnology, 231 SCIENCE 242 (1986). Between World War 11 and the 1970s, basic research in the biomedical sciences was dominated by public funding. See generally M. KENNEY, BIOTECHNOLOGY: THE UNIVERSITY-INDUSTRIAL COm- PLEX 13-18 (1986); 1 B. SMITH & J. KARLESKY, THE STATE OF ACADEMIC SCIENCE: THE UNIVER- SITIES IN THE NATION'S RESEARCH EFFORT 11-48 (1977), and sources cited therein. I use the term "basic research" to refer to "pure" research directed solely toward expanding human knowledge, as opposed to "applied" research directed toward solving practical problems. See infra notes 38-47 and accompanying text. I shall argue below that, whatever validity this dichotomy may have in other contexts, it is difficult to maintain in the context of contemporary biotechnology re- search. See infra notes 93-109 and accompanying text. 2. Between 1980 and 1984, industrial funding for research and development at universities and colleges increased 93%, from $237,025,000 in 1980 to $457,227,000 in 1984, while federal funding increased only 31%, from $4,096,029,000 to $5,386,578,000. NATIONAL SCIENCE FOUNDATION, NA- TIONAL PATTERNS OF SCIENCE AND TECHNOLOGY RESOURCES 65 (1986) (Table 51). Adjusted for inflation, the total investment in research and development at universities and colleges rose only 4% between 1980 and 1983. This extremely modest rise reflects cutbacks in federal funding, particularly in the health area. Id. at 12-13. Since 1980, federal support has increasingly shifted toward defense-related fields at the expense of other categories of research and development. While the federal government's share of total national research and development outlays remained at a relatively constant 46-47% between 1980 and 1986, federal defense-related expenditures rose from 22% of this total in 1980 to an estimated 30% in 1986, while federal civilian-related expenditures fell from 18% in 1980 to an estimated 10% in 1986. Id. at 41 (Table 13). During this period, federal research and development funding for national defense increased approximately 183%, from $14,946,000,000 in 1980 to an estimated $42,360,000,000 in 1986, while federal funding for health-related research and development increased only 38%, from $3,694,000,000 in 1980 to an estimated $5,108,000,000 in 1986. Id. at 45 (Table 22). Biomedical research would have fared worse under budgets proposed by the Reagan administration, but Congress has repeatedly approved funding increases significantly greater than the administration's recommen- dations. See Culliton, Pressure to Cut the Deficit Creates Uncertainty for Biomedical Research, 232 SCIENCE 564 (1986); Norman, Science Escapes Brunt of Budget Ax, 231 SCIENCE 785, 787 (1986). 3. See generally Commercialization of Academic Biomedical Research: Hearings Before the Sub- comm. on Investigations and Oversight and the Subcomm. on Science, Research and Technology of the House Comm. on Science and Technology, 97th Cong., 1st Sess. (1981) [hereinafter Commerciali- zation Hearings]; D. NELKIN, SCIENCE AS INTELLECTUAL PROPERTY (1984); TWENTIETH CEN- TURY FUND, INC., THE SCIENCE BUSINESS (1984) [hereinafter THE SCIENCE BUSINESS]; Grobstein, Biotechnology and Open University Science, 10 SCI. TECH. & HUM. VALUES, Spring 1985, at 55. Concerns about the impact of industrial funding on biomedical research are by no means confined to the effects of intellectual property law. Other potential problems include distortion of the biomedi- cal research agenda as academic scientists seek to attract corporate sponsors; bypassing of traditional peer review mechanisms in allocating research funds, with consequent deterioration in the quality of research; alteration of the research and subsequent employment opportunities available to graduate students; divided loyalties of faculty affiliated with industry; compromise of university autonomy in 19871 Biotechnology Research asserted that granting property rights in research discoveries is antithetical to the norms of science.' In this Article I examine the interaction of intellectual property rights with research science norms in biotechnology-related fields to determine whether the two systems of rules and incentives conflict, and if so, how. Biotechnology research presents an unusual, if not unprecedented,' juxta- position of the incentives of the patent law system with the norms and such matters as faculty appointments; and loss of university credibility for impartial expertise on scientific issues. See generally M. KENNEY, supra note 1; Blumenthal, Gluck, Louis, Stoto & Wise, University-Industry Research Relationships in Biotechnology: Implications for the University, 232 SCIENCE 1361 (1986); Markle & Robin, Biotechnology and the Social Reconstruction of Molecular Biology, 10 Scs. TECH. & HUM. VALUES, Winter 1985, at 70; Note, Patent and Trade Secret Protec- tion in University-Industry Research Relationships in Biotechnology, 24 HARV. J. ON LEGIS. 191, 201-08 (1987); Note, Ties That Bind: Conflicts of Interest in University-Industry Links, 17 U.C. DAvis L. REv. 895 (1984). 4. For example, microbiologist Jonathan King has claimed that concern for preserving proprietary rights is retarding communication among university scientists, thereby undermining a previously shared "professional canon" of promoting the dissemination of knowledge for the public welfare: The openness, the free exchange of ideas and information, the free exchange of strains, of protein, of techniques, have been a critical component in the creativity and productivity of the biomedical research community ... This freedom of communication stemmed from the fact that all of the investigators shared the same professional canon; the increase of knowledge of health and disease for the general benefit of the citizenry ... Individuals planning to profit personally from commercial development, by, for example, assigning patents to their own firms, tend to cut down communication with their colleagues ... This secrecy is inimical to the function and effectiveness of universities in educating future generations of scientists, as well as retarding the research effort. Commercialization Hearings, supra note 3, at 62-63 (testimony of Dr. Jonathan King). 5. Throughout this Article I use the word "norms" in a normative rather than descriptive sense. By "research science norms" I mean socially inculcated beliefs within the research science community about how scientists should behave, as opposed to descriptions of how they actually do behave. See infra notes 8-28 and 110-54 and accompanying text. 6. In some fields, such as chemistry and engineering, industrial applications of research have been sufficiently immediate that universities and academic researchers have engaged in industry-sponsored research projects for years. In these areas, researchers have long confronted industry's concern for protecting intellectual property. See Commercialization Hearings, supra note 3, at 28-60 (testimony of Dr. Paul Gray); D. DICKSON, THE NEW POLITICS OF SCIENCE 89-90 (1984); Servos, The Indus- trial Relations of Science: Chemical Engineering at MIT, 1900-1939, 71 Isis 531 (1980); Smith & Hounshell, Wallace H. Carothers and Fundamental Research at DuPont, 229 SCIENCE 436 (1985). But in other fields, including many fields within the biomedical sciences, the gap between basic and applied research has been wider. See Peters & Fusfeld, Current U.S. University-Industry Research Connections, in NATIONAL SCIENCE BOARD, UNIVERSITY-INDUSTRY RESEARCH RELATIONSHIPS: SELECTED STUDIES 1, 20-21 (1982). In these fields, there have been fewer occasions to reconcile conflict between scientific norms and patent law. The sudden juxtaposition of commercial incentives and scientific norms has been particularly strik- ing in the biomedical sciences, in part because of the strong public interest in health-related research and in part because of the rapid onset and proliferation of university-industry research relationships in biotechnology fields following decades of predominantly public funding. Martin Kenney describes the prevailing attitude among biomedical researchers prior to the biotechnology boom as follows: The role of the state as a funder of basic (non-commodity-oriented) research fostered a power- ful ideology-one of scientists working for the public good to improve the health status of Americans. According to this ideology, industry's motives-especially that of profitabil- ity-were suspect, and the applied science orientation of industry was considered to be scientif- ically uninspiring to scientists. M. KENNEY, supra note 1, at 32. In light of this ideology, it is not surprising that the privatization of new knowledge as intellectual property has been resisted in the biomedical research community. The Yale Law Journal [Vol. 97: 177 incentives of a basic research community. Given the growing economic significance of research-oriented industries, it seems likely that patent law will have an increasing impact on the conduct of basic research. Thus, the experience of the biomedical research community with patent law may offer a preview of phenomena that will soon become more widespread. If patent law is in fact unsalutary for the conduct of basic research, it would be useful to figure out why. In Section I, I analyze the divergence between intellectual property law and research science norms as traditionally conceived. I begin by describ- ing a conventional view of the norms of research science concerning the dissemination of new knowledge and the rights of scientists in their dis- coveries. Against this background, I examine impediments to intellectual property protection for basic research discoveries in the biomedical sci- ences under traditional patent and trade secret doctrine. In Section II, I analyze the conjunction of the norms and incentives of research science and the rules and incentives of intellectual property in contemporary bio- technology research. First, I describe recent developments that have helped bring about this conjunction. Next, I analyze the operation of scientific norms concerning publication of research results in the specific context of contemporary biotechnology research, as illustrated by the records of a recent controversy within the Journal of Biological Chemistry. I then ex- amine patent law doctrine as applied to biotechnology inventions to deter- mine whether and how patent law conflicts with scientific norms in this particular research context. I conclude that, although there are substantial parallels between the two systems, the conjunction may nonetheless cause delay in the dissemination of new knowledge and aggravate inherent con- flict between the norms and the reward structure of science. I. THE TRADITIONAL DIVERGENCE BETWEEN INTELLECTUAL PROPERTY AND SCIENTIFIC NORMS IN BASIC RESEARCH Given the shared focus of research science and patent law on promoting invention and the dissemination of new knowledge, the proposition that scientific norms and patent laws have coexisted for centuries without 7. Patent laws have existed in this country since 1790 and in Europe since the 17th century. The origins of patent law are generally traced to the English Statute of Monopolies, 1623, 21 Jac., ch. 3, reprinted in 6 D. CHISUM, PATENTS app. at 8-1 (1987). Patents granted in the American colonies were primarily special enactments of colonial legislatures in favor of individual inventors. B. BUGBEE, GENEsIs OF AMERICAN PATENT AND COPYRIGHT LAW 57 (1967). The first United States patent statute was enacted in 1790, Patent Act of 1790, ch. 7, 1 Stat. 109-12, and revised in 1793, Patent Act of 1793, ch. 11, 1 Stat. 318-23. A more substantial revision occurred in the Patent Act of 1836, ch. 357, 5 Stat. 117-25, which created a patent office to examine patent applications for conformance to statutory criteria prior to the issuance of patents. The law has since been revised on numerous occasions. See Patent Act of 1839, ch. 88, 5 Stat. 353-55; Patent Act of 1870, ch. 230, 16 Stat. 198-217; Patent Act of 1897, ch. 391, 29 Stat. 692-94; Patent Act of 1903, ch. 1019, 32 Stat. 1225-27; Patent Act of 1928, ch. 730, 45 Stat. 732; Patent Act of 1939, ch. 450, 53 Stat. 1212; Patent Act of 1952, Pub. L. No. 593, 66 Stat. 792 (codified as amended at 35 U.S.C. §§ 1-376 (1982 & 1987] Biotechnology Research working out their differences calls for some explanation. Conflict may have been forestalled in the past by mutual avoidance: Particularly in the biomedical sciences, both legal doctrine and scientific norms have discour- aged attempts to protect basic research discoveries as intellectual property. A. A Traditional View of the Norms of Science Some observers have asserted that, prior to the recent growth of com- mercial interest in biotechnology, neither research scientists nor the insti- tutions funding basic scientific research in the biomedical sciences showed much inclination to pursue patent rights." Universities, where much of the research was conducted, encouraged the dissemination of research results through publication and occasionally showed a positive aversion to patent- ing discoveries. Public sponsorship of basic research permitted university researchers to avoid industrial sponsors who might have exerted pressure to protect in- tellectual property. Government sponsors gave universities and research- ers little incentive to identify patentable inventions or pursue patent rights. While the patent policies of the federal agencies that fund univer- Supp. 111 1985)). Robert Merton traces the origins of the ethos and norms of modem research science back to 17th- century Puritanism. R. MERTON, The Puritan Spur to Science, in THE SOCIOLOGY OF SCIENCE 228-53 (1973) [hereinafter THE SOCIOLOGY OF SCIENCE], originally published as R. MERTON, Mo- tive Forces of the New Science, in SCIENCE, TECHNOLOGY AND SOCIETY IN SEVENTEENTH-CENTURY ENGLAND 80-111 (1970). 8. Commercialization Hearings, supra note 3, at 62-63 (testimony of Dr. Jonathan King); THE SCIENCE BUSINESS, supra note 3, at 30-31 (1984) (background paper by Nicholas Wade). See gener- ally R. MERTON, The Normative Structure of Science, in THE SOCIOLOGY OF SCIENCE, supra note 7, at 275, originally published as Science and Technology in a Democratic Order, 1 J.L. & POL. Soc. 115 (1942). Merton cites a handful of counterexamples of prominent research scientists outside the biomedical sciences who have been patent holders, including Einstein, Millikan, Compton, and Langmuir. Id. at 275 n.16, citing W. HAMILTON, PATENTS AND FREE ENTERPRISE 154 (TNEC Monograph No. 31, 1941). As recently as 1975, academic scientists Cesar Milstein and Georges Kbihler decided not to patent their path-breaking and commercially valuable invention of monoclonal antibody-producing hybridoma cells. See M. KENNEY, supra note 1, at 129. Kenney suggests that they may have believed it inappropriate to patent such a potentially life-saving discovery. Id. 9. David Dickson quotes a 1934 policy statement by the president and fellows of Harvard Univer- sity that "'no patents primarily concerned with therapeutics or public health may be taken out by any member of the university, except with the consent of the president and fellows; nor will such patents be taken out by the university itself except for dedication to the public.'" D. DICKSON, supra note 6, at 89. Harvard abandoned this policy in 1975 after signing an agreement to give an exclusive world- wide license to Monsanto Corporation for inventions made in the course of research under a 12-year, $23 million grant from Monsanto to the Harvard Medical School. See M. KENNEY, supra note 1, at 58-60, 78; Culliton, Harvard and Monsanto: The $23-Million Alliance, 195 SCIENCE 759-63 (1977). A notable early counterexample to the generalization that universities were averse to patenting discoveries is the University of Wisconsin, which has administered patents and licenses resulting from research discoveries of faculty through a non-profit foundation since 1925. See Blumenthal, Epstein & Maxwell, Commercializing University Research: Lessons from the Experience of the Wisconsin Alumni Research Foundation, 314 NEW ENG. J. MED. 1621 (1986). 10. As recently as 1982, a National Science Foundation study reported that in many academic fields there appeared to be a "psychological barrier to interacting with industry." Peters & Fusfeld, supra note 6, at 39. The Yale Law Journal [Vol. 97: 177 sity research varied considerably prior to the Patent and Trademark Act Amendments of 1980,11 a common practice was to require that rights in inventions made in the course of government-sponsored research be as- signed to the government. This practice accorded with the prevailing wisdom since World War II that private ownership of inventions made through public funding was contrary to the public interest. The disinclination to secure patents also had a normative component: It was thought contrary to scientific norms to claim exclusive rights in re- search discoveries. These norms derive in part from the notion that mak- ing new observations available to the scientific community for evaluation and extension in further research facilitates the progress of science.1 In the biomedical sciences, this notion has been fortified by a belief that new knowledge should be made as widely available as possible in order to serve humanity.' A commonly held conception of the norms and incentives that guide the behavior of research scientists is set forth in the writings of sociologist Robert Merton.1 Merton identifies the overriding institutional goal of sci- 11. Pub. L. No. 96-517, 94 Stat. 3019 (codified at 35 U.S.C. §§ 200-11 (1982)). These revisions established a uniform policy permitting non-profit organizations (defined to include "universities and other institutions of higher education," 35 U.S.C. § 201(i)) and small business firms to retain title to patentable inventions made in the course of government-sponsored research. See infra notes 102-08 and accompanying text. 12. See Dobkin, Patent Policy in Government Research and Development Contracts, 53 VA. L. REV. 564, 568-84, 591-607 (1967). 13. President's Memorandum, 28 Fed. Reg. 10,943, 10,943-44 (1963); 1 U.S. DEP'T OF JUS- TICE, INVESTIGATION OF GOVERNMENT PATENT PRACTICES AND POLICIES: REPORT AND RECOM- MENDATIONS OF THE ATTORNEY GENERAL TO THE PRESIDENT 4-5 (1947). See generally Dobkin, supra note 12. 14. See infra text accompanying notes 17-28. 15. See infra text accompanying notes 17-19. 16. See Commercialization Hearings, supra note 3, at 63 (testimony of Dr. Jonathan King); M. KENNEY, supra note 1, at 32. These writings are collected in THE SOCIOLOGY OF SCIENCE, supra note 7. Merton is often cited in descriptions of the traditional ethos of research science. See, e.g., D. DICKSON, supra note 6, at 90 n.115; J. GASTON, ORIGINALrrY AND COMPETITION IN SCIENCE 5-6 & passim (1973); THE SCIENCE BUSINESS, supra note 3, at 29-30; Garfield, Citation Measures of the Influence of Robert K. Merton, 39 TRANSACTIONS N.Y. ACAD. Scs. SERIES II 61 (1980), and sources cited therein. The picture Merton paints of the scientific community has been criticized, however, as inaccurate and misconceived. See, e.g., J. RAVETZ, SCIENTIFIC KNOWLEDGE AND ITS SOCIAL PROBLEMS 311-12 (1971); Mulkay, Sociology of Science in the West, 28 CURRENT SOC. 1, 43-64 (1980), and sources cited therein; Stehr, The Ethos of Science Revisited, in THE SOCIOLOGY OF SCIENCE 172 (J. Gaston ed. 1978), and sources cited therein. The fact that scientists may depart from Merton's norms in practice does not necessarily mean that they lack normative authority within the scientific community. Nicholas Wade explains: Academic scientists quite regularly depart from the norms of universalism, communism, and disinterestedness-usually in the pursuit of personal recognition. Vigorously asserting one's claim to a discovery, even at the expense of others' legitimate claims, is surely not disinterested, and may not even be justified, but it is probably not in itself a serious threat to the purity or progress of science. Indeed, such behavior is part of the competitive attitude that characterizes many of the fastest-moving arenas of scientific inquiry. And the intensity of glory-seeking is sometimes moderated by the need to observe at least the appearance of Mertonian norms; thus The Double Helix, James Watson's account of the competition to discover the structure of DNA, occasioned a major stir within the scientific community not so much because it described unusual behavior but because it publicly acknowledged motives that scientists usually reserve Biotechnology Research ence as "the extension of certified knowledge."" The methodology for achieving this goal is empirical research. Merton describes four interre- lated norms that derive from the institutional goal and the methodology: 1 9 universalism, communism, disinterestedness, and organized skepticism. "Universalism" means that the veracity of claimed observations is to be determined on the basis of impersonal criteria without regard to the iden- tity of the scientist who makes the observation. Since what is true in one laboratory will also be true in another, all scientists, regardless of their personal qualities or national or institutional affiliations, may draw on and contribute to the same body of certified knowledge." "Communism" means that scientific findings are a product of social collaboration and should be dedicated to the scientific community. All discoveries build on what has been learned previously and contribute to what may be learned in the future. "Disinterestedness" means that scientists should seek truth rather than seeking to further their own interests by advancing spurious claims. "Organized skepticism" means that the scientific community should subject the claims and beliefs of its members to empirical scrutiny before accepting them as true. Merton describes a scientific reward structure that fortifies these norms. The scientific community rewards those who make original contributions to the common stock of knowledge by giving them professional recogni- 2 4 tion. The emphasis on originality creates pressure to publish as quickly as possible in order to avoid being forestalled by others who are doing 2 5 competitive research on the same problems. Upon publication, scientists lose exclusive use of their scientific property. The reward structure of sci- ence thereby reinforces Merton's norm of communism. By offering recog- nition and esteem to those who contribute to its shared body of knowledge, for private confessions. THE SCIENCE BUSINESS, supra note 3, at 30. The presence of a norm is indicated by community embarrassment and disapproval at its violation. For example, in acknowledging violations of scientific norms concerning the sharing of new knowl- edge in the biomedical sciences, Jonathan King stresses that such behavior is frowned upon: "I don't mean to say there isn't professional jealously [sic]. We have ambition and we have fame and recogni- tion, but it is considered a departure from the normal and you are embarrassed when it comes out. It is not what you are supposed to be doing." Commercialization Hearings, supra note 3, at 63. For evidence that Merton's norms continue to influence how researchers in the biomedical sciences believe they ought to behave, see infra notes 112-54 and accompanying text. 18. R. MERTON, The Normative Structure of Science, in THE SOCIOLOGY OF SCIENCE, supra note 7, at 267. 19. id. 20. Id. at 270-73. 21. Id. at 273-75. 22. Id. at 275-77. 23. Merton defines organized skepticism as "[t]he temporary suspension of judgment and the de- tached scrutiny of beliefs in terms of empirical and logical criteria." Id. at 277. This skepticism is applied to societal assumptions about the world as well as to claims of other scientists. Id. at 277-78. 24. R. MERTON, Priorities in Scientific Discovery, in THE SOCIOLOGY OF SCIENCE, supra note 7, at 286, originally published in 22 AM. Soc. REv. 635 (1957). 25. R. MERTON, Behavior Patterns of Scientists, in THE SOCIOLOGY OF SCIENCE, supra note 7, at 325-27, originally published in 105 PROC. AM. PHIL Soc'Y 470 (1961). The Yale Law Journal [Vol. 97: 177 the scientific community insures that scientists' self-interest will coincide with the public good. This reward structure also reinforces Merton's norm of disinterested- ness, in that publication involves submitting research results to the scru- tiny of other scientists. Efforts to gain recognition by making false claims are unlikely to succeed; false claims may be exposed if other scientists are 2 7 results. the fraudulent to replicate unable According to this model, both the norms and the rewards of science promote prompt disclosure of new discoveries through publication. Upon publication, the scientist's observations enter the public domain, permit- ting the scientific community to check the veracity of her claims and to build upon them in further research." To the extent that it might inter- fere with publication of new claims or limit the ability of other scientists to use published knowledge, intellectual property law has been perceived within the scientific research community as conflicting with the traditional norms and rewards of science. B. Traditional Intellectual Property Protection for Basic Research Discoveries At the same time, legal doctrine has discouraged the protection of basic research discoveries in the biomedical sciences as intellectual property. In most fields of technology, intellectual property protection can be obtained through the federal patent system, state trade secrecy law, or actual se- crecy. " Limitations on patentable subject matter and requirements for patentability have operated, however, to exclude basic research discoveries 26. Merton believed that proprietary rights in discoveries conflicted with the scientific norm of communism, R. MERTON, supra note 18, at 275, but in fact a similar communal value may be reflected in the patent law policy of requiring disclosure and ultimate dedication to the public as the quid pro quo of a patent monopoly. See infra text accompanying notes 155-206. 27. See generally Fraud in Biomedical Research: Hearings Before the Subcomm. on Investiga- tions and Oversight of the House Comm. on Science and Technology, 97th Cong., 1st Sess. 19-24, 55-58, 352-55 (1981) (testimony of Dr. Philip Handler, Dr. John C. Long, and Dr. Patricia Woolf, respectively) [hereinafter Fraud Hearings]. 28. The only "payment" owed by users of the published research is citation to the original author. J. RAVETZ, supra note 17, at 247, 255. The frequency of citation to a publishing scientist's work is a measure of the esteem in which it is held. Id. at 257. 29. A notable exception to this general rule is computer software, which may be protected under the copyright statute but not under the patent statute. See Chisum, The Patentability of Algorithms, 47 U. PIr. L. REV. 959 (1986); Davidson, Protecting Computer Software: A Comprehensive Anal)- sis, 1983 ARIz. ST. L.J. 611; Samuelson, CONTU Revisited: The Case Against Copyright Protection for Computer Programs in Machine-Readable Form, 1984 DUKE L.J. 663; cf. Diamond v. Diehr, 450 U.S. 175 (1981) (processes incorporating computer programs may be patented, although com- puter programs themselves may not be patented). For an argument that genetically engineered organisms are works of authorship entitled to copyright protection, see Kayton, Copyright in Living Genetically Engineered Works, 50 GEO. WASH. L. REv. 191 (1982). Recently, Congress has provided sui generis copyright-like protection for semiconductor chip masks. Semiconductor Chip Protection Act of 1984, Pub. L. No. 98-620, 98 Stat. 3347 (codified at 17 U.S.C. §§ 901-14 (Supp. III 1985)). See generally Symposium: The Semiconductor Chip Protection Act of 1984 and Its Lessons, 70 MINN. L. REV. 263 (1985). Biotechnology Research 1987] from patent protection. Even without patent protection, some basic re- search discoveries can be protected through the law of trade secrecy. De- pending on the scope of permissible subject matter under applicable state law, however, basic research discoveries may not qualify for legal protec- tion as trade secrets either, although they might still be protected as a practical matter through a program of actual secrecy. Both forms of se- crecy require substantial nondisclosure, thereby preventing researchers from attaining recognition and conflicting with traditional scientific norms. 1. Doctrinal Impediments to Patent Protection A patent gives an inventor" the exclusive right to make, use, and sell the invention in the United States for a period of seventeen years." At the end of this period, the invention becomes freely available to all. To qualify for patent protection, an invention must fall within patentable subject 3 2 3 matter and must satisfy the further statutory requirements of novelty, 3 4 5 utility, and nonobviousness." In exchange for the property rights con- ferred by the patent system, the inventor must disclose her invention to the public. Insofar as patent law not only permits but even requires disclo- sure, it is more congruent with scientific norms than either legal trade secrecy or actual secrecy. Nonetheless, patent protection has often been unavailable for basic research discoveries in the biomedical sciences. Traditional patent law doctrine has confined the reach of patent protec- tion to inventions in applied technology, as distinguished from basic scien- tific research. This focus on applied technology derives in part from the language of the Constitution, which authorizes Congress "to promote the Progress of Science and useful Arts, by securing for limited Times to Au- thors and Inventors the exclusive Right to their respective Writings and Discoveries." " This provision is traditionally read distributively. Under this reading, the function of the patent law is to promote the progress of "useful Arts" (i.e., applied technology) by securing exclusive rights to 30. Patent applications must generally be made by the inventor or inventors, 35 U.S.C. §§ 111, 115, 116 (1982), although in some instances persons other than the inventor who have proprietary interests in the invention may apply for a patent. 35 U.S.C. §§ 117, 118 (1982). Patents and patent applications may be assigned by the inventor to someone else. 35 U.S.C. §§ 152, 261 (1982). 31. 35 U.S.C. §§ 154, 271(a) (1982); see infra notes 207-14 and accompanying text. 32. 35 U.S.c. § 101 (1982). 33. 35 U.S.C. §§ 101, 102 (1982). 34. 35 U.S.c. §§ 101, 112 (1982). 35. 35 U.S.c. § 103 (1982); Graham v. John Deere, 383 U.S. 1, 13-19 (1966). 36. 35 U.S.c. § 112 (1982). 37. Cf Commercialization Hearings, supra note 3, at 79 (testimony of Dr. William Raub, Na- tional Institutes of Health) ("[Mlany biomedical scientists, perhaps most, regard the patent process as a means of institutionalized secrecy, whereas it is in fact a time-tested way to assure broad and ready access to proprietary information."); see also infra notes 162-206 and accompanying text (comparing disclosure under patent system with disclosure motivated by scientific norms). 38. U.S. CONsr. art. I, § 8, cI. 8. The Yale Law Journal [Vol. 97: 177 "Inventors" in their "Discoveries," while the function of the copyright law is to promote the progress of "Science" (i.e., knowledge in general) by 3 ' 9 securing exclusive rights to "Authors" in their "Writings. So under- stood, the Constitution only authorizes Congress to extend patent protec- tion to inventions in applied technology. Congress has accordingly defined patentable subject matter to limit pat- ent protection to applied technology. The present patent statute defines as patentable subject matter "any new and useful process, machine, manu- facture, or composition of matter, or any new and useful improvement thereof." Applying this definition and its predecessors in earlier legisla- tion, the cases have denied patent protection to theoretical or abstract dis- coveries,"' laws of nature, products of nature, principles, abstract ideas, mathematical formulae, and algorithms. Patents are issued only for the discovery of new means to achieve useful results." To the extent that ba- sic research discoveries comprise explanations for existing means of achieving useful results or principles that are only put into practical use later by the work of others, they may not be protected by patent." Patent law doctrine further restricts the patentability of basic research discoveries through the requirement that a patent applicant make a disclo- sure demonstrating that the invention is "operable and capable of use" before a patent will issue. This requirement impedes the patenting of basic research discoveries in two ways. First, if technical problems remain to be solved before the discovery may be put to practical use, it may not yet be patented. Second, an invention with no demonstrated "utility" may not be patented, however interesting and significant it may be to research scientists. The Supreme Court has held that a patent applicant must demonstrate a specific and substantial utility for a claimed invention. The 39. See, e.g., In re Bergy, 596 F.2d 952, 958-59 (C.C.P.A. 1979), affd sub noma. Diamond v. Chakrabarty, 447 U.S. 303 (1980) ("[Tlhe constitutionally-stated purpose of granting patent rights to inventors for their discoveries is the promotion of progress in the 'useful Arts,' rather than in sci- ence. . . . [T]he present day equivalent of the term 'useful arts' employed by the Founding Fathers is 'technological arts.' "). See generally 1 D. CHisuM, supra note 7, § 1.01 n.9 (1987) (Congress may enact patent legislation only to promote "useful arts," not to promote "science" or knowledge in general). 40. 35 U.S.C. § 101 (1982). 41. Mackay Radio & Tel. Co. v. Radio Corp. of Am., 306 U.S. 86, 94 (1938); In re King, 801 F.2d 1324, 1328 (Fed. Cir. 1986); 1 D. CHisuM, supra note 7, § 1.01. 42. Funk Bros. Seed Co. v. Kalo Inoculant Co., 333 U.S. 127 (1948). Parker v. Flook, 437 U.S. 584 (1978); Gottschalk v. Benson, 409 U.S. 63 (1972). 44. 1 D. CHISUM, supra note 7, § 1.01. 45. Id. § 1.0312]. As Chisum remarks, some courts have noted the irony of denying patent protec- tion to the discoverers of fundamental scientific principles, while extending protection to "those lesser geniuses who put such discoveries to practical uses." Id. (quoting opinion of Frank, J., in Katz v. Horni Signal Mfg. Corp., 145 F.2d 961, 961 (2d Cir. 1944)). The Supreme Court has justified the distinction on the ground that scientific principles are the "basic tools" of science and technology. Gottschalk v. Benson, 409 U.S. 63, 67 (1972). Judge Frank has offered another rationale: Great scientists are not motivated by pecuniary gain, so the prospect of a patent monopoly will not enhance their productivity. Katz, 145 F.2d at 961. 46. This requirement derives from 35 U.S.C. §§ 101, 112 (1982). See 1 D. CHISUM, supra note 7, at § 4.04. 19871 Biotechnology Research 187 fact that the claimed invention "may be an object of scientific research" does not satisfy the utility standard. Thus, under existing law, many basic research discoveries are not yet ripe for patent protection. Even in the area of applied technology, patent protection for inventions in the biomedical sciences has been retarded by cases holding that medical and agricultural processes may not be patented. While the rationale for these cases is not entirely clear," at least one court seems to have believed that a patent monopoly on methods of medical treatment is contrary to the ethics of medical doctors and the public interest: Doctors and surgeons have seldom thought it desirable to patent their new procedures for human relief. . . . The professional ethics of doctors and surgeons are more consistent with the widespread use of their medical and surgical discoveries for the benefit of mankind than in obtaining a monopoly to control their discoveries for personal commercial advantage. In this respect it would seem also that public interest is here involved. " Finally, prior to the Supreme Court's 1980 decision in Diamond v. Chakrabarty, patent protection for biological materials was retarded by the longstanding belief that living organisms and cells were unpatentable products of nature." One explanation for this limitation on patentable 47. Brenner v. Manson, 383 U.S. 519, 534-35 (1966). 48. See generally I D. CHISUM, supra note 7, § 1.03[3], and cases cited therein. Medical prod- ucts (such as drugs), as opposed to medical processes, may be patentable. See Ruskin v. Coe, 58 F. Supp. 424 (D.D.C. 1945). 49. The origin of the rule that medical procedures may not be patented is apparently the holding in Morton v. New York Eye Infirmary, 17 F. Gas. 879 (S.D.N.Y. 1862) (No. 9865). The inventors in that case discovered that ether could be used as an anesthetic for patients undergoing surgery. The court acknowledged that the inventors had made one of the "greatest discoveries of modern times," but concluded nonetheless that it was not a patentable invention: "A discovery may be brilliant and useful, and not patentable. . . . The new force or principle brought to light must be embodied and set to work, and can be patented only in connection or combination with the means by which, or the me- dium through which, it operates." Id. at 884. The court followed similar reasoning in Wall v. Leck, 66 F. 552 (9th Cir. 1895), extending the exclusion from patent protection to agricultural methods. In Ex parte Brinkerhoff, 24 Comm'r Ms. Dec. 349 (1883), reprinted in 27 J. PAT. OFF. SOC'Y 797, 798 (1945), the Commissioner of Patents offered another rationale for excluding medical proce- dures from patent protection-that the effectiveness of such treatments is uncertain, and granting a patent might have a tendency to deceive the public into believing that the method is more foolproof than it is. Subsequent decisions have retreated somewhat from the rules of Morton and Brinkerhoff, extending patent protection to a skin test for detecting the susceptibility of a human being to scarlet fever, Dick v. Lederle Antitoxin Laboratories, 43 F.2d 628 (S.D.N.Y. 1930), and a method of injecting medica- ments by a pressure jet, Ex parte Scherer, 103 U.S.P.Q. (BNA) 107 (P.T.O. Bd. App. 1954). 50. Martin v. Wyeth, Inc., 96 F. Supp. 689, 695 (D. Md. 1951), affd, 193 F.2d 58 (4th Cir. 1951). 51. 447 U.S. 303 (1980); see infra notes 60-66 and accompanying text. 52. See generally I D. CHISUM, supra note 7, § 1.02[7][a]-[b], and cases cited therein; Biggart, Patentability, Disclosure Requirements, Claiming and Infringement of Microorganism-Related In- ventions, in GENETICALLY ENGINEERED MICROORGANISMS & CELLS: THE LAW & THE BUSINESS C (I. Kayton ed. 1981) [hereinafter GENETICALLY ENGINEERED ORGANISMS]. Prior to Chakrabarty, two federal statutes provided intellectual property protection for new plants. The Plant Patent Protection Act of 1930, 46 Stat. 376 (codified at 35 U.S.C. §§ 161-164 (1982)), The Yale Law Journal [Vol. 97: 177 subject matter is that in order to be patentable a product must be new, and naturally-occurring organisms are not new. New processes using preexisting naturally-occurring products could be patented, but the prod- ucts themselves could not. An applicant who seeks a patent monopoly on a naturally-occurring product would deprive the public of the use of something "which nature has produced and which nature has intended to all men."1 the use of equally for be The Supreme Court's decision in Funk Bros. Seed Co. v. Kalo Inocu- lant Co. " suggested a broader, if somewhat less coherent, limitation on patentable subject matter under the "products of nature" doctrine. The claimed invention in that case was a combination of selected strains of naturally-occurring bacteria. When inoculated into leguminous plants, the 5 7 bacteria enabled the plants to fix nitrogen from the air. Although the mixture of strains was a new product in that the selected strains did not naturally occur in combination, the Supreme Court held the patent inva- lid. The precise basis for the decision is unclear. Justice Douglas, writing for the majority, suggested both that the invention was a product of nature outside the scope of patentable subject matter and that the combination of species failed to satisfy the nonobviousness or "invention" requirement. extended patent protection to asexually reproduced plants, and the Plant Variety Protection Act of 1970, Pub. L. No. 91-577, 84 Stat. 1542 (codified at 7 U.S.C. §§ 2321-2583 (1982)), allowed breed- ers of sexually reproduced plants to obtain patent-like protection from the Department of Agriculture. 53. Once the patent applicant has modified the product in any significant respect from its natu- rally-occurring state, this "product of nature" objection would seem to be overcome, since the claimed invention would then be a "manufacture" or "composition of matter" and thus within the statutory subject matter. I D. CHIsuM, supra note 7, § 1.02[7][a]. 54. In re Mancy, 499 F.2d 1289, 1294 (C.C.P.A. 1974) (distinction between new products and new processes); Guaranty Trust Co. of New York v. Union Solvents Corp., 54 F.2d 400, 411 (D. Del. 1931) (same). 55. Ex parte Latimer, 1889 Dec. Comm'r Pat. 123, 125 (1889). 56. 333 U.S. 127 (1948). 57. Different species of bacteria were effective for different crops, and multiple strains existed within each species. The patentee was the first to identify strains of bacteria that could be combined without inhibiting one another's effectiveness, permitting farmers to buy a single mixed culture inocu- lant usable for multiple crops. Id. at 129-30. 58. "The qualities of these bacteria, like the heat of the sun, electricity, or the qualities of metals, are part of the storehouse of knowledge of all men. They are manifestations of laws of nature, free to all men and reserved exclusively to none." Id. at 130. 59. [T]here is no invention here unless the discovery that certain strains of the several species of these bacteria are non-inhibitive and may thus be safely mixed is invention. But we cannot so hold without allowing a patent to issue on one of the ancient secrets of nature now disclosed. All that remains, therefore, are advantages of the mixed inoculants themselves. They are not enough. Id. at 132. In a concurring opinion, Justice Frankfurter suggested a more cogent basis for holding the patent invalid-that the patent claims were overly broad in extending to any mixture of compatible strains, rather than just the particular mixture derived by the patentee, and that the same failure to identify the particular compatible strains used by the patentee also made the description of the invention inadequate. Id. at 133-35. The description problem for inventions requiring the use of living organ- isms is now generally resolved by depositing cultures of the organisms in recognized depositories. See infra notes 167-78 and accompanying text. 1987] Biotechnology Research Thirty-two years later, in Diamond v. Chakrabarty, the Supreme Court held in a five to four decision that a living, genetically-engineered microorganism itself fell within the statutory categories of patentable sub- ject matter as either a "manufacture" or a "composition of matter." The majority held that living things are not categorically excluded from patent protection, noting that the relevant distinction is "not between living and inanimate things, but between products of nature, whether living or not, 8 1 and human-made inventions." While the Chakrabarty decision makes plain that the subject matter of patents extends to genetically-engineered microorganisms and that living things are not per se excluded from patent protection, the Supreme Court did not reach the issue of whether natu- rally-occurring microorganisms that have been newly isolated or purified also fall within the ambit of "manufactures" or "compositions of 6 2 matter." Since the Chakrabarty decision, the Patent and Trademark Office 6 3 (PTO) has extended patent protection to plants and indicated that non- naturally-occurring polyploid oysters would be patentable if the invention were nonobvious.11 The Commissioner of Patents recently issued a notice announcing that the PTO "now considers nonnaturally occurring non- human multicellular living organisms, including animals, to be patentable subject matter within the scope of 35 USC 1012"65 The Commissioner of 60. 447 U.S. 303 (1980). 61. Id. at 313. Chakrabarty modified an existing bacterial strain by introducing new de- oxyribonucleic acid (DNA) plasmids into a bacterial cell, thereby giving the organism the capacity to break down multiple components of crude oil. In so doing, he "produced a new bacterium with mark- edly different characteristics from any found in nature and one having the potential for significant utility." Id. at 310. The Court concluded that the resulting organism "is not nature's handiwork, but [Chakrabarty's] own." Id. 62. A companion case, In re Bergy, involved a claim to a biologically pure culture of an organism occurring in nature only in impure form. 563 F.2d 1031 (C.C.P.A. 1977), vacated sub nom. Parker v. Bergy, 438 U.S. 902 (1978), on remand, In re Bergy, 596 F.2d 952, cert. granted sub nom. Parker v. Bergy, 444 U.S. 924 (1979), vacated and remanded with instructions to dismiss as moot sub nom. Diamond v. Chakrabarty, 444 U.S. 1028 (1980). The Court of Customs and Patent Appeals held that the claims in both Chakrabarty and Bergy recited patentable subject matter. 596 F.2d at 952. Dicta from the Court of Customs and Patent Appeals in In re Bergy rejected the argument that a "biologi- cally pure culture" of an organism existing naturally only in impure form is an unpatentable product of nature, id. at 972-73, but so far no appellate court holding has resolved the issue. The Patent and Trademark Office (PTO) has been issuing patents on such organisms, and it seems likely that the Court of Appeals for the Federal Circuit would follow the Bergy dicta if the issue were to arise. See generally South Corp. v. United States, 690 F.2d 1368 (Fed. Cir. 1982) (Court of Appeals for Fed- eral Circuit will consider prior decisions of Court of Customs and Patent Appeals binding as precedent). 63. In re Hibberd, 227 U.S.P.Q. (BNA) 443, 444 (P.T.O. Bd. App. & Int. 1985). Plants may also qualify for more limited protection under the Plant Patent Act, 35 U.S.C. §§ 2321-2364 (1982), or the Plant Variety Protection Act, 7 U.S.C. §§ 2321-2583 (1982). The Plant Patent Act grants "the right to exclude others from asexually reproducing the [protected] plant or selling or using the plant so reproduced," 35 U.S.C. § 163, while the Plant Variety Protection Act grants the right "to exclude others from selling the [protected] variety, or offering it for sale, or using it in producing (as distin- guished from developing) a hybrid or different variety therefrom . . . ." 7 U.S.C. § 2483(a). 64. In re Allen, 2 U.S.P.Q.2d (BNA) 1425 (P.T.O. Bd. App. & Int. 1987). 65. Nonnaturally Occurring Non-Human Animals Are Patentable Under § 101, 33 Pat. Trade- mark & Copyright J. (BNA) No. 827, at 664 (Apr. 23, 1987). The Yale Law Journal [Vol. 97: 177 Patents cited Chakrabarty for the proposition that Congress intended the subject matter of patents to "include anything under the sun that is made by man," but reiterated that products found in nature may not be patented. In short, although patent protection has previously been unavailable for basic research discoveries in general and for those in the biomedical sci- ences in particular, the trend of recent authority is towards increasing availability of patent protection for biotechnology-related inventions. 2. Doctrinal, Practical, and Normative Impediments to Protection Through Secrecy Even in the absence of patent protection, however, inventors can some- times protect their research discoveries through some form of secrecy. The term "trade secrecy" is often used ambiguously to refer to two different types of protection: legal trade secrecy and actual secrecy. Legal trade se- crecy affords a remedy in tort to persons who disclose certain kinds of information in confidence against those who breach this confidence or who otherwise misappropriate the information. Actual secrecy is a practical, nonlegal strategy for protection that may be effective in circumstances where not all of the legal requirements for trade secrecy protection have been satisfied. Since in many jurisdictions legal trade secrecy doctrine might exclude from protection research discoveries not yet put to use in a 66. Id. It is not clear at this time whether Congress will permit the new policy of the PTO on the patentability of non-naturally-occurring, non-human animals to go into effect. Shortly after the Com- missioner's announcement, the Senate adopted an amendment to a supplemental appropriations bill, H.R. 1827, 100th Cong., 1st Sess. (1987), that would have barred the use of funds appropriated for fiscal year 1987 "for the purpose of granting any patent for vertebrate or invertebrate animals, modi- fied, altered, or in any way changed through engineering technology, including genetic engineering." Id., amendment no. 245, 133 CONG. REC. S7268 (daily ed. May 28, 1987); see Senate Votes Morato- rium on Patenting of Animals, 34 Pat. Trademark & Copyright J. (BNA) No. 833, at 124 (June 4, 1987). The Senate amendment was stricken, however, in conference, after the PTO indicated that it would not grant patents on animals during the remainder of fiscal year 1987. 100th Cong., 1st Sess., 133 CONG. REC. H5654 (daily ed. June 27, 1987) (Joint Explanatory Statement of the Committee of Conference); 34 Pat. Trademark & Copyright J. (BNA) No. 838, at 277 (July 16, 1987). A currently pending House bill would impose a two-year moratorium on the issuance of animal patents. H.R. 3119, 100th Cong., 1st Sess., 133 CONG. REC. H7206 (daily ed. Aug. 5, 1987); see House Bill Would Impose Two-Year Moratorium on Patenting Animals, 34 Pat. Trademark & Copyright J. (BNA) No. 841, at 351 (Aug. 6, 1987). In the meantime, the House Subcommittee on Courts, Civil Liberties and the Administration of Justice has held hearings on the animal patenting issue. See House Panel Holds Third Hearing on Animal Patents, 34 Pat. Trademark & Copyright J. (BNA) No. 845, at 460 (Sept. 3, 1987); House Panel Resumes Hearings on Animal Patenting Issue, 34 Pat. Trademark & Copyright J. (BNA) No. 840, at 321 (July 30, 1987); Hearing Ponders Patentability of Geneti- cally Altered Animals, 34 Pat. Trademark & Copyright J. (BNA) No. 835, at 191 (June 18, 1987). 67. For example, Roman Saliwanchik states that "[ain inventor has the right to hold any inven- tion in secret as a trade secret." R. SALIWANCHIK, LEGAL PROTECTION FOR MICROBIOLOGICAL AND GarIC ENGINEERING INVENTIONS 9 (1982) (emphasis added). This statement is clearly incorrect if he means to refer to legal trade secrecy, see infra text accompanying notes 68-88, yet his use of the terms "right" and "trade secret" suggests that it is indeed legal protection that he has in mind. 68. Actual secrets may also receive legal protection outside of trade secrecy law, through laws against trespass or theft. Biotechnology 19871 Research business, actual secrecy may offer the only feasible strategy for protecting such discoveries. Trade secrecy is usually governed by state common law, although an increasing number of jurisdictions have enacted statutory trade secret pro- tection, " and a handful of federal statutes protect trade secret rights once they have accrued under state law. Trade secret doctrine varies some- what among jurisdictions, but the basic requirements are similar in most states. State law generally limits the permissible subject matter of trade secret protection and further requires that the owner of a trade secret demonstrate some measure of actual secrecy, reasonable efforts to main- 7 3 tain secrecy, and misconduct by the defendant in acquiring, using, or secret. the trade disclosing In most jurisdictions, the permissible subject matter is quite broad, but it does have limits that potentially exclude basic research discoveries. The 7 5 definition in the first Restatement of Torts, which many jurisdictions follow, is susceptible to such a construction: A trade secret may consist of any formula, pattern, device or compi- lation of information which is used in one's business, and which gives him an opportunity to obtain an advantage over competitors who do not know or use it. It may be a formula for a chemical compound, a process of manufacturing, treating or preserving mater- ials, a pattern for a machine or other device, or a list of customers ... [I]t is not simply information as to single or ephemeral events in the conduct of the business . . . A trade secret is a process or device for continuous use in the operation of the business. Generally 69. See infra note 79; ARK. STAT. ANN. §§ 70-1001-1007 (1985); IDAHO CODE §§ 48-801 to -807 (1984); MASS. ANN. LAWS ch. 93, §§ 42, 42A (Law. Co-op. 1982); N.M. STAT. ANN. §§ 30-16 to -24 (1978); N.C. GEN. STAT. §§ 66-152 to -157 (1985). 70. See, e.g., Freedom of Information Act, 5 U.S.C. § 552b(c)(4) (1982) (exempting trade secrets from disclosure obligation of government agencies); Trade Secrets Act, 18 U.S.C. § 1905 (1982) (prohibiting disclosure of trade secrets by federal employees). 71. See, e.g., SI Handling Systems v. Heisley, 753 F.2d 1244 (3d Cir. 1985); American Alloy Steel Corp. v. Ross, 149 Cal. App. 2d 215, 308 P.2d 494 (1957); Revcor, Inc. v. Fame, Inc., 85 Il. App. 2d 350, 228 N.E.2d 742 (1967). 72. See, e.g., SI Handling Systems v. Heisley, 753 F.2d 1244, 1256 (3d Cir. 1985); K-2 Ski Co. v. Head Ski Co., 506 F.2d 471, 473-74 (9th Cir. 1974); Emery Co. v. Marcan Prod. Corp., 389 F.2d 11, 16 (2d Cir. 1968); Koehring Co. v. E.D. Etnyre & Co., 254 F. Supp. 334, 339 (N.D. I1. 1966). 73. See, e.g., Revcor, Inc. v. Fame, Inc., 85 Il. App. 2d 350, 355, 228 N.E.2d 742, 745 (1967). 74. See, e.g., E.I. duPont de Nemours & Co. v. Christopher, 431 F.2d 1012, 1014-15 (5th Cir. 1970), cert. denied, 400 U.S. 1024 (1971); Franke v. Wiltschek, 209 F.2d 493, 500 (2d Cir. 1953). 75. RESTATEMENT OF TORTS, § 757 comment b (1939). 76. See, e.g., Ruesch v. Ruesch Int'l Monetary Servs., 479 A.2d 295 (D.C. 1984); Koch Eng'g Co. v. Faulconer, 227 Kan. 813, 610 P.2d 1094 (1980); Engineered Mechanical Servs. v. Langlois, 464 So. 2d 329 (La. Ct. App. 1984); Jet Spray Cooler v. Crampton, 385 N.E.2d 1349 (Mass. 1979); Delta Filter Corp. v. Morin, 108 A.D.2d 991, 485 N.Y.S.2d 143 (N.Y. App. Div. 1985); Felmlee v. Lockett, 466 Pa. 1, 351 A.2d 273 (1976). The Restatement (Second) of Torts covers neither trade secrets nor the broader topic of "Miscella- neous Trade Practices" under which trade secrecy was subsumed in the first Restatement. See 4 RESTATEMENT (SECOND) OF TORTS 1-2 (1979). The Yale Law Journal [Vol. 97: 177 it relates to the production of goods, as, for example, a machine or formula for the production of an article." This definition could be read to exclude information about ongoing re- search on the ground that it has not yet ripened into "a process or device for continuous use in the operation of a business." Trade Secrets Act "trade secret" under the Uniform The definition of is considerably more amenable than the Restatement definition to a con- struction that permits protection of knowledge generated in the course of scientific research. Section 1(4) of the Act defines "trade secret" as follows: "Trade secret" means information, including a formula, pattern, compilation, program, device, method, technique, or process, that: (i) derives independent economic value, present or potential, from not being generally known to, and not being readily ascertainable by proper means by, other persons who can obtain economic value from its disclosure or use, and (ii) is the subject of efforts that are reasonable under the circum- stances to maintain its secrecy. As the Commissioners' Comment to this definition makes clear, the Uni- form Trade Secrets Act goes beyond the Restatement to extend trade se- cret protection both to information not yet put to use and to negative in- formation. Thus, in those jurisdictions that have enacted the Uniform Trade Secrets Act, trade secret protection may be more readily available 77. RESTATEMENT OF TORTS, supra note 75, at § 757 comment b. 78. Another type of research information that might be denied protection under a strict reading of the Restatement is "negative information," such as data on failed experiments. A researcher might well wish to keep such information secret, since knowing to avoid these dead ends might save competitor time and money. Yet this kind of information could be viewed as "simply information as to single or ephemeral events in the conduct of the business" and therefore not protectible as a trade secret. The Uniform Trade Secrets Act was approved by the National Conference of Commissioners on Uniform State Laws in 1979, and approved again with amendments in 1985. UNIF. TRADE SECRETS Acr, 14 U.L.A. 537-51 (1980 & Supp. 1986). It has been adopted by state legislatures in California, CAL. CIv. CODE §§ 3426-3426.10 (West 1984), CAL. CIV. PROc. CODE § 2019(e) (West 1987); Connecticut, CONN. GEN. STAT. ANN. §§ 35-50 to -58 (West 1983); Delaware, DEL. CODE ANN. tit. 6, §§ 2001-2007 (1986); Indiana, IND. CODE ANN. §§ 24-2-3-1 to -8 (West 1982); Kansas, KAN. STAT. ANN. §§ 60-3320 to -3330 (1983); Louisiana, LA. REV. STAT. ANN. §§ 51:1431-:1439 (West Supp. 1984); Minnesota, MINN. STAT. ANN. §§ 325C.01-.08 (West 1980); Montana, MONT. CODE ANN. §§ 30-14-401 to -409 (1985); North Dakota, N.D. CENT. CODE §§ 47-25.1-01 to -08 (1983); and Washington, WASH. REV. CODE §§ 19.108.010-.940 (1981). 80. The pertinent language of the Commissioners' Comment reads as follows: The definition of "trade secret" contains a reasonable departure from the Restatement of Torts (First) definition which required that a trade secret be "continuously used in one's business." The broader definition in the proposed Act extends protection to a plaintiff who has not yet had an opportunity or acquired the means to put a trade secret to use. The definition includes information that has commercial value from a negative viewpoint, for example the results of lengthy and expensive research which proves that a certain process will not work could be of great value to a competitor. UNIF. TRADE SECRETS AcT § 1 comment, 14 U.L.A., at 543 (emphasis in original). 1987] Biotechnology Research for research discoveries than it is in jurisdictions that follow the Restate- ment definition. The requirements of actual secrecy and reasonable efforts to maintain secrecy create considerable practical difficulties. Although absolute secrecy is not required, ' the claimed trade secret may not be generally known or 8 2 readily ascertainable by proper means. Efforts to maintain secrecy are costly and may be foiled. Microorganism cultures are especially difficult to maintain as trade secrets because they are easily stolen without detection and propagate rap- idly. A small sample of a culture can supply a competitor with bountiful quantities of commercially valuable organisms. The practical difficulty in- volved in protecting such materials as trade secrets is illustrated by a noto- rious early biotechnology trade secret case involving microorganism cul- tures used by American Cyanamid to produce antibiotics and steroids. Although the cultures were maintained as proprietary materials within the company, employees purloined sample cultures of the microorganisms from an American Cyanamid research facility and sold them to foreign pharmaceutical firms. The incident led to criminal convictions of two American Cyanamid employees" as well as civil trade secret actions for 8 5 damages. The final requirement of a trade secret claim-wrongdoing by the de- fendant in acquiring, using, or disclosing the trade secret-further limits the effectiveness of trade secrecy as a strategy for protection in competitive research fields. This requirement insulates from liability anyone who de- 8 6 rives the trade secret through independent research, reverse engineer- 81. Trade secrecy is not lost through disclosure in the course of a confidential relationship or pursuant to an agreement to keep the secret-indeed, the law of trade secrets protects the confidential- ity of such disclosures. 82. See, e.g., SI Handling Systems v. Heisley, 753 F.2d 1244 (3d Cir. 1985); K-2 Ski Co. v. Head Ski Co., 506 F.2d 471 (9th Cir. 1974); A.H. Emery Co. v. Marcan Prod. Corp., 389 F.2d 11 (2d Cir. 1968); Sandlin v. Johnson, 152 F.2d 8 (8th Cir. 1945); Juliano v. Hobart Mfg. Co., 200 F. Supp. 453 (D. Mass. 1961), affd, 303 F.2d 830 (1st Cir. 1962). 83. See, e.g., E.I. duPont de Nemours & Co. v. Christopher, 431 F.2d 1012 (5th Cir. 1970), cert. denied, 400 U.S. 1024 (1971). Typical measures include labeling documents or laboratory notebooks "Confidential," limiting disclosure on a "need to know" basis, and obtaining confidentiality agree- ments from employees and other persons to whom disclosure is made. See Sobel, Preserving Trade Secret Protection, in PRACTISING LAW INSTITUTE, PATENT, COPYRIGHTS, TRADEMARKS, & LITER- ARY PROPERTY COURSE HANDBOOK SERIES No. 224, PROTECTING TRADE SECRETS 567 (1986) [hereinafter PROTECTING TRADE SECRETS]. 84. United States v. Bottone, 365 F.2d 389 (2d Cir.), cert. denied, 385 U.S. 974 (1966) (criminal prosecution under National Stolen Property Act). 85. American Cyanamid Co. v. Sharff, 309 F.2d 790 (3d Cir. 1962) (civil contempt proceeding against defendant who refused to answer questions in deposition in civil action for misappropriation of trade secrets); American Cyanamid Co. v. Fox, 140 U.S.P.Q. (BNA) 199 (N.Y. Sup. Ct. 1964) (civil action for misappropriation of trade secrets and breach of employment agreement). 86. See, e.g., Speedry Chem. Prods. v. Carter's Ink Co., 306 F.2d 328, 330 (2d Cir. 1962) (right to independent discovery of secret has long history); Greenberg v. Croydon Plastics Co., 378 F. Supp. 806, 812 (E.D. Pa. 1974) (others in field who arrive at independent discovery may use plaintiff's trade secret). The Yale Law Journal [Vol. 97: 177 ing,8" or information obtained from publicly available sources." Once the "secret" becomes generally known to other scientists through independent discovery, the first discoverer loses protection. Thus, even where state law extends protection to research discoveries, the protection may not last for long. Trade secrecy presents a further problem for scientists who are condi- tioned by traditional scientific norms and rewards. The parsimonious hoarding of information necessary to maintain secrecy interferes with sci- entific communication and frustrates the pursuit of scientific recognition. Public disclosure, such as publication in a scientific journal or presenta- tion of a paper, ends actual secrecy and with it trade secret protection. Moreover, publication violates the requirement that the owner of the 9 0 maintain its secrecy. reasonable efforts to trade secret take While both trade secrecy and actual secrecy foreclose scientific recogni- tion in exchange for protection that may prove fragile and short-lived, trade secrecy may be more disruptive of scientific communication than ac- tual secrecy. Actual secrecy is only one of the elements of a trade secret claim; compliance with the other requirements, such as reasonable efforts to maintain secrecy, may also impede scientific communication. A public disclosure that facilitates reverse engineering may be enough to terminate 9 1 the right to enforce a trade secret claim even before it ends actual se- crecy. If researchers who wish to earn recognition in the scientific com- 87. See, e.g., Smith v. Dravo Corp., 203 F.2d 369, 375 (7th Cir. 1953) (person may legally obtain trade secret through inspection and analysis of product). 88. Sinclair v. Aquarius Elec., 42 Cal. App. 3d 216, 226, 116 Cal. Rptr. 654, 661 (1974) (by definition, trade secret is something which has not been placed in public domain). 89. See, e.g., Fishing Concepts, Inc. v. Ross, 226 U.S.P.Q. (BNA) 692, 695 (D. Minn. 1985) (disclosure of claimed trade secret in advertising textbook ends status as trade secret). The more typical avenue for terminating trade secrecy through public disclosure is the sale of a product from which the secret can be reverse engineered. See, e.g., Midland-Ross Corp. v. Sunbeam Equip. Corp., 316 F. Supp. 171, 177 (W.D. Pa.), affd, 435 F.2d 159 (3d Cir. 1970) (sale of product ends trade secrecy where nature of trade secret is ascertainable by inspection of product). But cf. Smith v. Dravo Corp., 203 F.2d 369 (7th Cir. 1953) (fact that trade secret could have been discovered through inspection of product does not preclude trade secret action against defendant who instead obtained and used information improperly); Tabor v. Hoffman, 118 N.Y. 30, 23 N.E. 12 (1889) (surreptitious duplication of patterns for manufacture of pump actionable notwithstanding fact that patterns could be derived from pumps themselves with sufficient time and effort); Classic Instruments v. VDO-Argo Instruments, 73 Or. App. 732, 700 P.2d 677 (Ct. App. 1985) (use of blueprints com- municated in confidence was actionable notwithstanding fact that trade secrets could be ascertained from product sold on open market). 90. The content of this requirement varies with the circumstances of each case. Disclosure to persons who have no need to know the secret, or who are not obligated to maintain its confidentiality, ends trade secret protection. Sandlin v. Johnson, 152 F.2d 8 (8th Cir. 1945); Crown Indus. v. Kaw- neer Co., 335 F. Supp. 749 (N.D. Il1. 1971). Requiring reasonable efforts to preserve secrecy serves to alert potential misappropriators to the owner's claim of trade secrecy and to weed out spurious after- the-fact claims of trade secrecy involving information that has been freely disclosed. 91. See Van Prods. Co. v. General Welding & Fabricating Co., 419 Pa. 248, 264-65, 213 A.2d 769, 778 (1965) (patent specification destroys trade secrecy after patent granted). 92. For example, suppose the scientists at American Cyanamid who discovered the tetracycline- producing microorganisms had wanted to gain recognition for their discovery in the scientific commu- nity through publication, but the company was unwilling to release the microorganism culture to the public. The company might have permitted publication of a description of the research, provided that Biotechnology 1987] Research munity can publish general descriptions of their research while keeping essential aspects of their inventions to themselves, they may find actual secrecy to be a more agreeable strategy than legal trade secrecy. II. CONJUNCTION OF INTELLECTUAL PROPERTY AND BASIC RESEARCH IN BIOTECHNOLOGY A. Developments Bringing About the Conjunction Scientists working in biotechnology-related fields are increasingly likely to be concerned simultaneously with the norms and rewards of research science and the rules and incentives created by intellectual property law. Research involving recombinant DNA and hybridoma technologies can create new organisms with the capacity to make desired products."' The obvious commercial potential of these technologies has attracted the inter- est of industrial sponsors. The products of biotechnology research are often readily commercialized-especially when the research is conducted in affiliation with commercial companies. The interest of industrial sponsors in biotechnology research is not con- fined to product development. In this field, the traditional dividing line between basic and applied research is blurred. Not only has the historical time lag between the two collapsed, but it has become difficult to charac- 9 8 terize given research problems as belonging in one category or the other. the microorganism itself were withheld from the public. In so doing, the company might have compro- mised its position in the event of a subsequent trade secret action. Arguably, the publication would have constituted a failure to use "reasonable efforts to maintain secrecy" or made the invention "read- ily ascertainable by proper means." Nonetheless, even without a legal trade secret remedy, American Cyanamid might have been able to maintain the microorganism culture as an actual secret for a while by keeping it within the company. 93. See generally Grobstein, supra note 3. 94. Recombinant DNA technology can alter the genetic information of cells, enabling the cells to make desired proteins. Every living cell contains genetic information in the form of deoxyribonucleic acid (DNA) inherited from the progenitors of the organism of which that cell is a part. When cells replicate, the new cells in turn inherit a copy of the parent cells' DNA. DNA is a long molecule in the form of a double helix composed of two chains of subunits called nucleotides. One of the compo- nents of each nucleotide is a base, of which there are four different kinds: adenine, guanine, cytosine, and thymine. DNA provides the cell with instructions on how to make the proteins it needs to survive. These instructions are coded in the sequence in which the four bases occur. Each group of instructions corresponding to an individual protein is called a gene and comprises a small portion of one chain of the DNA molecule. All organisms use the same genetic code for protein synthesis, although different organisms make different proteins in accordance with the different DNA sequences in their cells. Recombinant DNA technology involves taking genetic information from one cell and splicing it into the DNA of another cell-or even combining the DNA of two different organisms. In this manner, simple, single-celled organisms such as bacteria can be programmed to make human proteins. The recombinant organism can then be used as a living factory to make the protein encoded in the cloned gene. See W. KEETON, BIOLOGICAL SCIENCE 63-64, 665-66 (3d ed. 1980). Hybridoma technology involves the fusion of antibody-producing lymphoid cells with tumor cells to create new cell lines that can make desired antibodies. 95. A 1984 study reveals that industry may be funding as much as one-quarter of all biotechnol- ogy research in universities, with nearly one-half of all biotechnology companies funding such re- search. Blumenthal, Gluck, Louis & Wise, supra note 1. 96. For example, cells contain more genetic information than they actually use. Learning what causes cells to "express" certain genes and not others would not only answer major questions in The Yale Law Journal [Vol. 97: 177 Academic and industrial researchers are often working on the same or closely related problems, whether competitively or collaboratively. " Note- worthy scientific discoveries are made in industrial laboratories, and pat- entable inventions are made in university laboratories. In addition, the capital markets are placing a dollar value on intellec- tual property long before a product is ready for market. The prospectuses that high technology companies use to attract investment capital advertise the companies' affiliations with university facilities and researchers. These affiliations enhance the companies' credibility with investors. They also supply companies with a window on new scientific developments, 9 9 which further expedites commercialization. Sometimes the scientists themselves have equity interests in the companies sponsoring their work. Even if they do not, however, research scientists tend to be more aware of the legal and commercial rights arising out of their work when their sponsors take an interest in those rights. When industry funds basic research, the scientists and universities in- volved may be contractually bound to preserve patent rights in sponsored research. But even publicly funded research is increasingly likely to be protected as intellectual property. Many universities have in-house patent or licensing departments or use the services of outside patent development 1 0 2 organizations. "' The 1980 Patent & Trademark Act Amendments en- courage universities to patent and license inventions made in the course of government-sponsored research. Under this legislation, universities must 0 3 report any invention which "is or may be patentable" to the sponsoring agency within a reasonable time; if they do not, the goyernment may re- 0 4 ceive title to the invention. Universities may elect to retain patent rights, subject to a non-exclusive license to the agency for use of the inven- 0 5 tion, but only if they agree to file timely patent applications once their 1 06 research results have been published. Government "march-in rights," through which the sponsoring agency can license the university's inven- biology, but might also suggest ways of facilitating the manufacture of desired proteins or suppressing the expression of disease-causing genes. This knowledge would therefore be valuable to both scientists and industry. 97. D. DICKSON, supra note 6, at 74-77; Blumenthal, Gluck, Louis & Wise, supra note 1; Grob- stein, supra note 3. 98. Commercialization Hearings, supra note 3, at 7 (testimony of Dr. Donald Kennedy); see, e.g., CALGENE, INC., PRELIMINARY PROSPEcTus 25, 28-29 (May 23, 1986); INTEGRATED GENET- ICS, INC., AMENDMENT No. 1 To FORM S-1, at 17, 22, 25-26 (filed with SEC Apr. 24, 1986). 99. Peters & Fusfeld, supra note 6, at 34-36. 100. Blumenthal, Gluck, Louis, Stoto & Wise, supra note 3, at 1364. 101. M. KENNEY, supra note 1, at 74-77; Blumenthal, Epstein & Maxwell, supra note 9; Wil- liams, Business Aspects of Biotechnology 25, in GENETICALLY ENGINEERED ORGANISMS, supra note 102. Pub. L. No. 96-517, 94 Stat. 3019 (1980) (codified at 35 U.S.C. §§ 200-212 (1982)). 103. See 35 U.S.C. § 201(d); see also 35 U.S.C. § 202(c)(1). 104. See 35 U.S.C. § 202(c)(1). 105. See 35 U.S.C. § 202(c)(4). 106. See 35 U.S.C. § 202(a), (c)(2), (c)(3). Biotechnology Research 1987] tions itself if the university fails to do so, encourage universities to exploit 10 7 their patents. The statute also requires universities to share patent roy- 0 8 alties with inventors.' This personal financial stake gives researchers an incentive to be alert to patent rights. Finally, most biotechnology research makes use of patented inven- 10 9 tions. Thus, even researchers who have no interest in protecting their own intellectual property run up against the intellectual property system when they try to gain access to the inventions of others. In short, intellectual property law pervades contemporary biotechnology research, causing significant concern about the impact of secrecy and ex- clusive rights on the behavior of research scientists in fields where scien- tific norms call for publication and dedication of research results to the public. B. The Norms of Science in Contemporary Biotechnology Research Wholly apart from any conflict between the law of intellectual property and traditional scientific norms, contemporary biotechnology research presents scientists with a conflict between scientific norms and scientific rewards. As we have seen, in theory these norms and rewards are congru- ent: Publication of original research results both rewards the publishing scientist with professional recognition and satisfies the norm of permitting other scientists to replicate one's research results. But for research involv- ing the use of unique biological materials, such as bacterial strains and other types of self-replicating cells, publication in writing alone may not be sufficient to satisfy this replicability norm. To replicate the authors' results, subsequent investigators may need access to identical materials. By sharing access to unique materials, however, the publishing scientist not only enables other scientists to replicate her claims; she also allows them to compete with her more effectively in making new discoveries. The professional advantage that exclusive access to unique research materials gives her may motivate her to retain exclusivity for as long as possible in order to maximize the discoveries for which she can claim priority. Of course, scientists in other fields of research also run the risk of losing future claims to priority when they publish their current research results. But at some point reluctance to help the competition is likely to be over- come by desire to earn recognition for priority of discovery, which is es- tablished through prompt publication. Delaying publication may handicap the competition, but the dilatory scientist thereby runs the risk that some- 107. See 35 U.S.C. § 203. 108. See 35 U.S.C. § 202(c)(7)(C). 109. Cf Fox, Patents Encroaching on Research Freedom, 224 SCIENCE 1080 (1984) (discussing concern in biotechnology research community about impact of restrictions on research use of patented products and processes). 110. See infra notes 112-46 and accompanying text. The Yale Law Journal [Vol. 97: 177 one else will publish first and get all the credit. So long as scientists must publish their results in a form that permits replication in order to earn recognition for priority, the reward structure of science operates to rein- force scientific norms. In biotechnology, however, the requirement of publication is a less ef- fective mechanism for enforcing scientific norms than in some other fields. As noted above, replication of research using unique biological materials generally requires access to more than the published text. Since the mater- ials themselves cannot be made a part of the text, publishing scientists could conceivably gain recognition through publication without sharing their materials. Withholding materials is a relatively inconspicuous depar- ture from scientific norms. It occurs after publication and is not apparent from the written text."' Thus, publishing scientists with exclusive access to such materials have an opportunity to gain recognition while retaining a future advantage over their research competitors. This conflict between norms and incentives is aggravated when the materials (or the discoveries they facilitate) have potential commercial value. The records of a controversy that arose when a team of scientists from a commercial biotechnology firm submitted a manuscript to the Journal of Biological Chemistry (JBC) in the summer of 1980 dramatically illustrate 1 12 this conflict. The JBC, a scientific journal that frequently publishes reports of research using biological materials, had a policy requiring au- thors to make available to other researchers any biological materials re- 11 8 ferred to in published manuscripts. The manuscript submitted by the commercial scientists reported the development of a recombinant organism 1 1 4 containing a newly synthesized gene. The authors described in detail how they synthesized the DNA insert. But in response to a question from 111. A publication may designate expressly that materials are available from an independent source, such as a cell culture depository, but the absence of such a designation does not necessarily mean that the author would be unwilling to supply the materials. Some authors would rather supply the materials themselves in order to be sure of the identity and viability of the sample. See infra note 112. The subject of commercial biotechnology commanded considerable public attention at this time, which roughly coincided with the Supreme Court's Chakrabarty decision and the initial public offering of stock by Genentech, Inc. See Hamilton, Biotech's First Superstar, Bus. WEEK, Apr. 14, 1986, at 68. 113. This "availability" policy was set forth as Instruction #c.4 in the written instructions to authors, published in each new volume of the JBC, as follows: The MICROORGANISMS and TISSUE CULTURE STRAINS referred to in a manuscript should be identified by an American Type Culture Collection number, or that of a comparable collection. If this is not possible, it is understood that the authors will make the particular strains used available to interested investigators upon request. Well mapped strains of Escherichia coli should be sent to the E. coli Genetic Center at Yale University. 255 J. BIOLOGICAL CHEMISTRY 2-3 (1980). 114. The authors reported the transformation of a well-known strain of bacteria with a recombi- nant DNA plasmid containing a newly synthesized gene. There were thus a number of different biological materials "referred to" in the manuscript: the original bacterial strain, the original plasmid, the synthesized gene, the recombinant plasmid containing the synthesized gene, and the transformed bacterial strain containing the recombinant plasmid. In keeping with the JBC's policy, the original bacterial strain was identified by a collection number. 19871 Biotechnology Research an editor, they indicated that their company's policy prevented them from giving out the recombinant strain. This event prompted an extensive discussion of the journal's "availabil- 1 15 ity" policy at a meeting of the JBC editorial board, culminating in an informal vote of approximately seventy-five percent in favor of retaining percent opposed.1 ' and twenty-five the policy The editorial board decided to seek the advice of an ad hoc committee"" consisting of six academic scientists and two industry scientists on the fol- lowing question: "Should this Instruction to Authors be retained as is, 115. The content of this discussion is summarized in an enclosure to Letter from William F. Harrington to Nathan 0. Kaplan (July 11, 1980) [hereinafter Summary] (on file at Yale Law Jour- nal). Dr. Harrington is on the faculty of the McCollum-Pratt Institute, The Johns Hopkins Univer-" sity. Dr. Kaplan is a professor in the Department of Chemistry, University of California at San Diego. 116. Arguments raised in favor of retaining the availability policy emphasized three core norms: 1. Replicability. Scientists should publish their research claims in a form that permits replica- tion so that the scientific community can verify their claims: "Once an experiment is published anyone should be able to reproduce it, otherwise it is worthless and might as well just not be published." Summary, supra note 115. Replication requires access to the publishing scientist's unique materials. 2. Dedication. Publishing scientists should make their scientific advances available to the public: [T]he act of publishing is the giving of a scientific advance to the public in a form which permits other scientists to reproduce the experiments. . . . [To withhold any necessary ingre- dient which is not generally available is morally equivalent to withholding essential informa- tion from the text of the publication and authors should in general be required to make availa- ble to other scientists those rare materials that go into the published experiments. Id. This dedication norm is violated by withholding essential information or rare materials that the author is able to supply. 3. Progress. Scientists should publish their research in a form that facilitates the progress of science: "The work cannot be repeated or extended without the availability of the strain or culture." Id. (emphasis added). 117. Arguments raised against the policy included: 1. Inconsistency. The policy is inconsistent in that it requires authors to make living materials available but does not require authors to make non-living materials available: We don't require distribution of organic compounds. . . .[A] complicated organic synthesis is more difficult to repeat than a DNA recombination experiment. . . .[I]t would be enough to insist (as in all other experiments) that enough experimental detail be given to know what was done and to permit, in most cases, repetition by a knowledgeable investigator. Id. The editors acknowledged parenthetically that "this [repetition] could not be done if a rare cell line from a clinical case is involved." Id. 2. Free riders. The policy permits free riders to benefit from the work of authors without compensating them: Such a rule forces an investigator in a university to supply his strain to a profit-making organ- ization without any financial return to his laboratory or university, regardless of the amount of profit eventually received. A similar argument concerns a single investigator developing a spe- cial strain after years of work. (One suggestion . . . is that the rule have a time-delay of one or two years before a culture has to be deposited ... . Id. 3. Nonenforceability. The policy would be difficult or impossible to enforce. If the JBC were to enforce the policy and other journals did not, authors would choose to publish elsewhere: [W]ould some authors not publish (or at least not in this journal) if such a requirement were enforced? Thus, would we be hindering, rather than helping, communication by such a policy? If all the journals don't have the same policy, would the JBC be losing papers in some of the most active and exciting fields? [As for those authors who do publish in the JBC,] [t]he journal cannot act as a policeman. Id. 118. Id. The Yale Law Journal [Vol. 97: 177 or deleted in the light of recent advances in molecular biology[?]" modified Each of the scientists responded by letter." These eight letters, written as of proprietary rights in biotechnology inventions were just be- the effects ginning to be felt, illustrate a range of views within the biomedical re- as to what the norms of science are, what pur- search community itself poses they serve, and how they actually function in practice. viewpoints. Two The six academic scientists represented three distinct 2 1 two favored favored retaining the policy without modification.' Another rather than continuing the policy as a statement of the views of the JBC, 2 2 the as a condition of publication.' The remaining two favored modifying to policy to limit it to those materials that are strictly necessary in order scientists permit replication of the published work. ' The two industry the need for were the most hostile toward the policy, questioning even 2 4 publication in replicable form.' 1. Retention of the Policy The two scientists who favored retaining the policy without modifica- 2 "5 tion' came closest to embracing Merton's traditional norms of commu- of science," "the nism and universalism, emphasizing "the best interest freedom of science,' and "the general ethic of bacterial and phage ge- neticists."12' Both rejected the argument raised at the editorial board com- meeting that, in treating organisms differently from new chemical 2 8 are pounds, the policy is inconsistent.' They pointed out that organisms chemicals can be difficult to synthe- relatively easy to reproduce, whereas size and supply.' One of the two expressly opposed a different rule for scientists were briefed on the issue with a 119. At the time their views were solicited, the eight at the editorial board meeting. See written summary of arguments raised for and against the policy and accompanying text. The scientists referred to these arguments in their indi- supra notes 115-17 vidual letters. 120. I am grateful to the editors of the Journal of Biological Chemistry and to Paul Berg, Donald Brown, William Harrington, Nathan Kaplan, Philip Leder, Daniel Nathans, Jesse Rabinowitz, Or- rie Friedman, and I.S. Johnson for permission to use their letters. 121. See infra notes 125-31 and accompanying text. 122. See infra notes 132-36 and accompanying text. 123. See infra notes 137-39 and accompanying text. 124. See infra notes 140-45 and accompanying text. 125. These two scientists were Dr. Nathan 0. Kaplan, see supra note 115, and Dr. Philip Leder, Institutes of Health and is now Chair- who was then with the Public Health Service of the National of Genetics at Harvard Medical School. See Letter from Nathan 0. Kaplan man of the Department (Aug. 5, 1980) [hereinafter Kaplan letter] (on file at Yale Law Journal); to William F. Harrington Letter from Philip Leder to William F. Harrington (July 25, 1980) [hereinafter Leder letter] (on file at Yale Law Journal). 126. Kaplan letter, supra note 125. 127. Leder letter, supra note 125. 128. See supra note 117. for any investigator to regrow the organism and 129. Dr. Kaplan explains: "[O]ne cell is enough thereby have the organism available for his studies. Living systems can be kept and maintained but chemicals can be utilized and not replaced unless someone redoes the synthesis or preparation." Kaplan letter, supra note 125. Dr. Leder simply noted that the two situations involved "clearly differ- ent . . . logistics . . . " Leder letter, supra note 125. 1987] Biotechnology Research recombinant organisms, reasoning that such a rule would lead to different 1 3 standards for academic and commercial scientists. The other acknowl- edged the existence of competing incentives both to protect intellectual property and to earn scientific recognition, but nonetheless opposed com- promising scientific norms to accommodate either "proprietary interests" 1 3 1 or "the all-too-human foibles of individual scientists. 2. Policy as Statement of Principle The two scientists who favored retaining the policy as a statement of 13 2 principle rather than as a binding contractual term conceded the impor- tance of the traditional norms' but nonetheless felt that the JBC was 3 4 unable to enforce the policy. Both recognized that the policy is some- 3 5 times violated in practice, and suggested that it be reformulated as an exhortation, rather than a requirement for publication. 3. Modification of the Poicy The remaining two academic scientists favored a narrower policy re- quiring availability only to the extent necessary for replication of pub- 1 3 7 lished experiments. Both drew a distinction in this regard between or- 130. See Kaplan letter, supra note 125. Dr. Kaplan objected to scientists earning recognition without observing scientific norms: "[Slome of the people involved in commercial laboratories would like to . . .receive recognition for their research yet not follow the general ethics of scientists ... One cannot have it two ways." Id. 131. Leder letter, supra note 125. 132. These two scientists were Dr. Paul Berg, Department of Biochemistry, Stanford University Medical School, and Dr. Donald Brown, Department of Embryology, Carnegie Institution of Wash- ington. See Letter from Paul Berg to William F. Harrington (July 31, 1980) [hereinafter Berg letter] (on file at Yale Law Journal); Letter from Donald D. Brown to William F. Harrington (July 28, 1980) [hereinafter Brown letter] (on file at Yale Law Journal). 133. Dr. Berg praised the objective embodied in the policy as "the ideal we should be striving towards" and "a principle that is nearly as defensible as motherhood." Berg letter, supra note 132. 134. Dr. Berg thought that "without comparable admonitions by the other journals. . . it would be a useless gesture." Id. Dr. Brown concurred that the idea behind the availability policy was sound, but thought that "to make it a law is a little like Prohibition, it is unrealistic and unenforceable." Brown letter, supra note 132. 135. Dr. Berg noted that the policy "reaches for a kind of openness and sharing in science that many of us sould [sic] like to see practiced universally." Berg letter, supra note 132. Dr. Brown felt sure that previous JBC authors had "refused or delayed sending [organisms] . . .to their competi- tors," and added that "[t]he problem is not made new by recombinant DNA proliferation, nor by its increasing popularity with industry." Rather, he attributed the reluctance of scientists to share strains to the fact that they "want time to explore their discoveries." Brown letter, supra note 132. 136. Dr. Berg proposed that the JBC "construct a statement that sets forth clearly and forcefully the JBC's belief on [sic] the responsibilities of researchers and authors to make available, along with the data of their investigations and experiments, the biological material used in acquiring that infor- mation" and urge other journals to do the same. Berg letter, supra note 132. Dr. Brown favored refraining the policy to state that authors are "encouraged" rather than required to make their strains available. Brown letter, supra note 132. 137. These two scientists were Dr. Daniel Nathans, Department of Microbiology, The Johns Hopkins University School of Medicine, and Dr. Jesse C. Rabinowitz, Department of Biochemistry, University of California, Berkeley. See Letter from Daniel Nathans to William F. Harrington (Aug. 7, 1980) [hereinafter Nathans letter] (on file at Yale Law Journal); Letter from Jesse C. Rabinowitz to William F. Harrington (Aug. 28, 1980) [hereinafter Rabinowitz letter] (on file at Yale Law The Yale Law Journal [Vol. 97: 177 ganisms isolated from nature and organisms synthesized by the authors, arguing that authors should be required to make only the former availa- ble. They justified this distinction primarily on the basis of the repli- cability standard." Both also noted that departures from the policy were not uncommon in the case of newly-synthesized organisms, even among 3 9 noncommercial scientists.' 4. Nonenforcement The two industry scientists both asserted that if the policy were en- forced, the JBC would lose manuscripts from scientists in commercial lab- 4° oratories. One of them noted that enforcement of the policy could ulti- mately undermine the goal of disseminating new knowledge by deterring scientists in commercial laboratories from publishing,"" and added that 4 2 the problem is not confined to commercial scientists. He added that Journal). 138. Dr. Nathans thought that the JBC should also encourage authors to make available new mutants or recombinants described in published reports, as well as new isolates from nature, in order to facilitate research. But although it would be "desirable" to have such mutants or recombinants freely available once they have been reported in the literature, he felt that it is neither desirable nor possible "to coerce people into giving out their materials." Dr. Nathans nonetheless favored retaining the policy as a binding condition of publication for "new isolates from nature of microbes or tissue culture cells" where "the principle of verifiability of research results operates." Nathans letter, supra note 137. Dr. Rabinowitz felt that both mutants and new isolates from nature should be made available because "it is not possible to characterize such organisms exactly or completely or to be certain that two organisms isolated by identical experimental procedures are, in fact, genetically identical unless they can be compared by genetic analysis." Rabinowitz letter, supra note 137. He noted, however, that where the manuscript provided a description of the experiment sufficient to permit replication, as would be the case for products of chemical synthesis and some recombinant organisms, it would not violate scientific norms to withhold the materials. Id. 139. Dr. Nathans noted that scientists who have constructed a new organism "may want to limit distribution of strains that were difficult to construct or especially important for their future work." He added that commercial incentives were compounding the problem: "Add to this an increasing number of recombinants with potential commercial value and you have a widespread interest in con- trolling the availability of new strains." Nathans letter, supra note 137. Dr. Rabinowitz observed that investigators often avoid making available organisms that they have derived themselves, yet these transgressions have generally been overlooked. He attributed the reluc- tance of authors to share their strains not to competing commercial interests, but to the interests of the authors in retaining exclusive access to a new research tool: "[Miost investigators recognize that when mutants or organisms isolated by enrichment techniques are involved, . . . free exchange of these materials has frequently not occurred. The usual reason for this has probably been to protect the advantage that the original investigator possesses having isolated an organism." Rabinowitz letter, supra note 137. 140. These two scientists were Dr. Orrie M. Friedman, then President and Scientific Director and now Chairman of the Board of Collaborative Research, Inc., and Dr. I.S. Johnson, Vice Presi- dent, Lilly Research Laboratories, A Division of Eli Lilly and Company. See Letter from Orrie M. Friedman to William F. Harrington (Aug. 4, 1980) [hereinafter Friedman letter] (on file at Yale Law Journal); Letter from I.S. Johnson to William F. Harrington (Aug. 18, 1980) [hereinafter Johnson letter] (on file at Yale Law Journal). 141. I can say that under some circumstances commercial labs such as ours clearly would not be able to make certain of such strains available publicly. If as a result we would be barred from publication it would be regrettable from the point of view of both the individual scientist who did the work and of the others who would be deprived of the results of the work. Friedman letter, supra note 140. 142. Dr. Friedman asserted that "[t]he fact is that ... organisms . . . referred to in published Biotechnology Research 1987] necessary to ensure replicability of re- availability of strains may not be 1 3 scientist, going a step further, questioned sults. The other industry whether the JBC should even require that authors make available those materials necessary for replication. He noted that a policy requiring authors to enable readers to reproduce their work would "reduce publica- intellectual tion incentive," citing competing incentives to preserve property. of the policy "as a Ultimately, the JBC decided to retain the wording than as a contractual agreement between the statement of principle, rather Journal and an author." The chairman of the ad hoc committee concluded the statement, that a weakening of "understood" to "urged" or "en- such as changing the word couraged," would be interpreted as a compromise in values set by most agree should not occur. Moreover, since the the Journal, which in the past, we see no statement seems to have worked satisfactorily need to alter it at this time as long as the Journal makes no attempt an oral or written pledge or to enforce the statement by requiring authors. the ethics of policing otherwise a straightforward conflict A superficial view of this episode suggests norms and competing incentives arising from the desire between scientific The ultimate decision of the JBC to preserve intellectual property rights. of the policy the same but not to enforce it lends to keep the wording support to the view that while the traditional norms continue to hold some moral appeal for scientists, the research science community may be losing its ability to enforce its own norms against competing commercial incentives. papers often are not generally available upon request, often even from academic investigators." He defended this practice on the ground that "in particular instances this [complete access] would place an intolerable burden upon the investigator." Id. to make this point: "Moreover it may not be 143. Dr. Friedman used the language of patent law essential that they be available, provided the publication contains sufficient information so that one the organism etc. on truly skilled in the art (as opposed to a casual practitioner) is able to reproduce (patent specification must enable any person "skilled in the his own." Id.; cf. 35 U.S.C. § 112 (1982) and use invention). art to which it pertains" to make 144. I cannot accept the position, expressed by some, that a manuscript that is not wholly enabling is valueless. Many such manuscripts add to the fund of information, sometimes in a highly significant manner ... As a matter of fact, many manuscripts are readily accepted for publication even though they report work that, for one reason or another, cannot or will not be verified by others. Johnson letter, supra note 140. Dr. Johnson illustrated his point with the following example: "[A]n investigator collects thousands of sheep pituitaries and, by an arduous and lengthy procedure, succeeds in isolating a few milligrams of a previously unreported hormone. It is doubtful that the work will be verified independently. Nev- Id. ertheless, a manuscript reporting it would be unconditionally accepted for publication." 145. Dr. Johnson noted that such a policy would leave an author "with the options of disclosing His decision, all too often, will be to not publish." Id. the trade secret or not publishing. 146. Letter from William F. Harrington to Nathan 0. Kaplan (Feb. 3, 1981) (on file at Yale Law Journal) (discussing suggestions of ad hoc committee). The Yale Law Journal [Vol. 97: 177 But a closer reading of the record and a broader view of the context changes the picture somewhat. First, the letters illustrate that even where the norms are relatively clear, scientists often violate them for reasons un- related to commercial incentives. The most widely shared norm seems to be that publishing scientists should enable subsequent investigators to rep- licate their research results. All of the scientists invoked this replicability norm in their analysis of the functions served by the policy. Even the scientists who favored modifying or eliminating the availability require- ment nonetheless felt that the Journal should at least encourage availabil- ity of strains to promote replicability of results. Strikingly, however, most of the committee members observed that academic scientists as well as commercial scientists often depart from the norms-including even the 1 4 7 widely-shared replicability norm-in practice. This suggests that the problem does not originate with commercial interests or intellectual prop- erty incentives. Indeed, the arguments raised against the policy focused on tensions within the scientific community rather than conflicting incentives from the outside. No one adopted the argument raised at the editorial board meet- ing that the policy unfairly benefited "free riders" at the expense of pub- 14 8 lishing scientists, perhaps because of a sense that "free riding" on the published work of other scientists is an integral feature of scientific re- search. Nonetheless, several of the scientists cited the desire of individu- als to retain the exclusive benefits of a discovery for themselves as an ex- planation, if not a justification, for departures from the norms and for the 1 50 the policy. of enforcing difficulty consequent If the financial rewards of intellectual property protection tempt scien- 147. See, e.g., Brown letter, supra note 132; Friedman letter, supra note 140; Nathans letter, supra note 137; Rabinowitz letter, supra note 137. Even Dr. Kaplan, who came closest to claiming that the replicability norm is in fact descriptive of the behavior of academic scientists, went only so far as to state that "the academic people . . . are, from my experience, usually willing to give up the organism." Kaplan letter, supra note 125 (emphasis added). In a personal interview, Dr. Kaplan conceded that academic scientists sometimes fail to observe this norm in practice. He gave me a copy of a facetious form letter, then circulating as a joke among scientists, declining a request for a sample of a strain and setting forth a list of possible excuses to be selected or deleted as appropriate. The letter reads in part as follows: I regret to inform you that we are unable to send the strains you wish because* we have a graduate student who is studying this particular strain and it would be unfair to distribute it at this time/the problem you outline does not seem very significant/we have just concluded your particular study/we suspect you are lying about the purpose for which you want the strains/ we suspect you are the referee with whom we had considerable trouble recently. I might add that I, personally, am not in favour of withholding mutant strains. *Delete as appropriate The joke itself shows tension within the academic community over adherence to the norms. (This letter is on file at The Yale Law Journal.) 148. See supra note 117. 149. R. MERTON, supra note 18. 150. See, e.g., Brown letter, supra note 132; Johnson letter, supra note 140; Nathans letter, supra note 137; Rabinowitz letter, supra note 137. No one took up the suggestion raised at the editorial board meeting that it was particularly unfair to compel academic scientists to make their materials available to profit-making organizations. See supra note 117. 19871 Biotechnology Research tists to depart from scientific norms, the records of the JBC controversy suggest that the temptation may simply reinforce conflicting incentives within the reward structure of science itself. Even among purely academic scientists, the norms may never have had the force that the community would like them to have. Given that professional recognition rewards only the first researcher to arrive at a particular result, it is not surprising that scientists are reluctant to give their materials to their research 1 5 1 competitors. Moreover, when the JBC episode is considered in light of subsequent responses to the same problem by other scientific journals, it is not at all clear that scientific norms are losing ground. In the seven years since the JBC first confronted this issue, a number of prominent scientific journals have adopted policies similar to the JBC's. This spate of new policies addressing the issue may signify a number of things. It may indicate that the scientific community believes its norms are threatened and in need of fortification. Perhaps increasing violations of these norms have led jour- nals to respond with express policy statements. Another possibility is that there was no need for specialized policies for biological materials until the recent increase in recombinant DNA research. Thus, the biotechnology boom may have made it necessary to clarify the implications of longstand- 1 53 materials are involved. norms where biological ing Whatever the explanation, the recent record hardly suggests that tradi- tional scientific norms are crumbling in the face of intellectual property incentives as commercial interest in biotechnology research becomes more 1 5 4 pervasive. Although there are signs of tension over adherence to the norms, commercial pressures appear to be only a new aggravation of a familiar problem. 151. See supra text accompanying note 111. 152. The following journals now have express editorial policies requiring or encouraging the availability of biological materials referred to in manuscripts: 46 CELL No. 6 (Sept. 12, 1986) ("Infor- mation for Contributors"); 117 DEV. BIOLOGY No. 1 (Sept. 1986) ("Information for Authors"); 165 J. BACTERIOLOGY ii (Jan. 1986) ("Instructions to Authors"); 132 J. GEN. MICROBIOLOGY No. 7 (July 1986) ("Instructions to Authors"); 57 J. VIROLOGY ii (Jan. 1986) ("Instructions to Authors"); 153 VIROLOGY No. 2 (Sept. 1986) ("Notice to Authors"). 153. In fact, several journals have more general policies requiring authors to give sufficient infor- mation about their materials and methods to permit reproduction of their experiments, although they make no specific reference to biological materials. Such policies are set forth in, e.g., 156 ANALYTICAL BIOCHEMISTRY ix (1986) ("Instructions to Authors"); 56 BIOLOGY OF THE CELL 93 (1986) ("In- structions to Authors"); 32 CLINICAL CHEMISTRY 2 (Jan. 1986) ("Information for Authors"); 97 J. BIOCHEMISTRY iii (1985) ("Instructions to Authors"); 23 J. CLINICAL MICROBIOLOGY iii (Jan. 1986) ("Instructions to Authors"); 137 J. IMMUNOLOGY No. 6 (Sept. 15, 1986) ("Information for Contributors"); 187 J. MOLECULAR BIOLOGY No. 2 (Jan. 20, 1986) ("Instructions to Authors"); 232 SCIENCE xi (1986) ("Information for Contributors"). 154. See generally OFFICE OF TECHNOLOGY ASSESSMENT, COMMERCIAL BIOTECHNOLOGY: AN INTERNATIONAL ANALYSIS (1984). Indeed, a recent survey of over 1200 faculty members at 40 major universities in the United States found that biotechnology researchers with industrial support both publish at higher rates and patent more frequently than colleagues without such support. Blumenthal, Gluck, Louis, Stoto & Wise, supra note 3, at 1361. The Yale Law Journal [Vol. 97: 177 C. Patent Rights in Biotechnology Inventions Without patent protection, scientists who wanted to preserve intellec- tual property rights in their discoveries would have to resort to either ac- tual secrecy or trade secrecy. As we have seen, these two strategies conflict with scientific norms and rewards in that they preclude disclosure of the 5 5 discoveries for which protection is sought.' Patent law interferes with scientific communication less than either actual secrecy or trade secrecy because it is premised on disclosure rather than secrecy. Patent law pro- motes disclosure in two ways. First, in order to obtain a patent, an inven- tor must include in the patent application a full description of the inven- tion and how to make and use it. Although this application is confidential while pending,1 it becomes freely available to the public once a patent issues. Second, by granting a property right that survives disclosure, patent law removes an obstacle to disclosure of inventions that would otherwise have to be kept secret in order to preserve their commer- 1 58 value. cial On the other hand, patent law may be less consistent with scientific norms than secrecy protection in a different respect. Patent law grants a much more extensive right to exclude others from the use of a discov- ery,' thus potentially foreclosing further research to a greater degree than secrecy. A patent provides a remedy against all who make, use, or sell the invention without permission, however innocently, whereas trade secret law provides a remedy only against wrongdoers.' ' For exam- ple, independent discovery is not a defense to a patent infringement suit, although it is a defense to a claim for misappropriation of trade secrets. Because of the broad scope of the exclusive rights granted by a patent, patent law may offer considerably stronger protection than trade se- crecy-as well as greater potential disruption of ongoing research-in competitive research fields. By providing such broad exclusive rights, patent law may aggravate preexisting conflict between scientific norms and the reward structure of science. Scientific norms call for the dedication of discoveries to the re- search community without restrictions on subsequent use. But the reward structure of science grants recognition to the first scientist to make a dis- covery, creating an incentive to foreclose other scientists from doing com- petitive research. Ordinarily, the need to publish in order to gain recogni- 155. See supra text accompanying notes 89-92. 156. See 35 U.S.C §§ 111, 112 (1982). 157. See 35 U.S.C. § 122 (1982). 158. Indeed, once they are assured of an enforceable property right, patent owners may have an incentive to disseminate information about their inventions in order to attract potential licensees. See Kitch, The Nature and Function of the Patent System, 20 J.L. & ECoN. 265, 278 (1977). 159. See infra notes 207-42 and accompanying text. 160. See infra notes 211-13 and accompanying text. 161. See supra notes 86-88 and accompanying text. 1987] Biotechnology Research tion constrains the incentive to keep discoveries out of the hands of research competitors. Although publication may help competitors to make future discoveries, the reward structure impels scientists to make this con- tribution to the scientific community in order to gain current recognition for discoveries already made. But if patent law gives inventors the right to exclude others from using their discoveries in competitive research even after publication, it could potentially upset this balance of incentives. A system of exclusive rights that survives disclosure could thus undermine the mechanisms of the scientific community for gaining free access to new knowledge for use in subsequent research. I now turn to a more detailed consideration of the disclosure obligations and exclusive rights provided under the patent laws. 1. Disclosure A comparison of the patent disclosure requirements for microorganism- related inventions with the scientists' analysis of the functions of the JBC's availability policy shows considerable congruence between patent law policy and scientific norms concerning disclosure. If anything, patent law appears more rigorous than the scientific community with respect to the substance of its disclosure requirement. The two systems differ signifi- cantly, however, regarding the timing of disclosure. Patent disclosure is unlikely to occur until after a patent issues,"' which is often years later than disclosure to the scientific community would otherwise occur. a. Enablement Sections 111 and 112 of the Patent Act require an inventor to set forth in her patent application a written description of the invention, an expla- nation that will "enable" a skilled practitioner to "make and use" the invention, and the "best mode" contemplated by the inventor of "carrying out" her invention. This disclosure requirement serves two functions. First, it ensures that the public will inherit the full benefit of the inven- tion upon expiration of the patent term. Second, it permits the PTO to determine that the applicant has in fact developed an operative, useful embodiment of the invention. The first function parallels scientific norms by calling upon patent applicants, like publishing scientists, to dedicate 162. The statement in text assumes that the inventor has filed only a United States patent appli- cation. If the applicant also seeks foreign patent protection under the patent laws of other nations or international patent protection under the Patent Cooperation Treaty or the European Patent Conven- tion, the application will be published 18 months after the date of first filing. Patent Cooperation Treaty, done June 19, 1970, art. 21, 28 U.S.T. 7647, 7666-67, T.I.A.S. No. 8733, at 7666-67; European Patent Convention, October 5, 1973, art. 93, translated in K. HAERTEL, EUROPEAN PAT- ENT CONVENTION 19, 62-63 (1980), reprinted in 78 PAT. & TRADEMARK REv. 31, 39 (1980). See generally R. SALIWANCHIK, supra note 67, at 139-46. 163. 35 U.S.C. §§ 111, 112 (1982). [Vol. 97: 177 The Yale Law Journal their inventions to the public. The second function parallels the scien- calling upon patent applicants, like publish- tists' replicability standard by ing scientists, to supply sufficient information and materials to demon- strate to a knowledgeable audience that they have in fact achieved what claim. they materials, the patent dis- When an invention involves unique biological themselves if the written closure obligation extends to the materials called for expressly in the statute is insufficient to enable rep- description lication of the applicant's invention. Although microorganisms were not commonly considered patentable subject matter until the Supreme Court decided Diamond v. Chakrabarty, patent protection has long been available for processes using microorganisms and for the products of such 1 6 7 by processes. With the commercial development of antibiotics produced of bacterial strains, inventors sought a way to make enabling disclosures such inventions. Where the bacteria involved were generally available, the by explaining how to procure the necessary disclosure could be made 1 6 materials. ' But where a new or rare strain was involved, the only way to enable others to make or use (in patent law parlance, to "practice") the invention was to supply the strain. For these cases, patent attorneys de- vised the procedure of depositing samples of bacterial strains with recog- nized depositories from which the strains would become available to the public upon issuance of the patent.' 164. See infra notes 172-75 and accompanying text. 165. See infra note 175 and accompanying text. 166. 447 U.S. 303 (1980). The majority in Chakrabarty cited three prior instances in which patents had issued on microorganisms. The first was a patent issued to Louis Pasteur on yeast in and 1968. Id. at 314 1873, and the second and third were patents issued on microogranisms in 1967 n.9. 167. Note, Microorganisms and the Patent Office: To Deposit or Not to Deposit, That Is the Question, 52 FORDHAM L. REv. 592, 595-96 (1984). 168. See, e.g., Merck & Co. v. Chase Chem. Co., 273 F. Supp. 68 (D.N.J. 1967). 169. See, e.g., Ex parte Kropp, 143 U.S.P.Q. (BNA) 148 (P.T.O. Bd. App. 1959); Ex parte Davisson, 118 U.S.P.Q. (BNA) 520, 521 (P.T.O. Bd. App. 1957); Levy & Wendt, Microbiology and Applications, 37 J. PAr. a Standard Format for Infra-Red Absorption Spectra in Antibiotic Patent OFF. Soc'y 855, 855-59 (1955). It is interesting to compare the statutory disclosure requirements under the Plant Patent Act, 35 Act, 7 U.S.C. §§ 2321-2583 (1982), U.S.C. §§ 161-164 (1982), and the Plant Variety Protection those under the Patent Act. The Plant Patent Act, first promulgated in 1930, ch. 312, 46 Stat. with 376, extends patent protection to new varieties of asexually reproduced plants. The disclosure require- § 112, except for the following qualification: "No plant patent ment is the same as that in 35 U.S.C. section 112 of this title if the description is as com- shall be declared invalid for noncompliance with plete as is reasonably possible." 35 U.S.C. § 162. This provision has been construed to permit disclo- sures failing to satisfy the section 112 requirement of a description of the manner and process of making the invention. See In re Greer, 484 F.2d 488, 490-91 (C.C.P.A. 1973); cf. Funk Bros. Seed Co. v. Kalo Inoculant Co., 333 U.S. 127, 133-35 (1948) (Frankfurter, J., concurring) (arguing that patent for mixed culture of bacterial inoculants should be denied not because combination of bacterial strains was "work of nature," but because, inter alia, description of invention was inadequate for Patent Act allows patent protection for some failure to identify particular strains used). The Plant with section 112. inventions that it might not be possible to describe in accordance 1542, provides The Plant Variety Protection Act, enacted in 1970, Pub. L. No. 91-577, 84 Stat. by the Department of Agriculture for novel varieties of sexually patent-like protection administered reproduced plants. The disclosure provisions expressly call for both a written description and a Biotechnology Research 1987] The Court of Customs and Patent Appeals first addressed the adequacy 7 0 of microorganism deposits under section 112 in In re Argoudelis.1 The discussion of the function of deposits in this and subsequent cases parallels the analysis of the JBC's availability policy by the scientists. In re Argoudelis predates the JBC controversy by ten years. It involved a patent application on new antibiotic compounds and a process for pre- paring the compounds from microorganisms. The applicants deposited samples of their microorganisms in the permanent culture collection of the United States Department of Agriculture before the filing date of their patent application and referred to this deposit in the specification. At the request of the applicants, the depository agreed not to release the organ- ism without their written permission until a patent issued. Reversing the PTO, the Court of Customs and Patent Appeals ap- proved the use of a deposit to "disclose" a microorganism obtained from nature. The court held that section 112 did not require that the general public have access to the culture before the patent issued, reasoning that as long as the materials are available to the PTO during the pendency of the patent application, the disclosure is sufficient to demonstrate comple- tion of the invention as of the filing date. Since patent applicants in other fields need not make their patent disclosures public unless and until a patent issues, the court felt that no earlier public disclosure should be 1 7 required for microorganism-related inventions. In a concurring opinion, Judge Baldwin emphasized the dual role played by enablement in patent law. First, enablement ensures that the 1 7 2 public will receive its quid pro quo for the patent monopoly. Second, it demonstrates that as of the filing date the applicant is entitled to a pat- 1 7 3 ent. The first function does not call for public disclosure until the pat- deposit: An application for a certificate recognizing plant variety rights shall contain: (2) A description of the variety setting forth its novelty and a description of the genealogy and breeding procedure, when known .... (3) A declaration that a viable sample of basic seed necessary for propagation of the variety will be deposited and replenished periodically in a public repository .... 7 U.S.C. § 2422. In requiring a deposit in addition to a written description, this more recent statute follows the approach of the courts in interpreting section 112 of the Patent Act. 170. 434 F.2d 1390 (C.C.P.A. 1970). 171. Id. at 1393. 172. The first aspect of that role is to provide the assurance that the public will, in fact, receive something in return for the patent grant. This consideration is, of course, the full and complete disclosure of how to make and use the claimed invention. Thus, the patent adds a measure of worthwhile knowledge to the public storehouse . ... Id. at 1394. 173. Most, if not all, of the determinations involving the patentable merits of a claimed inven- tion vis-a-vis the prior art require that the application filing date be regarded as the prima facie date of invention . . . . Because the filing date is so important in determining patent rights, it is essential that there be no question that, at the time an application for patent is filed, the invention claimed therein is fully capable of being reduced to practice (i.e., that no technological problems, the resolution of which would require more than ordinary skill and reasonable time, remain in order to obtain an operative, useful embodiment). The Yale Law Journal [Vol. 97: 177 7 4 application contain ent issues." It does, however, require that the patent enough information for the PTO to determine whether the public will receive an enabling disclosure as of the issue date. The second function requires only that the disclosure demonstrate successful invention as of the filing date. Since Argoudelis' specification, including the deposited orga- nisms, was sufficient to show successful invention by the filing date, and the disclosure would become available to the public upon issuance of a 112.175 satisfied section specification patent, the Id. at 1395 (emphasis in original). 174. It should be apparent, however, that this first aspect of the enabling disclosure require- ment of section 112, requires only that the adequacy of the teaching disclosure be measured as of the issue date of the patent. There is no sense in making an applicant publicly disclose any part of his invention, much less its very essence, before he has been assured that he will obtain the protection he is seeking in return for that disclosure. Id. at 1394-95 (footnote omitted). 175. Judge Baldwin later reiterated this analysis in a majority opinion in In re Hawkins, 486 F.2d 569 (C.C.P.A. 1973), a case that did not involve microorganisms. The same approach has been followed in subsequent cases analyzing the use of microorganism deposits to satisfy the enablement requirement. In Feldman v. Aunstrup, 517 F.2d 1351 (C.C.P.A. 1975), the court extended the reasoning of In re Argoudelis to a case involving the deposit of a microorganism in a foreign depository. At the time the patent application was filed, the deposit was subject to a restriction that the organism could not be released without the depositor's consent. This restriction was removed after the filing date and before issuance of a patent. Id. at 1352-53. The Court held that the specification was sufficient to determine whether the inventor was entitled to a patent as of the filing date, and that release of the restrictions ensured that the specification would be fully enabling to the public upon issuance of a patent. Id. at The court also held that deposit in a foreign country could satisfy the enablement requirement. Id. at 1356. This issue is now moot because the United States has become a signatory to the Budapest Treaty on the International Recognition of the Deposit of Microorganisms for the Purposes of Patent Procedure, Apr. 28, 1977, 32 U.S.T. 1241, T.I.A.S. No. 9768. See 961 Official Gazette Pat. Off. 21-36 (Aug. 23, 1977). The Budapest Treaty establishes minimum requirements for maintaining an international depository for microorganisms and authorizes such depositories to receive and store de- posits and dispense samples in accordance with the Treaty and the patent laws of the signatory na- tions. 999 Official Gazette Pat. Off. 2 (Oct. 7, 1980). See generally Meyer, Problems and Issues in Depositing Microorganisms for Patent Purposes, 65 J. PAT. OFF. Soc'y 455 (1983) (discussing de- pository requirements under international treaties and section 112). The PTO responded to the Argoudelis decision by promulgating an administrative rule concerning the deposit of microorganisms. See MANUAL OF PATENT EXAMINING PROCEDURE, § 608.01 (c) (5th ed. 1983, rev. 1985). Under this rule, whenever an invention requires the use of a microorganism that is not known and readily available to the public, the applicant must deposit the organism not later than the filing date "in a depository affording permanence of the deposit and ready accessibility thereto by the public if a patent is granted." Id. The conditions of deposit must assure access to the deposit by the PTO during pendency of the patent application and unrestricted availability of the culture to the public upon issuance of the patent. The PTO has construed this rule strictly. For example, the PTO invoked the rule to invalidate an agreement by a depository to require that persons requesting samples of a microorganism identify the party making the request and undertake not to transmit the organism to third parties. Restrictions Voided on Availability of Deposited Yeast Strains to Public, 30 Pat. Trademark & Copyright J. (BNA) No. 732, at 104 (1985). In In re Lundak, 773 F.2d 1216 (Fed. Cir. 1985), the Court of Appeals for the Federal Circuit rejected the PTO's deposit rule insofar as it requires deposit before the filing date. That case involved a patent application on biological materials useful for producing immunoglobulins. The inventor ap- parently filed a patent application in the mistaken belief that the materials had already been placed in a depository. In fact, the materials were not deposited until one week after the filing date. The PTO rejected the application for failure to meet the enablement requirement of section 112. The Federal Circuit reversed. Following the reasoning of In re Argoudelis, the court noted that the materials were available from the inventor during the pendency of the patent application under a provision of the patent statute giving the PTO the right to request "specimens or ingredients" from the applicant. See 1987] Biotechnology Research This analysis appears congruent with scientific norms. Issuance of a patent brings about publication of the disclosure in the specification."' In patent law, as in science, a researcher may keep her discoveries secret until publication. Thereafter, she must surrender to the public an ena- 1 7 7 bling disclosure that permits replication of her claims. Indeed, in re- quiring access to materials through a source other than the inventor, pat- ent law goes beyond the requirements of the scientific community.' Two further respects in which patent law may impose tougher disclo- sure requirements than the scientific community deserve mention. The first concerns disclosures that permit replication only after further experi- mentation. The second concerns disclosures that, while fully enabling, conceal from the public the "best mode" contemplated by the inventor of practicing the invention. b. Undue Experimentation Recall that a number of the scientists consulted by the JBC thought that authors should not be required to supply newly-engineered organisms described in their publications, as long as the authors give enough infor- mation about the method of synthesis to permit other investigators to re- 35 U.S.C. § 114 (1982). The court deemed recourse to this statutory provision sufficient to satisfy any need for access by the PTO to verify successful invention. Although availability of materials from the inventor may be sufficient to allow the PTO to deter- mine constructive reduction to practice, it is plainly not a substitute for deposit in satisfying the en- ablement requirement once a patent issues. Otherwise, an inventor could effectively prolong the patent monopoly by restricting access to the materials after the patent expires. Cf White Consol. Indus. v. Vega Servo-Control, 713 F.2d 788 (Fed. Cir. 1983) (disclosure that incorporates trade secret available only by license from patentee fails to satisfy enablement requirement). Inventors could thereby cir- cumvent the terms of the bargain offered by the patent system: a monopoly on the invention for a limited term in exchange for dedicating the invention to the public thereafter. See Brulotte v. Thys Co., 379 U.S. 29, 31 (1964). The PTO recently issued an advance notice of proposed rulemaking relating to the deposit of bio- logical materials for patent purposes. 52 Fed. Reg. 34,080 (1987), reprinted in 34 Pat. Trademark & Copyright J. (BNA) No. 846, at 485 (Sept. 10, 1987). The proposed rule would update PTO practice in light of the decision of the Federal Circuit in In re Lundak; the Budapest Treaty on the Interna- tional Recognition of the Deposit of Microorganisms for the Purposes of Patent Protection, see supra; and the PTO's decision to extend patent protection to plants outside the Plant Patent Act in In re Hibberd, 227 U.S.P.Q. (BNA) 443, 444 (P.T.O. Bd. App. & Int. 1985). 176. In some countries, patent applications are published before a patent issues. See supra note and sources cited therein; see also Meyer, supra note 175. 177. The timing of disclosure may nonetheless be quite different under the two systems. See infra notes 201-06 and accompanying text. 178. Recall that a promise of availability from the author was an acceptable substitute for deposit under the JBC policy. See supra note 113 and accompanying text. Under some circumstances, scien- tists may be reluctant to make their organisms available through a depository. For example, scientists may not trust the depository to maintain and distribute viable culture samples that will continue to show the characteristics described in prior publications. See Hampar, Patenting of Recombinant DNA Technology: The Deposit Requirement, 67 J. PAT. OFF. Soc'y 569, 605-07 (1985). Of course, patent law does not prohibit patent holders from maintaining their own cultures in addition to those main- tained by the depository and making samples from their own laboratories available to corroborate their research claims. Thus, although patent law may go beyond scientific norms in requiring that organisms be made available through a depository, it should not deter scientists who are willing to make their organisms available from seeking patent protection. [Vol. 97: 177 The Yale Law Journal peat what was done. Patent law also permits a disclosure that demands as long as replica- some further effort in order to replicate the invention, ' 8 0 The specification is, tion does not require "undue experimentation." however, defective if replication requires "more than the skill of the art" ' ' to which the invention pertains or "an inordinate amount of work." The courts and the PTO have analyzed the "undue experimentation" standard in the context of microorganism-related inventions in Tabuchi v. 8 2 Jackson.2 Parte and Ex Nubel Tabuchi v. Nubel involved a priority contest between two inventors of a and isocitric acid using yeasts fermentation method for making citric acid The case turned on whether a specification that of the genus Candida. failed to supply a depository accession number for the particular Candida applicant satisfied the enabling disclosure requirement strain used by the of section 112.184 The applicant argued that it would have been possible to select a citric-acid-producing strain from publicly available Candida strains without undue experimentation by following the process disclosed in the specification. To prove his point, the applicant hired an expert in the field who read the application, selected a strain from Candida strains according to available to the public in depositories, and made citric acid the process described in the specification, all in a matter of fifteen working days. The court concluded that the disclosure "would not compel those of skill in the art to perform undue experimentation in selecting a ordinary '85 suitable strain" and that the enablement requirement was therefore satisfied. the discovery of three distinct microorganism Ex Parte Jackson involved same new antibiotic. The patent ap- strains, each of which produced the plicant classified the strains as belonging to a new species and deposited each of the three strains in a recognized depository. The application in- cluded claims to the process of producing the antibiotic from any microor- ganisms belonging to the same species as the deposited strains, ' and to the deposited strains or any mutations the same process employing 179. See, e.g., Friedman letter, supra note 140; Nathans letter, supra note 137; Rabinowitz letter, note 137; supra notes 137-45 and accompanying text. supra 180. W.L. Gore & Assocs. v. Garlock, Inc., 721 F.2d 1540, 1557 (Fed. Cir. 1983), cert. denied, 469 U.S. 851 (1984); White Consol. Indus., Inc. v. Vega Servo-Control, Inc., 713 F.2d 788, 791 (Fed. Cir. 1983). 181. 2 D. CHISUM, supra note 7, § 7.03141[a], [bi. 182. 559 F.2d 1183 (C.C.P.A. 1977). 183. 217 U.S.P.Q. (BNA) 804 (P.T.O. Bd. App. 1982). 184. Tabuchi claimed priority on the basis of a Japanese patent application. The issue before the court was whether this application met the enablement requirement of section 112 of the Patent Act, 35 U.S.C. § 112 (1982). Section 119 of the Patent Act, 35 U.S.C. § 119 (1982), allows inventors to provided that the claim United States patent priority on the basis of a filing date in a foreign country, application satisfies the disclosure requirements of section 112. Kawai v. Metlesics, 480 F.2d foreign 880 (C.C.P.A. 1973). 185. Tabuchi, 559 F.2d at 1189. 186. See 217 U.S.P.Q. at 805. Biotechnology Research 1 7 A divided Board of Appeals affirmed the rejection of the former thereof. claim, but reversed the rejection of the latter claim. The majority reasoned that the specification did not enable one skilled in the art to isolate other antibiotic-producing strains of the same species from nature without un- it was possible that the deposited strains due experimentation-indeed, the other were the only strains in existence suitable for the process."" On microorganisms are a common occurrence and "can be hand, mutations of 1 8 9 by a variety of known procedures." By deposit- intentionally produced ing the three strains, the patent applicant enabled skilled practitioners to derive mutant strains, and he was therefore entitled to extend his process mutants. claims to the use of such Here again patent law parallels and reinforces the scientific commu- nity's replicability norm. The "undue experimentation" cases on one hand reinforce the disclosure obligation and on the other hand limit the scope of invention. A patent applicant may what may be claimed as a patentable not, by withholding materials, leave too much work to be done by those invention. Conversely, the who would use the specification to practice the applicant is limited to claiming as her invention such advances as patent has fairly put within reach of the public. Patent applicants her disclosure to achieve who limit their disclosures so as to require inordinate effort replication thereby limit the scope of their monopoly. c. Best Mode Section 112 of the Patent Act further requires that the specification "set forth the best mode contemplated by the inventor" of carrying out the Unlike the enablement requirement, which is an objective invention."" focusing on the quality of the inventor's disclosure, the "best standard is a subjective standard focusing on the inventor's own mode" requirement knowledge and belief at the time the patent application is filed. ' The "best mode" requirement goes beyond the scientific community's repli- makes an enabling disclosure of an opera- cability norm. An inventor who tive version of her invention will not receive a valid patent if she knows of and conceals a better way to practice the invention at the time the applica- tion is filed.' Once again, patent law gives teeth to scientific norms through its tougher enforcement mechanisms. An applicant's failure to disclose the 187. See id. at 805-06. 188. Id. at 807-08. 189. Id. at 806. 190. 35 U.S.C. § 112 (1982). 191. 2 D. CHISUM, supra note 7, § 7.0511], [2]. 192. See, e.g., In re Gay, 309 F.2d 769, 772 (C.C.P.A. 1962) ("Manifestly, the sole purpose of [the best mode requirement] is to restrain inventors from applying for patents while at the same time concealing from the public preferred embodiments of their inventions which they have in fact conceived."). The Yale Law Journal [Vol. 97: 177 "best mode" of practicing the invention is unlikely to come to the attention of the PTO in the course of the ex parte patent examination procedure. But it may well surface in the course of discovery in a subsequent patent infringement suit. Should it appear at that time that the inventor con- cealed her "best mode" when she filed her patent application, the patent 1 9 invalid. be held will d. Patent Disclosure and Scientific Norms Aside from considerations of timing to be examined below,"' the disclo- sure requirements of patent law seem to fortify scientific norms to a de- gree that surpasses the willingness of the scientific community to enforce its own norms. Patent law reinforces the scientific community's interest in the replicability of claimed discoveries by requiring that a patent applicant make a disclosure of her invention that will enable persons skilled in the art to which the invention pertains to make and use the invention without undue experimentation. Patent law reinforces the scientific community's interest in having new discoveries dedicated to the public by requiring that the patent application set forth the best mode contemplated by the inven- tor of practicing the invention. Publication of the specification upon issu- ance of the patent and dedication of the claimed invention to the public upon expiration of the patent term further promote these interests. Scien- tists who protect their discoveries under the patent laws should therefore be at least as likely as scientists influenced solely by the rewards of science to disclose their inventions to the scientific community in a form that satis- fies scientific norms. In some circumstances, scientists may choose to forgo patent protection in order to circumvent enforcement of these norms. For example, scientists might want to restrict access to their organisms in order to keep the or- ganisms out of the hands of their research or commercial competitors. If the scientific community is willing to publish the research of these scien- tists while allowing them to keep their organisms to themselves, it may be possible for scientists to earn recognition in the scientific community while undermining the community's interest in replicability and public dedica- tion of research claims. Even if the publication is enough to forfeit legal trade secrecy protection, a strategy of publishing research results while maintaining the materials as actual secrets might permit scientists to by- pass the strictures of both patent law and scientific norms. They could thus earn scientific rewards and preserve the benefits of exclusivity with- out ever giving an enabling disclosure of their inventions. 193. See Biggart, supra note 52, at 2-73. 194. E.g., Engelhard Indus. v. Sel-Rex Corp., 253 F. Supp. 832, 836-37 (D.N.J. 1966), afl'd, 384 F.2d 877 (3d Cir. 1967). 195. See infra text accompanying notes 201-06. Biotechnology Research It might seem that the protection of a patent would be adequate to persuade inventors to forgo the narrower and more fragile protection of secrecy, since a patentee has the right to enjoin others from making, using, or selling the invention during the term of the patent even after disclo- sure. But the patent law exchange of disclosure for a right to exclude is not always attractive where secrecy is a feasible alternative. For one thing, an invention that can be exploited in secrecy by the inventor can often also be exploited in secret by a patent infringer. The unrestricted deposit of an organism necessary for replicating such an invention may "enable" infringements that cannot be detected. ' Cell lines and microor- ganisms can be propagated so easily that it is difficult under the best of circumstances to keep track of who has access to samples.' An inventor may choose instead to keep the organism to herself. Another reason why biotechnology inventors may be unwilling to de- posit their organisms in exchange for patent protection is that a patent may not be broad enough to protect inventors from research competition. For reasons more fully explored below, the use of patented inventions in competitive research may not constitute patent infringement. Scientists who make an enabling disclosure of their inventions for patent purposes may find that they have enabled their research competitors to gain scien- tific credit for subsequent discoveries to which the originator of the mater- ials might otherwise have eventually claimed priority. If the patenting sci- entist does not gain a corresponding right to exclude research competitors from using the materials, she might prefer to keep her materials to herself and forego patent protection. Of course, withholding research materials after publication to avoid being forestalled by research competitors violates scientific norms. But the reward structure of science sets up an incentive to maximize claims to priority of discovery that conflicts with these norms. And if patent law enforces its enablement requirement more zealously than the scientific community enforces its own norms, scientists may find 196. See Kewanee Oil Co. v. Bicron Corp., 416 U.S. 470, 489-90 (1974). 197. Of course, not all inventions may be exploited in secret. If an invention may be readily reverse-engineered, patent protection will be more attractive than futile efforts at secrecy. See R. SALIWANCHIK, supra note 67, at 10-11. 198. For example, suppose a company has created a new recombinant organism producing a pro- tein that is indistinguishable from human insulin. The recombinant product differs from previously- sold insulin solely in its lower cost of production. The company has a choice of either patent protec- tion or trade secrecy. If the company chooses patent protection, competitors will gain access to the organism once the patent issues. Since the recombinant insulin is identical to insulin produced by other means, it will be impossible to detect whether competitors are infringing the patent. The PTO's prohibition against agreements between inventors and depositories permitting inventors to monitor who gets access to deposits, see supra note 175, can only increase the likelihood of undetectable infringement. 199. The ease of propagation can also make it difficult to keep microorganisms as trade secrets or actual secrets. See supra notes 84-85 and accompanying text. 200. See infra notes 226-42 and accompanying text. Journal The Yale Law [Vol. 97: 177 that secrecy provides a more congenial strategy for maximizing their pro- fessional and commercial rewards. e. Timing of Disclosure There is one important respect in which the incentives arising from scientific rewards and those arising from the patent laws diverge: The timing of disclosure may be quite different for scientists seeking patent protection than for scientists seeking scientific recognition alone. Both the patent laws and the scientific community reward priority of invention, ' new claims. Nonetheless, thus placing a premium on prompt disclosure of scientists motivated solely patent applicants are likely to publish later than by scientific norms and rewards. One reason for this difference is that basic research discoveries might be ripe for recognition in the scientific 2 °2 community before they are ripe for patent protection. But if a scientist, guided by scientific norms and rewards, publishes results of research that has not yet yielded a patentable invention, she may be unable to obtain a 2 03 patent for inventions developing out of this research at a later date. of patent protection may therefore lead scientists to defer publica- Pursuit tion until their research has yielded patentable inventions, prolonging the they have filed patent applications and period of secrecy until after communication and progress. thereby slowing down scientific Once an inventor has a patent application on file, she may publish her for patent protection on research results without impairing her prospects 20 4 those inventions already claimed. Even then, however, patent applicants who are uncertain whether their inventions are patentable may choose to 2 0 5 defer publication until a patent actually issues. Publication prior to that thereby forfeits secrecy time is a risky proposition because the inventor 20 6 patent rights. A patent protection without any assurance of obtaining 201. R. MERTON, supra note 24, at 286-324; see 35 U.S.C. §§ 101, 102 (1982). 202. This disparity in timing is in part a consequence of patent law mechanisms for limiting patent protection to applied technology. See supra notes 38-47 and accompanying text. 203. There are two ways in which early publication could jeopardize the scientist's eligibility for a publication that discloses a subsequently claimed invention bars United patent protection. First, is filed within one year of the publication date. 35 States patent protection unless the application U.S.C. § 102. Foreign patent protection may be forfeited immediately upon publication. Second, pub- lication could have untoward consequences for subsequent patentability because of the nonobviousness requirement of 35 U.S.C. § 103. An invention is generally deemed to have been made on the date the inventor files a patent application, and the patentability of the invention turns on whether the inven- tion is obvious in light of the "prior art" as of that date. The prior art includes the applicant's own publication. See, e.g., Mass. Inst. of Technology v. AB Fortia, 227 U.S.P.Q. (BNA) 428, 431-32 publication discloses enough information to make the (Fed. Cir. 1985). If the inventor's own prior later-claimed invention obvious, the inventor will not be able to obtain a patent. See Chisum, Sources of Prior Art in Patent Law, 52 WASH. L. REv. 1 (1976). may impair the inventor's prospects for obtaining patent protection on 204. But such publication broader claims at a later date. See In re Ruscetta, 255 F.2d 687 (C.C.P.A. 1958). 205. If the inventor also applies for foreign patent protection, publication may occur 18 months after the application is first filed. See supra note 162. 206. In the United States patent system, secrecy may be preserved prior to the issuance of a patent. 35 U.S.C. § 122 (1982). 1987] Biotechnology Research application may languish in the PTO for several years before a patent ultimately issues. Thus, the combination of uncertainty as to the patent- able status of research discoveries and time lags in the PTO is likely to slow down the dissemination of information in the scientific community. In sum, notwithstanding the congruence between the substantive disclo- sure requirement of the patent laws and scientific norms, patent disclosure may occur considerably later than disclosure motivated solely by scientific norms and rewards. 2. Exclusive Rights Disclosure, of course, is only one side of the patent law bargain. Even if patent applicants make broader disclosures of their inventions than scien- tists motivated only by scientific norms and rewards, they do so in ex- change for a seventeen-year monopoly. Publication traditionally marks the end of exclusivity in science, but public disclosure through issuance of a patent marks the beginning of exclusive rights in the patent system. An "enabling" disclosure may be small consolation to a scientific community that is disabled from using it. Although the patent statute permits patent holders to exclude others from using their inventions, they do not always choose to do so. Depend- ing on the nature of the invention and its market, a patentee may find it advantageous to exploit the invention directly as a monopolist, to suppress the invention entirely, or to license others to exploit the invention on an 2 7 exclusive or non-exclusive basis in exchange for royalties. Licensing of patented inventions may pose special problems where the inventions are useful primarily for subsequent research rather than for commercial applications. The serendipitous nature of research discoveries may make it difficult to place a value on the right to use a patented inven- tion before the outcome of a research project is known. Moreover, most scientific research relies on many prior discoveries. If researchers needed to obtain licenses from each of the prior inventors on whose work they 20 8 build, the royalties and transaction costs could mount quickly. There are several reasons why a patentee might elect not to extend licenses to research competitors. The patentee might want to supress the 207. For an economic analysis of the exploitation of patent monopolies, see F. SCHERER, INDUS- TRIAL MARKET STRUCTURE AND ECONOMIC PERFORMANCE 379-99 (1970); see also Machlup, An Economic Review of the Patent System, Study No. 15 of the Subcomm. on Patents, Trademarks, and Copyrights of the Senate Comm. on the Judiciary, 85th Cong., 2d Sess. (1958) (economic overview of patent system as whole). Nonexclusive licensing of patented inventions minimizes interference with subsequent research. Patent rights on the basic process for recombinant DNA, owned jointly by Stanford and the Univer- sity of California, are licensed on a nonexclusive basis to all interested companies. See M. KENNEY, supra note 1, at 257-59. 208. See generally Gordon, Fair Use as Market Failure: A Structural and Economic Analysis of the Betamax Case and Its Predecessors, 82 COLUM. L. REV. 1600 (1982). The Yale Law Journal [Vol. 97: 177 invention in order to bolster its position in a related market, as happens when corporations build up extensive patent portfolios to prevent potential 2 9 licensees from using competitive technologies. " Or the patentee might fear that licensed use of the patented invention in further research will facilitate inventing around the patent, thereby undermining the future 2 1 ° value of the patent. Finally, the patentee might wish to preserve exclu- sivity in subsequent research in order to maximize future claims to prior- ity of discovery for purposes of both intellectual property and scientific credit. The extent to which patent law actually permits patent holders to en- join the use of their inventions by others in subsequent research is un- clear. On its face, the statute seems to forbid any unauthorized use of the patented invention. Section 154 of the Patent Act by its express terms grants to patentees "the right to exclude others from making, using, or selling the invention throughout the United States" during the patent term. " Section 271 of the Patent Act supplies a corresponding definition of patent infringement: "Except as otherwise provided in this title, who- ever without authority makes, uses or sells any patented invention, within the United States during the term of the patent therefor, infringes the patent. 112 According to the cases, this seventeen-year period of exclusive use is the consideration paid by the public for the disclosure of an inven- tion which it will be free to use once the patent expires. "' The enabling disclosure in the patent specification assures that the public will receive its 21 4 of the patent. the bargain upon expiration end of This commonplace formulation of the patent law bargain is indeed at odds with the traditional mechanisms of the scientific community for building upon prior research. Scientific research promotes the dissemina- tion of new knowledge to research competitors by awarding recognition to those who make their claims freely available to the community. Patent law, in contrast, promotes investment in innovation by providing exclusive rights in inventions. ' But other features of the patent statute indicate that this exclusivity is 209. F. SCHERER, supra note 207, at 392. Patents for inventions made by universities and small businesses in the course of federally-sponsored research cannot be held back from exploitation in this way. See supra notes 102-08 and accompanying text. 210. For an example of how this might occur, see infra note 242. 211. 35 U.S.C. § 154. 212. 35 U.S.C. § 271(a). A patentee may obtain an injunction against infringement "in accor- dance with the principles of equity," as well as damages "in no event less than a reasonable royalty" and up to three times the amount of actual damages in the discretion of the court. 35 U.S.C. §§ 283, 284. Reasonable attorney fees are also available to the prevailing party "in exceptional cases." 35 U.S.C. § 285. 213. See, e.g., Brulotte v. Thys Co., 379 U.S. 29 (1964); United States v. Dubilier Condenser Corp., 289 U.S. 178 (1933). 214. E.g., In re Argoudelis, 434 F.2d 1390, 1394-95 (C.C.P.A. 1970) (Baldwin, J., concurring). 215. See supra notes 17-28 and accompanying text. 216. Kewanee Oil Co. v. Bicron Corp., 416 U.S. 470, 480 (1974). Biotechnology Research not absolute. If the public had absolutely no right to make, use, or sell the patented invention until the end of the patent term, it would be somewhat puzzling to require that the patentee give the public an enabling disclo- sure of the invention at the beginning of the patent term. The require- ment of early disclosure suggests that certain uses of patented inventions during the patent term do not constitute patent infringement. Perhaps the most important function of early disclosure is to facilitate improvement of the patented invention or invention around the patent. Disclosure during the patent term also permits verification of the suffi- 2 18 ciency of the disclosure. Other practitioners in the field of the invention may be in a better position than the patent examiner to determine 2 19 whether a disclosure is enabling. Since an insufficient disclosure makes the patent invalid and unenforceable, those who have a use for the pat- ented technology will be motivated to uncover defects in the specification in order to avoid liability to the patentee. If an insufficient disclosure were not made public until the expiration of the patent term, the public would already have tolerated "the embarrassment of an exclusive pat- 2 2 ent" ' without receiving its quid pro quo. Disclosure at the outset of the patent term also reveals the scope of the patent claims, thereby giving notice to potential infringers (and to potential licensees) of the metes and bounds of the patent monopoly. Finally, early disclosure may facilitate "the Progress of Science and the useful Arts" by giving the public ac- 2 25 cess to those parts of the specification that the patentee does not claim. At least the first two of these functions-inventing around the patent and checking the sufficiency of the disclosure-seem to contemplate some 217. See generally Yarway Corp. v. Eur-Control USA, 775 F.2d 268, 277 (Fed. Cir. 1985) ("negative incentive" to "design around" other patented products is benefit of patent system); State Indus. v. A.0. Smith Corp., 751 F.2d 1226, 1236 (Fed. Cir. 1985) (same). The prevalence and social value of inventing around patented technologies are subject to dispute. Compare Turner, The Patent System and Competitive Policy, 44 N.Y.U. L. REv. 449, 455 (1969) (patent forces competitors to invest resources in finding noninfringing ways of achieving same result as patented invention, which may in many instances be wasteful if it diverts resources from other un- solved problems) with Adelman, The Supreme Court, Market Structure and Innovation: Chakrabarty, Rohm and Haas, 27 ANTrrusT BULL. 457, 464 (1982) (efforts to invent around pat- ent are unlikely to occur unless competitor and patentee have different views of cost of developing alternative technology and will not result in net loss to society because patented invention might never have been made without patent system). 218. See infra notes 232-34 and accompanying text. 219. Cf generally Loom Co. v. Higgins, 105 U.S. 580, 586 (1882) (patent specifications should be addressed to those skilled in art); Mowry v. Whitney, 81 U.S. (14 Wall.) 620, 644 (1871) (same). 220. See, e.g., Grant v. Raymond, 31 U.S. (6 Pet.) 218, 239 (1832). 221. Graham v. John Deere Co., 383 U.S. 1, 9 (1965) (quoting Thomas Jefferson). 222. United States v. Dubilier Condenser Corp., 289 U.S. 178, 186-87 (1933); Grant v. Ray- mond, 31 U.S. (6 Pet.) 218, 247 (1832). See Evans v. Eaton, 20 U.S. (7 Wheat.) 356, 434 (1822) (specification warns innocent user of patented invention against infringement and permits public to check whether patentee is entitled to patent). 224. U.S. CONsr. art. I, § 8, cl. 2; see supra text accompanying notes 38-39. 225. See generally Alexander Milburn Co. v. Davis-Bournonville Co., 270 U.S. 390 (1926) (dis- closure of unclaimed invention in specification of issued patent puts such invention in public domain). The Yale Law Journal [Vol. 97: 177 "use" of the disclosed invention during the term of the patent, if only in the laboratory. And while the statute itself does not expressly permit any use of patented technology during the term of the patent, the cases have recognized an exception to patent infringement for "experimental use" of an invention. It is difficult to discern the scope of this exception with any degree of precision. Since experimental use becomes an issue only in infringement actions, judicial pronouncements on its reach address situations where pat- entees have found a defendant's activities sufficiently annoying to be worth the trouble of pursuing a lawsuit. This factor has undoubtedly skewed the universe of experimental use decisions toward cases that impli- cate commercial interests. Within this universe, the experimental use de- 2 7 rarely sustained. raised and fense is frequently The experimental use doctrine was first expounded by Justice Story in dictum in an 1813 decision, Whittemore v. Cutter. " In that case, the trial court had entered judgment on a verdict for the plaintiff in a patent in- fringement suit. On appeal, the defendant argued that the jury had been erroneously instructed "that the making of a machine fit for use, and with a design to use it for profit, was an infringement of the patent right." Justice Story noted that the instruction was actually favorable to the de- fendant because it limited infringement liability to the making of a ma- chine with an intent to use it for profit. He supposed that this limitation "was adopted by the court from the consideration, that it could never have been the intention of the legislature to punish a man, who constructed such a machine merely for philosophical experiments, or for the purpose of ascertaining the sufficiency of the machine to produce its described 23 0 effects. 226. Roche Prods. v. Bolar Pharmaceutical Co., 733 F.2d 858, 862-63 (Fed. Cir.), cert. denied, 469 U.S. 856 (1984); Pfizer, Inc. v. International Rectifier Corp., 217 U.S.P.Q. (BNA) 157, 160-61 (C.D. Cal. 1982). See generally 4 D. CHISUM, supra note 7, at § 16.03(1); Bee, Experimental Use as an Act of Patent Infringement, 39 J. PAr. OFF. Soc'y 357 (1957); Hantman, Experimental Use as an Exception to Patent Infringement, 67 J. PAT. OFF. Soc'y 617 (1985). 227. The Court of Appeals for the Federal Circuit has called the defense "truly narrow," Roche Products v. Bolar Pharmaceutical Co., 733 F.2d at 863, and one commentator has argued that the defense is inconsistent with the language of the statute and the policies of patent law. See Bee, supra note 226, at 359, 377. 228. 29 F. Cas. 1120 (C.C.D. Mass. 1813) (No. 17,600); see also Sawin v. Guild, 21 F. Cas. 554 (C.C.D. Mass. 1813) (No. 12,391). The doctrine continued to appear solely in dicta for the next 45 years. See, e.g., Jones v. Pearce, Webster's Pat. Case 122, 124-25 (K.B. 1832); Byam v. Bullard, 4 F. Cas. 934, 935 (C.C.D. Mass. 1852) (No. 2,262). It finally found its way into a published holding in 1858, when the Circuit Court for the Southern District of New York approved a jury instruction that the use of a patented invention "merely for experiment, and not with a view to profit" would not give rise to infringement liability. Poppenhusen v. New York Gutta Percha Comb Co., 19 F. Cas. 1059, 1063 (C.C.S.D.N.Y. 1858) (No. 11,283). 229. 29 F. Cas. at 1121. The defendant objected to this instruction on two grounds: first, that as a matter of statutory construction the making of a machine without using it could not constitute patent infringement; and, second, that the mere making of a patented machine without actually using it causes no damage and thus could not be the basis for infringement liability. Justice Story rejected both of these arguments. 230. Id. at 1121 (emphasis added). Biotechnology Research 1987] Although subsequent cases elaborated the doctrine in confounding 23 1 ways, Justice Story's original articulation seems, at least in theory, to meet the needs of the research science community. The first exemption from liability for making a patented device-"philosophical experi- ments"-appears to protect "pure" research from patent infringement lia- bility, while the second-"ascertaining the sufficiency of the machine to produce its described effects"-appears to permit replication of research claims. Elsewhere in the opinion Justice Story elaborates a patent law correla- tive to the scientific community's interest in replicability. In objecting to a jury instruction that omissions from the patent specification rendering it nonenabling would not invalidate the patent unless the omissions were deliberate, the defendant argued that a materially defective specification should provide a defense to an infringement action even without proof of intentional deception. Story felt compelled to reject the argument under the language of the 1793 Patent Act, but was clearly sympathetic none- theless to the defendant's position: [T]he monopoly is granted upon the express condition, that the party shall make a full and explicit disclosure, so as to enable the public, at the expiration of his patent, to make and use the invention or improvement in as ample and beneficial a manner as the patentee himself. If therefore it be so obscure, loose, and imperfect, that this cannot be done, it is defrauding the public of all the consideration, upon which the monopoly is granted .... It was probably with a view to guard the public against the injury arising from defective specifications, that the statute requires the let- Later that year Justice Story reiterated his conception of the experimental use doctrine in the case of Sawin v. Guild, 21 F. Cas. 554 (C.C.D. Mass. 1813) (No. 12,391), as follows: [Tihe making of a patented machine to be an offence. . . must be the making with an intent to use for profit, and not for the mere purpose of philosophical experiment, or to ascertain the verity and exactness of the specification. In other words, that the making must be with an intent to infringe the patent-right, and deprive the owner of the lawful rewards of his discovery. Id. at 555 (citation omitted). 231. For example, some later decisions suggest that the rationale for the experimental use defense is that experimental use causes no harm. See, e.g., Byam v. Bullard, 4 F. Cas. 934, 935 (C.C.D. Mass. 1852) (No. 2,262); Albright v. Celluloid Harness-Trimming Co., I F. Cas. 320, 323 (C.C.D.N.J. 1877) (No. 147). Since Justice Story rejected the argument that making a patented ma- chine without causing harm cannot give rise to liability in Whittemore, see supra note 229, this cannot have been his rationale for the defense. 232. Section 6 of that Act reads in part as follows: Provided always,. . . That the defendant in such action shall be permitted to plead. . . that the specification, filed by the plaintiff, does not contain the whole truth relative to his discov- ery, or that it contains more than is necessary to produce the described effect, which conceal- ment or addition shall fully appear to have been made, for the purpose of deceiving the public. Patent Act of 1793, ch. 11, 1 Stat. 318 (1793), quoted in 6 D. CmisuM, supra note 7, at app. 10-3 (emphasis in original). Despite the language of the statute, the Supreme Court subsequently held in Grant v. Raymond, 31 U.S. (6 Pet.) 218 (1832), that a defendant in an infringement suit could avoid liability by showing that the specification was defective even without any intent on the part of the patentee to deceive the public. The Yale Law Journal [Vol. 97: 177 ters patent to be examined by the attorney general, and certified to be in conformity to the law, before the great seal is affixed to them. In point of practice, this must unavoidably be a very insufficient se- curity, and the policy of the provision, that has changed the common law, may be very doubtful. In recognizing an exception to patent infringement for use of the inven- tion "for the purpose of ascertaining the sufficiency of the machine to produce its described effects," Justice Story may have been trying to cor- rect for this deficiency in the 1793 statute. If, as Justice Story thought, examination by the attorney general prior to patent issuance is "a very insufficient security" against defective disclosures, giving the public a right to check the adequacy of the disclosure without liability for infringement provides additional assurance that the public has received its quid pro quo for the patent monopoly. The most effective scrutinizers of the patent dis- closure are likely to be potential infringers-or at least potential users of the patented invention. These people have the greatest knowledge of the relevant field and care the most whether the claimed invention works as described. The experimental use doctrine thus gives the patentee's compet- itors a right during the patent term to vindicate the public's interest in having an enabling disclosure. Although a few decisions have allowed the experimental use defense where the defendant simply wanted to see if the claimed invention worked, " hardly any cases have allowed the defense to excuse otherwise infringing activities that were conducted "merely for the purpose of philo- ' 23 5 sophical experiments. When commercial companies have claimed that their uses of patented inventions were for "philosophical experiments," 23 6 the defense has generally been denied. 233. 29 F. Cas. at 1122. 234. See, e.g., Dugan v. Lear Avia, Inc., 55 F. Supp. 223, 229 (S.D.N.Y. 1944); Akro Agate Co. v. Master Marble Co., 18 F. Supp. 305, 333 (N.D.W. Va. 1937). But cf. National Meter Co. v. Thomson Meter Co., 106 F. 531 (C.C.S.D.N.Y. 1900) (use of patented invention to see if it works constitutes threatened infringement by business competitor sufficient to support injunction). 235. These cases have recently been reviewed in Hantman, supra note 226. 236. See, e.g., United States Mitis Co. v. Carnegie Steel Co., 89 F. 343, 351 (C.C.W.D. Pa. 1898) (noting that defendant's use of patented invention "was a commercial use, extending over a period of several months, and involved a very large product"); Bonsack Machine Co. v. Underwood, 73 F. 206, 211 (C.C.E.D.N.C. 1896) (noting that machine "has not been made simply as an experi- ment, but has been used for profit, that is, for the purpose of selling the [defendant's] patent"); Al- bright v. Celluloid Harness-Trimming Co., 1 F. Gas. 320, 323 (1877) (No. 147) (holding use of patented invention in trial manufacture "is a technical infringement, and is sufficient to authorize an injunction restraining ... future use" but not sufficient for award of damages); Poppenhusen v. Falke, 19 F. Cas. 1048, 1049 (C.C.S.D.N.Y. 1861) (No. 11,279) (noting that defendants "are rivals of the complainant in the very business to which his patents relate"). The Robinson patent treatise, published in 1890, took the position that an experimental use of a patented invention in order to use or sell the products of the experiment or with a view to adapting the invention to the experimenter's business is "hostile to the interests of the patentee" and thus constitutes infringement. W. ROBINSON, THE LAW OF PATENTS FOR USEFUL INVENTIONS § 898 (1890). Consistent with this view, defendants who used patented inventions in the course of producing products for sale or performing services for customers have met with little success. E.g., Spray Refrig- 19871 Biotechnology Research A rare case allowing the experimental use defense for the use of a pat- ented invention in research is Ruth v. Stearns-Roger Manufacturing 2 37 Co. In that case, the defendant made unauthorized sales of patented machines and replacement parts to several customers, including the Colo- rado School of Mines. In awarding damages, the court adopted the recom- mendations of a special master, who had distinguished between direct in- fringement for the sale of whole machines and contributory infringement 2 3 8 for the sale of parts. The special master had calculated damages for direct infringement on all sales of infringing machines, including those sold to the Colorado School of Mines. In calculating damages for contrib- utory infringement, however, the special master had excluded sales of re- placement parts to the Colorado School of Mines on the ground that these parts "were for use in laboratory machines used for experimental pur- 23 9 poses, and consequently did not contribute to an infringing use."1 Be- eration Co. v. Sea Spray Fishing, Inc., 322 F.2d 34, 36-37 (9th Cir. 1963) (use of patented apparatus and method for freezing fish "for the purpose of experimentation as to the desirability of using this method" constituted patent infringement where use occurred in course of commercial fishing operation of kind which patent was designed to serve); Imperial Chem. Indus., PLC v. Henkel Corp., 545 F. Supp. 635, 656 (D. Del. 1982) (defendant's manufacture of 1200 pounds of patented chemicals for "pilot plant use" infringed patent where defendant supplied samples to potential customers and made it known in market that chemicals were commercially available); Northill Co. v. Danforth, 51 F. Supp. 928, 929 (N.D. Cal. 1942) ("Defendant's experiments were evidently not made for philosophi- cal or amusement purposes but were made in connection with his business as a manufacturer and salesman of anchors."); Sprout, Waldron & Co. v. Bauer Bros., 26 F. Supp. 162, 169 (S.D. Ohio 1938) ("The defendant cannot escape on the ground of experimental use where the machines were used to operate upon customers' products in the ordinary course of business."); Radio Corp. of Am. v. Andrea, 15 F. Supp. 685, 687 (E.D.N.Y. 1936), modified on other grounds, 90 F.2d 612 (2d Cir. 1937) (assembling parts of patented combination to test them was infringement where assembly was "not a scientific research or an engineering inquiry" but rather "a step which the defendants appar- ently deemed necessary in the manufacture and sale of their product"); Pairpearl Prods., Inc. v. Jo- seph H. Meyer Bros., 58 F.2d 802, 804 (D. Me. 1932) (use of patented process for extracting pearl essence infringed patent notwithstanding defendant's claim that use was experimental as incident to search for new agent, where pearl essence extracted through patented process was sold to customers); Cimiotti Unhairing Co. v. Derboklow, 87 F. 997, 999 (C.C.E.D.N.Y. 1898) (use of patented machine for customers "in the ordinary course of business" was patent infringement, notwithstanding defend- ant's claim that he was "experimenting" with machines in order to see if he could improve them). Two recent decisions have denied the experimental use defense to generic drug manufactuiers who performed clinical tests of patented drugs prior to expiration of the patents, reasoning that such testing relates to a commercial activity or the operation of a business. Roche Prods., Inc. v. Bolar Pharmaceu- tical Co., 733 F.2d 858, 862 (Fed. Cir. 1984), cert. denied, 469 U.S. 856 (1984); Pfizer, Inc. v. International Rectifier Corp., 217 U.S.P.Q. (BNA) 157, 160-62 (C.D. Cal. 1982). Congress responded to the decision in Roche v. Bolar by enacting the Drug Price Competition and Patent Term Restoration Act of 1984, Pub. L. 98-417, 98 Stat. 1585. That Act amends § 271 of the Patent Act to exempt from infringement liability the use of a patented drug solely to develop the information that must be submitted under federal law in order to obtain permission to sell the drug. See 35 U.S.C.A. § 271(e) (West Supp. 1987). 237. 13 F. Supp. 697 (D. Colo. 1935). 238. The unlicensed sale of a patented machine directly infringes the patent regardless of the use to which the machine is put by the buyer, but the sale of parts does not constitute direct infringement unless the parts are themselves the subject of patent claims. Selling unpatented parts with the inten- tion that they be used in a patented machine might nonetheless constitute contributory infringement, but only if the buyer's use of the parts itself constitutes direct infringement. See 35 U.S.C. § 271(a), (c) (1982); Aro Mfg. Co. v. Convertible Top Replacement Co., 365 U.S. 336, 341-42 (1961); 4 D. CHISUM, supra note 7, § 17.03. 13 F. Supp. at 703. The Yale Law Journal [Vol. 97: 177 yond this conclusory statement, the only information offered in the opin- ion as to the nature of the use is that "[tihese machines were all used in 24 0 the laboratory and were cut up and changed from day to day." The case offers no guidance for distinguishing a philosophical experiment from an infringing use."" The scope of the experimental use defense remains unclear, making it difficult to assess the degree to which the exclusive rights of patent holders might impede the progress of science. Justice Story's original formulation of the experimental use doctrine seems in principle to accommodate the interests of the scientific community in replicating and building upon new knowledge in basic research. The first prong of Justice Story's experimen- tal use privilege, permitting "philosophical experiments," is not well de- fined in the cases, but it seems to permit subsequent researchers to use the patented invention at least in traditional basic research with no commer- cial implications. The second prong, allowing the defense where the use is for the purpose of testing the adequacy of the specification, parallels the scientific community's interest in replication of published claims. The difficulty arises in trying to apply these principles when basic re- search has commercial value. As patent protection for basic research dis- coveries becomes more commonplace, and as the use of patented inven- tions in research becomes more of a threat to the interests of patent holders, the experimental use defense is likely to be litigated in new con- texts requiring careful analysis of its purposes and scope. The optimal extent of the experimental use defense cannot be deter- mined without attention to its likely effects on the scientific community. Too narrow a defense could stifle basic research and impair the commu- nity's mechanisms for validating and building upon new knowledge. Too broad a defense could cause industrial sponsors to lose interest in biotech- nology research or to rely on secrecy in lieu of patent protection. There may be no way to avoid both of these potential problems completely. The case for allowing the defense appears strongest where the subse- 24 2 quent user is attempting to devise alternatives to the patented invention. 240. Id. 241. In a recent review of experimental use cases, Ronald Hantman approved the Ruth decision while questioning its precedential value: This case is unlike any other experimental use case because it is the only one in which the exception is claimed for use at a university. Few would deny the experimental use exception for research on patented technology performed at a university in furtherance of its educational function. Because of the limited scope of the research in Ruth, the case probably has little precedential value for experimental use cases not involving university research. Hantman, supra note 226, at 633. 242. The facts of a pending lawsuit, Scripps Clinic & Research Found. v. Genentech, Inc., 3 U.S.P.Q.2d (BNA) 1481 (N.D. Cal. 1987), illustrate how the issue might arise. The lawsuit concerns technology for producing Factor VIII:C, a protein used in treating hemophilia. Plaintiffs hold patents on methods for recovering Factor VIII:C from human blood plasma in purified and concentrated form and on the purified and concentrated protein itself. The patented methods are costly, however, and involve a risk of transmitting disease. These difficulties make the protein a likely candidate for pro- 19871 Biotechnology Research In such a case, the interests of the research user are congruent with the interests of the public and the scientific community in advancing the state of human knowledge. The patent holder, by contrast, has an interest in prolonging the period in which the public is dependent on the patented technology. If the patentee sees the research user as a competitor rather than a customer, she may refuse to license the invention. Without an ex- perimental use defense, it is possible that no one would be able to build on the inventor's discovery until the patent expired. At the other end of the spectrum, the case for allowing the defense ap- pears weakest where the research user is essentially consuming a patented invention in an unrelated research effort-for example, by using a pat- ented laboratory machine. To allow such a user to avoid infringement liability on the ground that the machine was used in research would evis- cerate patent protection for technologies used primarily in research labo- ratories. The user falls squarely within the patentee's market, and there is no reason to suppose that the patentee would refuse to negotiate a license. If patent protection permits scientists to exclude the scientific commu- nity from making use of published research discoveries, it could under- mine a critical mechanism of the scientific community for facilitating the progress of science. Without the exclusive rights granted by the patent system, the reward structure of science encourages scientists to dedicate their results to the scientific community by requiring that they publish in order to gain recognition. This system promotes the interests of the com- munity in the validation and extension of new knowledge by overriding the individual scientist's contradictory incentive to hoard her findings. Dedication of new discoveries to the community necessarily entails facili- tating the work of one's competitors, but this is the price individual scien- tists must pay for current recognition. As the records of the JBC contro- versy show, the norms of dedication and replicability cease to be self- enforcing when it is possible for scientists to earn recognition without fully dedicating their results to the community-for example, when scien- tists can publish their research claims without supplying otherwise un- available materials to their competitors. By granting exclusive rights after publication, patent law may offer scientists another mechanism for earn- duction through recombinant DNA technology. Genentech is charged with infringing plaintiffs' patents in the course of cloning the gene for Factor VIII:O. Genentech used purified Factor VIII:C made by the plaintiffs' process to determine the se- quence of amino acids in the protein, and used this information to identify and clone in a recombinant organism the human gene responsible for manufacturing Factor VIII:C. The district court held that Genentech's use of the patented invention in its laboratories infringed the patent. Id. at 1486. Since Genentech evidently never asserted the experimental use defense, the court did not reach the question of whether the "philosophical experiments" prong of the experimen- tal use doctrine privileges research use of a patented invention by a commercial company. Had Genentech raised the issue, this might have been an appropriate case for allowing the experimental use defense (at least as to the method claims), since Genentech's goal was to invent around the patent, rather than to exploit the patented technology itself. The Yale Law Journal [Vol. 97: 177 dedication for sev- ing recognition while deferring payment of the price of enteen years. It remains to be seen whether the experimental use defense will develop into a means of reconciling scientific norms with the patent monopoly. D. An Illustration of the Conjunction A recently-settled dispute illustrates some of the problems that arise when scientists influenced by scientific norms and rewards seek to pre- serve intellectual property rights. The dispute erupted between competing teams of scientists at the National Cancer Institute (NCI) in Bethesda, Maryland, and the Pasteur Institute in Paris. Both teams of scientists had been doing research related to acquired immune deficiency syndrome (AIDS), separately and collaboratively, for several years. Their competi- tive efforts to secure intellectual property rights and scientific credit for invention of a test to detect the presence of antibodies to the AIDS virus in human blood became the subject of two legal proceedings: a breach of contract action in the United States Claims Court" and a patent interfer- 2 4 ence proceeding in the PTO. NCI scientists, under the direction of Dr. Robert Gallo, began studying 24 5 human retroviruses before the onslaught of the AIDS epidemic. In 1978, the scientists isolated the first human retrovirus, which they called human T-cell lymphotropic virus because of its affinity for T-lympho- cytes. The same group discovered a second human retrovirus, associated with leukemia, in 1982. They renamed the first human retrovirus 4 6 HTLV-I and called the new retrovirus HTLV-II. This early research laid the groundwork for the subsequent discovery that AIDS is caused by 2 4 7 yet another retrovirus. Scientists at the Pasteur Institute began studying AIDS under the di- rection of Dr. Luc Montagnier in late 1982. In February of 1983, they isolated a virus from the lymph nodes of a patient suffering from persis- tent lymphadenopathy. They called their virus lymphadenopathy- associated virus or LAV. This virus ultimately proved to be associated with AIDS. During this period, the NCI and Pasteur Institute scientists 243. Institut Pasteur v. United States, No. 730-85 (Cl. Ct. Dec. 12, 1985) (complaint reprinted as attachment C to Lipsey, Brezner & Majofis, Protecting Trade Secrets in Biotechnology, in PROTECT- ING TRADE SECRETS, supra note 83, at 807, 861); see Institut Pasteur v. United States, 10 Cl. Ct. 304 (1986), rev'd, 2 U.S.P.Q.2d (BNA) 1048 (Fed. Cir. 1987). 244. See Norman, AIDS Patent Negotiations Break Down, 232 SCIENCE 819 (1986); Altman, Ruling May Settle AIDS Test Dispute, N.Y. Times, Apr. 30, 1986, at 10 col. 3. its genetic information in the form of ribonucleic acid 245. A retrovirus is a virus that carries called (RNA) rather than deoxyribonucleic acid (DNA) and reproduces itself by using an enzyme reverse transcriptase to make a DNA strand corresponding to its own viral RNA. This DNA then integrates itself into the chromosomes of the host cell and uses the biological mechanisms of the host to replicate the virus. See Gallo, The First Human Retrovirus, SCi. AM., Dec. 1986, at 88. 246. Id. 247. See Gallo, The AIDS Virus, SCl. AM., Jan. 1987, at 47. 1987] Biotechnology Research 24 8 began communicating with each other and sharing biological materials. The NCI scientists supplied materials from their prior work with ret- roviruses, which allowed the Pasteur Institute scientists to identify LAV as a retrovirus distinct from either HTLV-I or HTLV-II. The Pasteur Institute published this finding in May of 1983.49 Despite the propensity of the LAV virus to kill the cells in which it was grown, the Pasteur Institute scientists were able to propagate the virus in small quantities. By August of 1983, they had developed a test that de- tected antibodies to LAV in twenty percent of serum samples taken from AIDS patients. They filed patent applications for their test in London in September of 1983 and in the United States in December of 1983, 25 1 priority. filing for purposes of date of the European claiming the In the meantime, the NCI scientists were also attempting to isolate virus from AIDS patients. While they found evidence of retrovirus in 25 2 AIDS patients, they were having difficulty growing the virus in culture. The Pasteur group twice sent samples of their virus to NCI, but the NCI scientists assert that they were unable to propagate the Pasteur vi- rus. Finally, in November of 1983, the NCI group found a cell line that could survive infection with the AIDS virus, making it possible to grow the virus in significant quantities. Now the NCI group was able to characterize the virus, which they called HTLV-III, and to develop a more sensitive test to detect antibodies to the virus in blood samples. NCI published its findings in May of 1984.255 A patent was issued on NCI's test kit in May of 1985.56 The terms under which these materials were shared, and the actual use made of the shared materials, are disputed. Institut Pasteur v. United States, No. 730-85 (Cl. Ct. Dec. 12, 1985). 249. BarrE-Sinoussi, Chermann, Rey, Nugeyre, Chamaret, Gruest, Dauguet, Axler-Blin, Vizinet-Brun, Rouzioux, Rozenbaum & Montagnier, Isolation ofa T-Lymphotropic Retrovirusfrom a Patient at Risk for Acquired Immune Deficiency Syndrome (AIDS), 220 SCIENCE 868 (1983). 250. Norman, Patent Dispute Divides AIDS Researchers, 230 SCIENCE 640, 642 (1985). 251. See supra note 184. 252. Norman, AIDS Priority Fight Goes to Court, 231 SCIENCE 11 (1986). 253. Complaint, Institut Pasteur v. United States, No. 730-85 (Cl. Ct. Dec. 12, 1985). 254. Norman, supra note 252, at 12. 255. Gallo, Salahuddin, Popovic, Shearer, Kaplan, Haynes, Palker, Redfield, Oleske, Safai, White, Foster & Markham, Frequent Detection and Isolation of Cytopathic Retroviruses (HTLV-III) From Patients With AIDS and at Risk for AIDS, 224 SCIENCE 500 (1984); Popovic, Sarnagadharan, Read & Gallo, Detection, Isolation, and Continuous Production of Cytopathic Retroviruses (HTLV- Ill)from Patients with AIDS and pre-AIDS, 224 SCIENCE 497 (1984); Shaw, Broder, Essex & Gallo, Human T-Cell Leukemia Virus: Its Discovery and Role in Leukemogenesis and Immunosuppression, 30 ADV. INTERN. MED. 1 (1984); Shupbach, Popovic, Gilden, Gonda, Sarnagadharan & Gallo, Sero- logical Analysis of a Subgroup of Human T-Lymphotropic Retroviruses (HTLV-IlI) Associated with AIDS, 224 SCIENCE 503 (1984). Almost two years later, controversy over priority of discovery was aggravated by the disclosure that electron micrograph pictures of the AIDS virus accompanying these publications depicted cells infected with LAV supplied by the Pasteur Institute rather than cells in- fected with HTLV-III isolated by the NCI scientists. See Gilden, Gonda, Sarnagadharan, Popovic & Gallo, HTLV-Il Legend Correction, 232 SCIENCE 307 (1986) (letter to editor); Norman, A New Twist in AIDS Patent Fight, 232 SCIENCE 308 (1986). 256. U.S. Patent 4,520,113. NCI also filed a patent application, Serial No. 602,946, on a method of continuous production of HTLV-III retrovirus from patients with AIDS and pre-AIDS. This application is still pending. See In re Gallo, 231 U.S.P.Q. (BNA) 496 (Comm'r Pat. 1986). The Yale Law Journal [Vol. 97: 177 The Pasteur Institute initiated legal proceedings to establish its priority of invention for purposes of both patent rights and scientific credit. In the patent interference proceeding, the Pasteur Institute asserted that its re- searchers were the first to isolate the AIDS virus and to develop an AIDS antibody test kit, and that it was therefore entitled to the patent rights on this invention. In the Court of Claims action, the Pasteur Institute alleged that NCI had violated express and implied contracts with Pasteur in pat- enting and licensing its own test kit and denying appropriate recognition to the Pasteur Institute scientists. Crucial to this latter action was the extent of NCI's actual use of the virus supplied by the Pasteur Institute in developing the NCI test kit. The Pasteur Institute asserted that it sup- plied the virus to NCI solely for "research purposes" and not for "indus- trial purposes," and that NCI's HTLV-III "is, or is identical to, the LAV strain first isolated by Pasteur." President Ronald Reagan and French Prime Minister Jacques Chirac announced the settlement of both pro- 2 55 by Chirac. during a visit to Washington on March 31, 1987, ceedings This dramatic conclusion to the conflict indicates the extraordinary degree of public interest in AIDS research. This controversy illustrates a number of the themes developed in the foregoing analysis of the interaction of scientific norms with intellectual property law. First, apart from intellectual property concerns, it shows the dilemma confronting scientists who seek current scientific recognition in the course of a competitive research effort. Each team of scientists bene- fited from work done by the other, and, conversely, each team facilitated the work of the other through its own disclosures. These disclosures un- doubtedly earned scientific recognition for the publishing scientists and advanced the progress of science, but they may also have given a boost to the competition. The dispute also highlights the difference in timing between a discovery that is ripe for scientific recognition and an invention that is ripe for pat- ent protection. The Pasteur Institute developed a more primitive test kit first; NCI developed a more sensitive test kit later. In the patent interfer- ence proceeding, NCI contended that Pasteur was not entitled to a patent because the test kit that was the basis for Pasteur's patent application was useless without a method for growing the virus in quantity, and that NCI rather than Pasteur was the first to develop such a method. This argu- ment may encourage biotechnology inventors to defer publication and dis- semination of their discoveries until they are confident that their inven- 257. Complaint, Institut Pasteur v. United States, No. 730-85 (Cl. Ct. Dec. 12, 1985). 258. Barnes, AIDS Patent Dispute Settled, 236 Sci. 17 (1987). Under the settlement agreement, the parties will share patent rights and jointly contribute 80% of their royalties on the antibody test kits to an AIDS research foundation. Both teams of scientists will be recognized as joint inventors of the test kits. As part of the settlement, Drs. Gallo and Montagnier signed a chronology of key discov- eries in AIDS research. Biotechnology Research 1987] tions are patentable. Had the Pasteur and NCI scientists followed such a strategy of interim secrecy, it might have taken additional months or years for either team to develop its AIDS screening test. The Pasteur Institute's efforts to restrict dissemination of LAV even after publication of its discovery demonstrate that the scientific community may enforce its disclosure norms less rigorously than the patent law en- forces its disclosure requirement. In this regard, it may be significant that Science magazine, where much of the research of both groups was pub- lished, did not at that time have a policy requiring publishing authors to make their materials available to other researchers, although it has since 2 59 adopted such a policy. Without such a requirement, the Pasteur Insti- tute was able to restrict access to LAV to its research collaborators and to prevent its use for "industrial purposes" or its dissemination to other scientists without their express permission. Restrictions on access to the virus would not have been possible if a patent had already issued on an invention requiring use of the virus." On the other hand, a patent might have given either team the right to enjoin the other from using the patented invention in subsequent research. Laboratory use of the virus by scientists seeking to develop an antibody test kit for ultimate sale in the market arguably falls outside the scope of 2 6 the experimental use defense. ' Fortunately, in this case neither team of scientists sought to enjoin the other's research. It is clear in retrospect, however, that the research of each team represented a threat to both the commercial and the professional interests of the other. Both of these inter- ests provide an incentive to impede the research efforts of competitors. But while the scientific community does not empower inventors to enjoin the use of prior discoveries in subsequent research, patent law may. CONCLUSION The commercialization of basic research discoveries in biotechnology and the availability of patent protection for such discoveries have gener- ated concern in the scientific community about the impact of intellectual 259. The Science policy now provides: "When a paper is accepted for publication in Science, it is understood by the editors that any materials necessary to verify the conclusions of the experiments reported will be made available to other investigators under appropriate conditions." 232 SCIENCE xi (1986) ("Information for Contributors"). The Pasteur Institute's restrictions on NCI's use of LAV might constitute "appropriate conditions" for making its materials available to other investigators within the meaning of the new Science policy. 260. See supra notes 175-78 and accompanying text. The dispute also illustrates the risk of unde- tectable infringement of patents involving the use of microorganisms. The Pasteur Institute contended that NOI used LAV to develop its test kit, while NCI insisted that it isolated its own virus. Subse- quent studies indicated that the two viruses differ in their genetic sequences by approximately 1.5%, compared to a difference of approximately 6% relative to an AIDS virus isolated by another labora- tory. Norman, HTLV-Ill and LAV: Similar, or Identical?, 230 SCIENCE 643 (1985). Since AIDS viruses can undergo spontaneous mutation, this genetic similarity is consistent with but does not prove use of LAV by NCI. 261. See supra notes 234-41 and accompanying text. Journal The Yale Law [Vol. 97: 177 in biotechnology- property law on the behavior of research scientists of intellectual property rules in light of scientific related fields. Analysis reveals substantial parallels between the two sys- norms and incentives requirements of patent law fortify scientific norms tems. The disclosure research results in replicable form. To the extent promoting publication of go beyond scientific norms, mandating broader that these requirements necessary to earn recognition in the scientific commu- disclosure than is in favor of se- nity, some inventors may choose to forgo patent protection strategies crecy. Trade secrecy and actual secrecy, while offering feasible considera- for protection of some biotechnology-related inventions, conflict they bly more with scientific norms and rewards than patent law in that patent entail substantial nondisclosure. Even for inventors who choose dis- protection over secrecy, patent disclosure is likely to occur later than and rewards. Moreover, if the closure motivated solely by scientific norms the patent statute forbids anyone but the patentee from making or using may be prevented from invention for any purpose, subsequent researchers inventions. Insofar as patent law permits inventors to building on patented keep research competitors from using their discoveries, it could aggravate existing conflict between scientific norms and the incentives created by sci- entific rewards. The experimental use doctrine offers a potential mecha- nism for reconciling the patent monopoly with the interest of the research community in building upon prior discoveries through subsequent research. Of course, it is open to question whether the patent system should do anything at all to avoid aggravating existing conflict between the norms structure of science. The patent system formulates rules to and the reward agenda, which may not coincide with the aims of the advance its own If patent rules and incentives conflict with the norms scientific community. perhaps the scientific community should either and rewards of science, adjust its own norms and rewards, or do without patent protection. are several reasons why it might make sense for the patent sys- There tem to make some adjustments as well. First, the two systems are not far apart. Although patent law has tradi- tionally operated in the realm of applied technology, whereas scientific and rewards have governed basic research, both systems promote norms the creation and dissemination of new knowledge by rewarding priority of discovery in exchange for disclosure. As patent law moves into areas tradi- tionally defined as basic research, it will be able to achieve its ends more of the scientific community in effectively if it draws from the experience to achieve similar ends. developing norms and incentives the incentives of private research sponsors with Second, harmonizing the norms and incentives of research scientists can help attract vitally- research projects. While private sponsor- needed private dollars to basic ship may have some undesirable side effects, such as distorting the re- Biotechnology Research search agenda of the scientific community, it nonetheless offers some pro- tection against the vicissitudes of political support for federal research funding. Third, the extension of patent protection to living organisms demands a fresh look at traditional patent doctrine. As the PTO and the courts ex- plore the special challenges posed by this new kind of intellectual prop- erty, sensitivity to the interaction of patent rules with scientific norms and rewards can facilitate the progress of science in these fields while preserv- ing incentives for private sector funding of research. Finally, the patent system will influence the behavior of research scien- tists more effectively if it takes into account the norms and incentives that guide behavior in the scientific community. Although patent law has avoided interference with basic research in the past, the two systems of rules and incentives are beginning to converge. Legal rules are more likely to succeed in influencing the behavior of scientists if they resonate with scientists' conception of appropriate behavior than if they call for counter- intuitive departures from the norms of science. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The Yale Law Journal Unpaywall

Proprietary Rights and the Norms of Science in Biotechnology Research

The Yale Law JournalDec 1, 1987

Loading next page...
 
/lp/unpaywall/proprietary-rights-and-the-norms-of-science-in-biotechnology-research-nBGVBXAx08

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Unpaywall
ISSN
0044-0094
DOI
10.2307/796481
Publisher site
See Article on Publisher Site

Abstract

The Yale Law Journal Volume 97, Number 2, December 1987 Articles Norms of Proprietary Rights and the Science in Biotechnology Research Rebecca S. Eisenberg As basic research in biotechnology yields increasing commercial applica- tions, scientists and their research sponsors have become more eager to protect the commercial value of research discoveries through intellectual property law. Some scientists fear that these commercial incentives will weaken or even undermine the norms that have traditionally governed scientific research. In this Article, Professor Eisenberg examines the interaction of proprietary rights in inventions with these traditional scientific norms. Trade secrecy, she ar- research discoveries gues, is an undesirable strategy for protection of basic because it impedes dissemination of new knowledge to the scientific community. She finds that patent law is in many respects more congruent with scientific norms than trade secrecy because it is premised on disclosure rather than Eisenberg demonstrates, however, that the fit between the secrecy. Professor and incentives of the scientific community is patent system and the norms hardly perfect. Patent law may operate to delay the dissemination of knowl- edge to other researchers. Moreover, by granting rights to exclude others from using patented inventions for a term of years, the patent system threatens the interest of the scientific community in the free use and extension of new discov- eries. Professor Eisenberg concludes that greater sensitivity to the impact of © 1987 Rebecca S. Eisenberg of Law, University of Michigan Law School. I wish to thank Marty Add- t Associate Professor Gage, Judah Garber, Paul Goldstein, Don Herzog, Dick man, Ed Cooper, Mel Eisenberg, Julie Kaba, Jim Krier, Rick Lempert, Jessica Litman, Fred Schauer, Jonathan Weinberg, Peter Westen, and Virginia Wise for helpful comments on earlier drafts of this Article. The Yale Law Journal [Vol. 97: 177 patent law doctrine on scientific norms will help to reconcile the norms and incentives of these two systems. INTRODUCTION The commercial potential of recent advances in biotechnology has sub- stantially increased private investment in basic research in the biomedical sciences.' Some academic scientists have viewed this influx of private funding as a mixed blessing. Although it comes at a time when the con- tinuity of public funding for biomedical research is uncertain, it raises concerns about the potential impact of commercial incentives to protect intellectual property on traditions of open communication and free flow of information within the scientific community.' Indeed, some scientists have 1. See S. OLSON, BIOTECHNOLOGY: AN INDUSTRY COMES OF AGE (1986); Blumenthal, Gluck, Louis & Wise, Industrial Support of University Research in Biotechnology, 231 SCIENCE 242 (1986). Between World War 11 and the 1970s, basic research in the biomedical sciences was dominated by public funding. See generally M. KENNEY, BIOTECHNOLOGY: THE UNIVERSITY-INDUSTRIAL COm- PLEX 13-18 (1986); 1 B. SMITH & J. KARLESKY, THE STATE OF ACADEMIC SCIENCE: THE UNIVER- SITIES IN THE NATION'S RESEARCH EFFORT 11-48 (1977), and sources cited therein. I use the term "basic research" to refer to "pure" research directed solely toward expanding human knowledge, as opposed to "applied" research directed toward solving practical problems. See infra notes 38-47 and accompanying text. I shall argue below that, whatever validity this dichotomy may have in other contexts, it is difficult to maintain in the context of contemporary biotechnology re- search. See infra notes 93-109 and accompanying text. 2. Between 1980 and 1984, industrial funding for research and development at universities and colleges increased 93%, from $237,025,000 in 1980 to $457,227,000 in 1984, while federal funding increased only 31%, from $4,096,029,000 to $5,386,578,000. NATIONAL SCIENCE FOUNDATION, NA- TIONAL PATTERNS OF SCIENCE AND TECHNOLOGY RESOURCES 65 (1986) (Table 51). Adjusted for inflation, the total investment in research and development at universities and colleges rose only 4% between 1980 and 1983. This extremely modest rise reflects cutbacks in federal funding, particularly in the health area. Id. at 12-13. Since 1980, federal support has increasingly shifted toward defense-related fields at the expense of other categories of research and development. While the federal government's share of total national research and development outlays remained at a relatively constant 46-47% between 1980 and 1986, federal defense-related expenditures rose from 22% of this total in 1980 to an estimated 30% in 1986, while federal civilian-related expenditures fell from 18% in 1980 to an estimated 10% in 1986. Id. at 41 (Table 13). During this period, federal research and development funding for national defense increased approximately 183%, from $14,946,000,000 in 1980 to an estimated $42,360,000,000 in 1986, while federal funding for health-related research and development increased only 38%, from $3,694,000,000 in 1980 to an estimated $5,108,000,000 in 1986. Id. at 45 (Table 22). Biomedical research would have fared worse under budgets proposed by the Reagan administration, but Congress has repeatedly approved funding increases significantly greater than the administration's recommen- dations. See Culliton, Pressure to Cut the Deficit Creates Uncertainty for Biomedical Research, 232 SCIENCE 564 (1986); Norman, Science Escapes Brunt of Budget Ax, 231 SCIENCE 785, 787 (1986). 3. See generally Commercialization of Academic Biomedical Research: Hearings Before the Sub- comm. on Investigations and Oversight and the Subcomm. on Science, Research and Technology of the House Comm. on Science and Technology, 97th Cong., 1st Sess. (1981) [hereinafter Commerciali- zation Hearings]; D. NELKIN, SCIENCE AS INTELLECTUAL PROPERTY (1984); TWENTIETH CEN- TURY FUND, INC., THE SCIENCE BUSINESS (1984) [hereinafter THE SCIENCE BUSINESS]; Grobstein, Biotechnology and Open University Science, 10 SCI. TECH. & HUM. VALUES, Spring 1985, at 55. Concerns about the impact of industrial funding on biomedical research are by no means confined to the effects of intellectual property law. Other potential problems include distortion of the biomedi- cal research agenda as academic scientists seek to attract corporate sponsors; bypassing of traditional peer review mechanisms in allocating research funds, with consequent deterioration in the quality of research; alteration of the research and subsequent employment opportunities available to graduate students; divided loyalties of faculty affiliated with industry; compromise of university autonomy in 19871 Biotechnology Research asserted that granting property rights in research discoveries is antithetical to the norms of science.' In this Article I examine the interaction of intellectual property rights with research science norms in biotechnology-related fields to determine whether the two systems of rules and incentives conflict, and if so, how. Biotechnology research presents an unusual, if not unprecedented,' juxta- position of the incentives of the patent law system with the norms and such matters as faculty appointments; and loss of university credibility for impartial expertise on scientific issues. See generally M. KENNEY, supra note 1; Blumenthal, Gluck, Louis, Stoto & Wise, University-Industry Research Relationships in Biotechnology: Implications for the University, 232 SCIENCE 1361 (1986); Markle & Robin, Biotechnology and the Social Reconstruction of Molecular Biology, 10 Scs. TECH. & HUM. VALUES, Winter 1985, at 70; Note, Patent and Trade Secret Protec- tion in University-Industry Research Relationships in Biotechnology, 24 HARV. J. ON LEGIS. 191, 201-08 (1987); Note, Ties That Bind: Conflicts of Interest in University-Industry Links, 17 U.C. DAvis L. REv. 895 (1984). 4. For example, microbiologist Jonathan King has claimed that concern for preserving proprietary rights is retarding communication among university scientists, thereby undermining a previously shared "professional canon" of promoting the dissemination of knowledge for the public welfare: The openness, the free exchange of ideas and information, the free exchange of strains, of protein, of techniques, have been a critical component in the creativity and productivity of the biomedical research community ... This freedom of communication stemmed from the fact that all of the investigators shared the same professional canon; the increase of knowledge of health and disease for the general benefit of the citizenry ... Individuals planning to profit personally from commercial development, by, for example, assigning patents to their own firms, tend to cut down communication with their colleagues ... This secrecy is inimical to the function and effectiveness of universities in educating future generations of scientists, as well as retarding the research effort. Commercialization Hearings, supra note 3, at 62-63 (testimony of Dr. Jonathan King). 5. Throughout this Article I use the word "norms" in a normative rather than descriptive sense. By "research science norms" I mean socially inculcated beliefs within the research science community about how scientists should behave, as opposed to descriptions of how they actually do behave. See infra notes 8-28 and 110-54 and accompanying text. 6. In some fields, such as chemistry and engineering, industrial applications of research have been sufficiently immediate that universities and academic researchers have engaged in industry-sponsored research projects for years. In these areas, researchers have long confronted industry's concern for protecting intellectual property. See Commercialization Hearings, supra note 3, at 28-60 (testimony of Dr. Paul Gray); D. DICKSON, THE NEW POLITICS OF SCIENCE 89-90 (1984); Servos, The Indus- trial Relations of Science: Chemical Engineering at MIT, 1900-1939, 71 Isis 531 (1980); Smith & Hounshell, Wallace H. Carothers and Fundamental Research at DuPont, 229 SCIENCE 436 (1985). But in other fields, including many fields within the biomedical sciences, the gap between basic and applied research has been wider. See Peters & Fusfeld, Current U.S. University-Industry Research Connections, in NATIONAL SCIENCE BOARD, UNIVERSITY-INDUSTRY RESEARCH RELATIONSHIPS: SELECTED STUDIES 1, 20-21 (1982). In these fields, there have been fewer occasions to reconcile conflict between scientific norms and patent law. The sudden juxtaposition of commercial incentives and scientific norms has been particularly strik- ing in the biomedical sciences, in part because of the strong public interest in health-related research and in part because of the rapid onset and proliferation of university-industry research relationships in biotechnology fields following decades of predominantly public funding. Martin Kenney describes the prevailing attitude among biomedical researchers prior to the biotechnology boom as follows: The role of the state as a funder of basic (non-commodity-oriented) research fostered a power- ful ideology-one of scientists working for the public good to improve the health status of Americans. According to this ideology, industry's motives-especially that of profitabil- ity-were suspect, and the applied science orientation of industry was considered to be scientif- ically uninspiring to scientists. M. KENNEY, supra note 1, at 32. In light of this ideology, it is not surprising that the privatization of new knowledge as intellectual property has been resisted in the biomedical research community. The Yale Law Journal [Vol. 97: 177 incentives of a basic research community. Given the growing economic significance of research-oriented industries, it seems likely that patent law will have an increasing impact on the conduct of basic research. Thus, the experience of the biomedical research community with patent law may offer a preview of phenomena that will soon become more widespread. If patent law is in fact unsalutary for the conduct of basic research, it would be useful to figure out why. In Section I, I analyze the divergence between intellectual property law and research science norms as traditionally conceived. I begin by describ- ing a conventional view of the norms of research science concerning the dissemination of new knowledge and the rights of scientists in their dis- coveries. Against this background, I examine impediments to intellectual property protection for basic research discoveries in the biomedical sci- ences under traditional patent and trade secret doctrine. In Section II, I analyze the conjunction of the norms and incentives of research science and the rules and incentives of intellectual property in contemporary bio- technology research. First, I describe recent developments that have helped bring about this conjunction. Next, I analyze the operation of scientific norms concerning publication of research results in the specific context of contemporary biotechnology research, as illustrated by the records of a recent controversy within the Journal of Biological Chemistry. I then ex- amine patent law doctrine as applied to biotechnology inventions to deter- mine whether and how patent law conflicts with scientific norms in this particular research context. I conclude that, although there are substantial parallels between the two systems, the conjunction may nonetheless cause delay in the dissemination of new knowledge and aggravate inherent con- flict between the norms and the reward structure of science. I. THE TRADITIONAL DIVERGENCE BETWEEN INTELLECTUAL PROPERTY AND SCIENTIFIC NORMS IN BASIC RESEARCH Given the shared focus of research science and patent law on promoting invention and the dissemination of new knowledge, the proposition that scientific norms and patent laws have coexisted for centuries without 7. Patent laws have existed in this country since 1790 and in Europe since the 17th century. The origins of patent law are generally traced to the English Statute of Monopolies, 1623, 21 Jac., ch. 3, reprinted in 6 D. CHISUM, PATENTS app. at 8-1 (1987). Patents granted in the American colonies were primarily special enactments of colonial legislatures in favor of individual inventors. B. BUGBEE, GENEsIs OF AMERICAN PATENT AND COPYRIGHT LAW 57 (1967). The first United States patent statute was enacted in 1790, Patent Act of 1790, ch. 7, 1 Stat. 109-12, and revised in 1793, Patent Act of 1793, ch. 11, 1 Stat. 318-23. A more substantial revision occurred in the Patent Act of 1836, ch. 357, 5 Stat. 117-25, which created a patent office to examine patent applications for conformance to statutory criteria prior to the issuance of patents. The law has since been revised on numerous occasions. See Patent Act of 1839, ch. 88, 5 Stat. 353-55; Patent Act of 1870, ch. 230, 16 Stat. 198-217; Patent Act of 1897, ch. 391, 29 Stat. 692-94; Patent Act of 1903, ch. 1019, 32 Stat. 1225-27; Patent Act of 1928, ch. 730, 45 Stat. 732; Patent Act of 1939, ch. 450, 53 Stat. 1212; Patent Act of 1952, Pub. L. No. 593, 66 Stat. 792 (codified as amended at 35 U.S.C. §§ 1-376 (1982 & 1987] Biotechnology Research working out their differences calls for some explanation. Conflict may have been forestalled in the past by mutual avoidance: Particularly in the biomedical sciences, both legal doctrine and scientific norms have discour- aged attempts to protect basic research discoveries as intellectual property. A. A Traditional View of the Norms of Science Some observers have asserted that, prior to the recent growth of com- mercial interest in biotechnology, neither research scientists nor the insti- tutions funding basic scientific research in the biomedical sciences showed much inclination to pursue patent rights." Universities, where much of the research was conducted, encouraged the dissemination of research results through publication and occasionally showed a positive aversion to patent- ing discoveries. Public sponsorship of basic research permitted university researchers to avoid industrial sponsors who might have exerted pressure to protect in- tellectual property. Government sponsors gave universities and research- ers little incentive to identify patentable inventions or pursue patent rights. While the patent policies of the federal agencies that fund univer- Supp. 111 1985)). Robert Merton traces the origins of the ethos and norms of modem research science back to 17th- century Puritanism. R. MERTON, The Puritan Spur to Science, in THE SOCIOLOGY OF SCIENCE 228-53 (1973) [hereinafter THE SOCIOLOGY OF SCIENCE], originally published as R. MERTON, Mo- tive Forces of the New Science, in SCIENCE, TECHNOLOGY AND SOCIETY IN SEVENTEENTH-CENTURY ENGLAND 80-111 (1970). 8. Commercialization Hearings, supra note 3, at 62-63 (testimony of Dr. Jonathan King); THE SCIENCE BUSINESS, supra note 3, at 30-31 (1984) (background paper by Nicholas Wade). See gener- ally R. MERTON, The Normative Structure of Science, in THE SOCIOLOGY OF SCIENCE, supra note 7, at 275, originally published as Science and Technology in a Democratic Order, 1 J.L. & POL. Soc. 115 (1942). Merton cites a handful of counterexamples of prominent research scientists outside the biomedical sciences who have been patent holders, including Einstein, Millikan, Compton, and Langmuir. Id. at 275 n.16, citing W. HAMILTON, PATENTS AND FREE ENTERPRISE 154 (TNEC Monograph No. 31, 1941). As recently as 1975, academic scientists Cesar Milstein and Georges Kbihler decided not to patent their path-breaking and commercially valuable invention of monoclonal antibody-producing hybridoma cells. See M. KENNEY, supra note 1, at 129. Kenney suggests that they may have believed it inappropriate to patent such a potentially life-saving discovery. Id. 9. David Dickson quotes a 1934 policy statement by the president and fellows of Harvard Univer- sity that "'no patents primarily concerned with therapeutics or public health may be taken out by any member of the university, except with the consent of the president and fellows; nor will such patents be taken out by the university itself except for dedication to the public.'" D. DICKSON, supra note 6, at 89. Harvard abandoned this policy in 1975 after signing an agreement to give an exclusive world- wide license to Monsanto Corporation for inventions made in the course of research under a 12-year, $23 million grant from Monsanto to the Harvard Medical School. See M. KENNEY, supra note 1, at 58-60, 78; Culliton, Harvard and Monsanto: The $23-Million Alliance, 195 SCIENCE 759-63 (1977). A notable early counterexample to the generalization that universities were averse to patenting discoveries is the University of Wisconsin, which has administered patents and licenses resulting from research discoveries of faculty through a non-profit foundation since 1925. See Blumenthal, Epstein & Maxwell, Commercializing University Research: Lessons from the Experience of the Wisconsin Alumni Research Foundation, 314 NEW ENG. J. MED. 1621 (1986). 10. As recently as 1982, a National Science Foundation study reported that in many academic fields there appeared to be a "psychological barrier to interacting with industry." Peters & Fusfeld, supra note 6, at 39. The Yale Law Journal [Vol. 97: 177 sity research varied considerably prior to the Patent and Trademark Act Amendments of 1980,11 a common practice was to require that rights in inventions made in the course of government-sponsored research be as- signed to the government. This practice accorded with the prevailing wisdom since World War II that private ownership of inventions made through public funding was contrary to the public interest. The disinclination to secure patents also had a normative component: It was thought contrary to scientific norms to claim exclusive rights in re- search discoveries. These norms derive in part from the notion that mak- ing new observations available to the scientific community for evaluation and extension in further research facilitates the progress of science.1 In the biomedical sciences, this notion has been fortified by a belief that new knowledge should be made as widely available as possible in order to serve humanity.' A commonly held conception of the norms and incentives that guide the behavior of research scientists is set forth in the writings of sociologist Robert Merton.1 Merton identifies the overriding institutional goal of sci- 11. Pub. L. No. 96-517, 94 Stat. 3019 (codified at 35 U.S.C. §§ 200-11 (1982)). These revisions established a uniform policy permitting non-profit organizations (defined to include "universities and other institutions of higher education," 35 U.S.C. § 201(i)) and small business firms to retain title to patentable inventions made in the course of government-sponsored research. See infra notes 102-08 and accompanying text. 12. See Dobkin, Patent Policy in Government Research and Development Contracts, 53 VA. L. REV. 564, 568-84, 591-607 (1967). 13. President's Memorandum, 28 Fed. Reg. 10,943, 10,943-44 (1963); 1 U.S. DEP'T OF JUS- TICE, INVESTIGATION OF GOVERNMENT PATENT PRACTICES AND POLICIES: REPORT AND RECOM- MENDATIONS OF THE ATTORNEY GENERAL TO THE PRESIDENT 4-5 (1947). See generally Dobkin, supra note 12. 14. See infra text accompanying notes 17-28. 15. See infra text accompanying notes 17-19. 16. See Commercialization Hearings, supra note 3, at 63 (testimony of Dr. Jonathan King); M. KENNEY, supra note 1, at 32. These writings are collected in THE SOCIOLOGY OF SCIENCE, supra note 7. Merton is often cited in descriptions of the traditional ethos of research science. See, e.g., D. DICKSON, supra note 6, at 90 n.115; J. GASTON, ORIGINALrrY AND COMPETITION IN SCIENCE 5-6 & passim (1973); THE SCIENCE BUSINESS, supra note 3, at 29-30; Garfield, Citation Measures of the Influence of Robert K. Merton, 39 TRANSACTIONS N.Y. ACAD. Scs. SERIES II 61 (1980), and sources cited therein. The picture Merton paints of the scientific community has been criticized, however, as inaccurate and misconceived. See, e.g., J. RAVETZ, SCIENTIFIC KNOWLEDGE AND ITS SOCIAL PROBLEMS 311-12 (1971); Mulkay, Sociology of Science in the West, 28 CURRENT SOC. 1, 43-64 (1980), and sources cited therein; Stehr, The Ethos of Science Revisited, in THE SOCIOLOGY OF SCIENCE 172 (J. Gaston ed. 1978), and sources cited therein. The fact that scientists may depart from Merton's norms in practice does not necessarily mean that they lack normative authority within the scientific community. Nicholas Wade explains: Academic scientists quite regularly depart from the norms of universalism, communism, and disinterestedness-usually in the pursuit of personal recognition. Vigorously asserting one's claim to a discovery, even at the expense of others' legitimate claims, is surely not disinterested, and may not even be justified, but it is probably not in itself a serious threat to the purity or progress of science. Indeed, such behavior is part of the competitive attitude that characterizes many of the fastest-moving arenas of scientific inquiry. And the intensity of glory-seeking is sometimes moderated by the need to observe at least the appearance of Mertonian norms; thus The Double Helix, James Watson's account of the competition to discover the structure of DNA, occasioned a major stir within the scientific community not so much because it described unusual behavior but because it publicly acknowledged motives that scientists usually reserve Biotechnology Research ence as "the extension of certified knowledge."" The methodology for achieving this goal is empirical research. Merton describes four interre- lated norms that derive from the institutional goal and the methodology: 1 9 universalism, communism, disinterestedness, and organized skepticism. "Universalism" means that the veracity of claimed observations is to be determined on the basis of impersonal criteria without regard to the iden- tity of the scientist who makes the observation. Since what is true in one laboratory will also be true in another, all scientists, regardless of their personal qualities or national or institutional affiliations, may draw on and contribute to the same body of certified knowledge." "Communism" means that scientific findings are a product of social collaboration and should be dedicated to the scientific community. All discoveries build on what has been learned previously and contribute to what may be learned in the future. "Disinterestedness" means that scientists should seek truth rather than seeking to further their own interests by advancing spurious claims. "Organized skepticism" means that the scientific community should subject the claims and beliefs of its members to empirical scrutiny before accepting them as true. Merton describes a scientific reward structure that fortifies these norms. The scientific community rewards those who make original contributions to the common stock of knowledge by giving them professional recogni- 2 4 tion. The emphasis on originality creates pressure to publish as quickly as possible in order to avoid being forestalled by others who are doing 2 5 competitive research on the same problems. Upon publication, scientists lose exclusive use of their scientific property. The reward structure of sci- ence thereby reinforces Merton's norm of communism. By offering recog- nition and esteem to those who contribute to its shared body of knowledge, for private confessions. THE SCIENCE BUSINESS, supra note 3, at 30. The presence of a norm is indicated by community embarrassment and disapproval at its violation. For example, in acknowledging violations of scientific norms concerning the sharing of new knowl- edge in the biomedical sciences, Jonathan King stresses that such behavior is frowned upon: "I don't mean to say there isn't professional jealously [sic]. We have ambition and we have fame and recogni- tion, but it is considered a departure from the normal and you are embarrassed when it comes out. It is not what you are supposed to be doing." Commercialization Hearings, supra note 3, at 63. For evidence that Merton's norms continue to influence how researchers in the biomedical sciences believe they ought to behave, see infra notes 112-54 and accompanying text. 18. R. MERTON, The Normative Structure of Science, in THE SOCIOLOGY OF SCIENCE, supra note 7, at 267. 19. id. 20. Id. at 270-73. 21. Id. at 273-75. 22. Id. at 275-77. 23. Merton defines organized skepticism as "[t]he temporary suspension of judgment and the de- tached scrutiny of beliefs in terms of empirical and logical criteria." Id. at 277. This skepticism is applied to societal assumptions about the world as well as to claims of other scientists. Id. at 277-78. 24. R. MERTON, Priorities in Scientific Discovery, in THE SOCIOLOGY OF SCIENCE, supra note 7, at 286, originally published in 22 AM. Soc. REv. 635 (1957). 25. R. MERTON, Behavior Patterns of Scientists, in THE SOCIOLOGY OF SCIENCE, supra note 7, at 325-27, originally published in 105 PROC. AM. PHIL Soc'Y 470 (1961). The Yale Law Journal [Vol. 97: 177 the scientific community insures that scientists' self-interest will coincide with the public good. This reward structure also reinforces Merton's norm of disinterested- ness, in that publication involves submitting research results to the scru- tiny of other scientists. Efforts to gain recognition by making false claims are unlikely to succeed; false claims may be exposed if other scientists are 2 7 results. the fraudulent to replicate unable According to this model, both the norms and the rewards of science promote prompt disclosure of new discoveries through publication. Upon publication, the scientist's observations enter the public domain, permit- ting the scientific community to check the veracity of her claims and to build upon them in further research." To the extent that it might inter- fere with publication of new claims or limit the ability of other scientists to use published knowledge, intellectual property law has been perceived within the scientific research community as conflicting with the traditional norms and rewards of science. B. Traditional Intellectual Property Protection for Basic Research Discoveries At the same time, legal doctrine has discouraged the protection of basic research discoveries in the biomedical sciences as intellectual property. In most fields of technology, intellectual property protection can be obtained through the federal patent system, state trade secrecy law, or actual se- crecy. " Limitations on patentable subject matter and requirements for patentability have operated, however, to exclude basic research discoveries 26. Merton believed that proprietary rights in discoveries conflicted with the scientific norm of communism, R. MERTON, supra note 18, at 275, but in fact a similar communal value may be reflected in the patent law policy of requiring disclosure and ultimate dedication to the public as the quid pro quo of a patent monopoly. See infra text accompanying notes 155-206. 27. See generally Fraud in Biomedical Research: Hearings Before the Subcomm. on Investiga- tions and Oversight of the House Comm. on Science and Technology, 97th Cong., 1st Sess. 19-24, 55-58, 352-55 (1981) (testimony of Dr. Philip Handler, Dr. John C. Long, and Dr. Patricia Woolf, respectively) [hereinafter Fraud Hearings]. 28. The only "payment" owed by users of the published research is citation to the original author. J. RAVETZ, supra note 17, at 247, 255. The frequency of citation to a publishing scientist's work is a measure of the esteem in which it is held. Id. at 257. 29. A notable exception to this general rule is computer software, which may be protected under the copyright statute but not under the patent statute. See Chisum, The Patentability of Algorithms, 47 U. PIr. L. REV. 959 (1986); Davidson, Protecting Computer Software: A Comprehensive Anal)- sis, 1983 ARIz. ST. L.J. 611; Samuelson, CONTU Revisited: The Case Against Copyright Protection for Computer Programs in Machine-Readable Form, 1984 DUKE L.J. 663; cf. Diamond v. Diehr, 450 U.S. 175 (1981) (processes incorporating computer programs may be patented, although com- puter programs themselves may not be patented). For an argument that genetically engineered organisms are works of authorship entitled to copyright protection, see Kayton, Copyright in Living Genetically Engineered Works, 50 GEO. WASH. L. REv. 191 (1982). Recently, Congress has provided sui generis copyright-like protection for semiconductor chip masks. Semiconductor Chip Protection Act of 1984, Pub. L. No. 98-620, 98 Stat. 3347 (codified at 17 U.S.C. §§ 901-14 (Supp. III 1985)). See generally Symposium: The Semiconductor Chip Protection Act of 1984 and Its Lessons, 70 MINN. L. REV. 263 (1985). Biotechnology Research 1987] from patent protection. Even without patent protection, some basic re- search discoveries can be protected through the law of trade secrecy. De- pending on the scope of permissible subject matter under applicable state law, however, basic research discoveries may not qualify for legal protec- tion as trade secrets either, although they might still be protected as a practical matter through a program of actual secrecy. Both forms of se- crecy require substantial nondisclosure, thereby preventing researchers from attaining recognition and conflicting with traditional scientific norms. 1. Doctrinal Impediments to Patent Protection A patent gives an inventor" the exclusive right to make, use, and sell the invention in the United States for a period of seventeen years." At the end of this period, the invention becomes freely available to all. To qualify for patent protection, an invention must fall within patentable subject 3 2 3 matter and must satisfy the further statutory requirements of novelty, 3 4 5 utility, and nonobviousness." In exchange for the property rights con- ferred by the patent system, the inventor must disclose her invention to the public. Insofar as patent law not only permits but even requires disclo- sure, it is more congruent with scientific norms than either legal trade secrecy or actual secrecy. Nonetheless, patent protection has often been unavailable for basic research discoveries in the biomedical sciences. Traditional patent law doctrine has confined the reach of patent protec- tion to inventions in applied technology, as distinguished from basic scien- tific research. This focus on applied technology derives in part from the language of the Constitution, which authorizes Congress "to promote the Progress of Science and useful Arts, by securing for limited Times to Au- thors and Inventors the exclusive Right to their respective Writings and Discoveries." " This provision is traditionally read distributively. Under this reading, the function of the patent law is to promote the progress of "useful Arts" (i.e., applied technology) by securing exclusive rights to 30. Patent applications must generally be made by the inventor or inventors, 35 U.S.C. §§ 111, 115, 116 (1982), although in some instances persons other than the inventor who have proprietary interests in the invention may apply for a patent. 35 U.S.C. §§ 117, 118 (1982). Patents and patent applications may be assigned by the inventor to someone else. 35 U.S.C. §§ 152, 261 (1982). 31. 35 U.S.C. §§ 154, 271(a) (1982); see infra notes 207-14 and accompanying text. 32. 35 U.S.c. § 101 (1982). 33. 35 U.S.C. §§ 101, 102 (1982). 34. 35 U.S.c. §§ 101, 112 (1982). 35. 35 U.S.c. § 103 (1982); Graham v. John Deere, 383 U.S. 1, 13-19 (1966). 36. 35 U.S.c. § 112 (1982). 37. Cf Commercialization Hearings, supra note 3, at 79 (testimony of Dr. William Raub, Na- tional Institutes of Health) ("[Mlany biomedical scientists, perhaps most, regard the patent process as a means of institutionalized secrecy, whereas it is in fact a time-tested way to assure broad and ready access to proprietary information."); see also infra notes 162-206 and accompanying text (comparing disclosure under patent system with disclosure motivated by scientific norms). 38. U.S. CONsr. art. I, § 8, cI. 8. The Yale Law Journal [Vol. 97: 177 "Inventors" in their "Discoveries," while the function of the copyright law is to promote the progress of "Science" (i.e., knowledge in general) by 3 ' 9 securing exclusive rights to "Authors" in their "Writings. So under- stood, the Constitution only authorizes Congress to extend patent protec- tion to inventions in applied technology. Congress has accordingly defined patentable subject matter to limit pat- ent protection to applied technology. The present patent statute defines as patentable subject matter "any new and useful process, machine, manu- facture, or composition of matter, or any new and useful improvement thereof." Applying this definition and its predecessors in earlier legisla- tion, the cases have denied patent protection to theoretical or abstract dis- coveries,"' laws of nature, products of nature, principles, abstract ideas, mathematical formulae, and algorithms. Patents are issued only for the discovery of new means to achieve useful results." To the extent that ba- sic research discoveries comprise explanations for existing means of achieving useful results or principles that are only put into practical use later by the work of others, they may not be protected by patent." Patent law doctrine further restricts the patentability of basic research discoveries through the requirement that a patent applicant make a disclo- sure demonstrating that the invention is "operable and capable of use" before a patent will issue. This requirement impedes the patenting of basic research discoveries in two ways. First, if technical problems remain to be solved before the discovery may be put to practical use, it may not yet be patented. Second, an invention with no demonstrated "utility" may not be patented, however interesting and significant it may be to research scientists. The Supreme Court has held that a patent applicant must demonstrate a specific and substantial utility for a claimed invention. The 39. See, e.g., In re Bergy, 596 F.2d 952, 958-59 (C.C.P.A. 1979), affd sub noma. Diamond v. Chakrabarty, 447 U.S. 303 (1980) ("[Tlhe constitutionally-stated purpose of granting patent rights to inventors for their discoveries is the promotion of progress in the 'useful Arts,' rather than in sci- ence. . . . [T]he present day equivalent of the term 'useful arts' employed by the Founding Fathers is 'technological arts.' "). See generally 1 D. CHisuM, supra note 7, § 1.01 n.9 (1987) (Congress may enact patent legislation only to promote "useful arts," not to promote "science" or knowledge in general). 40. 35 U.S.C. § 101 (1982). 41. Mackay Radio & Tel. Co. v. Radio Corp. of Am., 306 U.S. 86, 94 (1938); In re King, 801 F.2d 1324, 1328 (Fed. Cir. 1986); 1 D. CHisuM, supra note 7, § 1.01. 42. Funk Bros. Seed Co. v. Kalo Inoculant Co., 333 U.S. 127 (1948). Parker v. Flook, 437 U.S. 584 (1978); Gottschalk v. Benson, 409 U.S. 63 (1972). 44. 1 D. CHISUM, supra note 7, § 1.01. 45. Id. § 1.0312]. As Chisum remarks, some courts have noted the irony of denying patent protec- tion to the discoverers of fundamental scientific principles, while extending protection to "those lesser geniuses who put such discoveries to practical uses." Id. (quoting opinion of Frank, J., in Katz v. Horni Signal Mfg. Corp., 145 F.2d 961, 961 (2d Cir. 1944)). The Supreme Court has justified the distinction on the ground that scientific principles are the "basic tools" of science and technology. Gottschalk v. Benson, 409 U.S. 63, 67 (1972). Judge Frank has offered another rationale: Great scientists are not motivated by pecuniary gain, so the prospect of a patent monopoly will not enhance their productivity. Katz, 145 F.2d at 961. 46. This requirement derives from 35 U.S.C. §§ 101, 112 (1982). See 1 D. CHISUM, supra note 7, at § 4.04. 19871 Biotechnology Research 187 fact that the claimed invention "may be an object of scientific research" does not satisfy the utility standard. Thus, under existing law, many basic research discoveries are not yet ripe for patent protection. Even in the area of applied technology, patent protection for inventions in the biomedical sciences has been retarded by cases holding that medical and agricultural processes may not be patented. While the rationale for these cases is not entirely clear," at least one court seems to have believed that a patent monopoly on methods of medical treatment is contrary to the ethics of medical doctors and the public interest: Doctors and surgeons have seldom thought it desirable to patent their new procedures for human relief. . . . The professional ethics of doctors and surgeons are more consistent with the widespread use of their medical and surgical discoveries for the benefit of mankind than in obtaining a monopoly to control their discoveries for personal commercial advantage. In this respect it would seem also that public interest is here involved. " Finally, prior to the Supreme Court's 1980 decision in Diamond v. Chakrabarty, patent protection for biological materials was retarded by the longstanding belief that living organisms and cells were unpatentable products of nature." One explanation for this limitation on patentable 47. Brenner v. Manson, 383 U.S. 519, 534-35 (1966). 48. See generally I D. CHISUM, supra note 7, § 1.03[3], and cases cited therein. Medical prod- ucts (such as drugs), as opposed to medical processes, may be patentable. See Ruskin v. Coe, 58 F. Supp. 424 (D.D.C. 1945). 49. The origin of the rule that medical procedures may not be patented is apparently the holding in Morton v. New York Eye Infirmary, 17 F. Gas. 879 (S.D.N.Y. 1862) (No. 9865). The inventors in that case discovered that ether could be used as an anesthetic for patients undergoing surgery. The court acknowledged that the inventors had made one of the "greatest discoveries of modern times," but concluded nonetheless that it was not a patentable invention: "A discovery may be brilliant and useful, and not patentable. . . . The new force or principle brought to light must be embodied and set to work, and can be patented only in connection or combination with the means by which, or the me- dium through which, it operates." Id. at 884. The court followed similar reasoning in Wall v. Leck, 66 F. 552 (9th Cir. 1895), extending the exclusion from patent protection to agricultural methods. In Ex parte Brinkerhoff, 24 Comm'r Ms. Dec. 349 (1883), reprinted in 27 J. PAT. OFF. SOC'Y 797, 798 (1945), the Commissioner of Patents offered another rationale for excluding medical proce- dures from patent protection-that the effectiveness of such treatments is uncertain, and granting a patent might have a tendency to deceive the public into believing that the method is more foolproof than it is. Subsequent decisions have retreated somewhat from the rules of Morton and Brinkerhoff, extending patent protection to a skin test for detecting the susceptibility of a human being to scarlet fever, Dick v. Lederle Antitoxin Laboratories, 43 F.2d 628 (S.D.N.Y. 1930), and a method of injecting medica- ments by a pressure jet, Ex parte Scherer, 103 U.S.P.Q. (BNA) 107 (P.T.O. Bd. App. 1954). 50. Martin v. Wyeth, Inc., 96 F. Supp. 689, 695 (D. Md. 1951), affd, 193 F.2d 58 (4th Cir. 1951). 51. 447 U.S. 303 (1980); see infra notes 60-66 and accompanying text. 52. See generally I D. CHISUM, supra note 7, § 1.02[7][a]-[b], and cases cited therein; Biggart, Patentability, Disclosure Requirements, Claiming and Infringement of Microorganism-Related In- ventions, in GENETICALLY ENGINEERED MICROORGANISMS & CELLS: THE LAW & THE BUSINESS C (I. Kayton ed. 1981) [hereinafter GENETICALLY ENGINEERED ORGANISMS]. Prior to Chakrabarty, two federal statutes provided intellectual property protection for new plants. The Plant Patent Protection Act of 1930, 46 Stat. 376 (codified at 35 U.S.C. §§ 161-164 (1982)), The Yale Law Journal [Vol. 97: 177 subject matter is that in order to be patentable a product must be new, and naturally-occurring organisms are not new. New processes using preexisting naturally-occurring products could be patented, but the prod- ucts themselves could not. An applicant who seeks a patent monopoly on a naturally-occurring product would deprive the public of the use of something "which nature has produced and which nature has intended to all men."1 the use of equally for be The Supreme Court's decision in Funk Bros. Seed Co. v. Kalo Inocu- lant Co. " suggested a broader, if somewhat less coherent, limitation on patentable subject matter under the "products of nature" doctrine. The claimed invention in that case was a combination of selected strains of naturally-occurring bacteria. When inoculated into leguminous plants, the 5 7 bacteria enabled the plants to fix nitrogen from the air. Although the mixture of strains was a new product in that the selected strains did not naturally occur in combination, the Supreme Court held the patent inva- lid. The precise basis for the decision is unclear. Justice Douglas, writing for the majority, suggested both that the invention was a product of nature outside the scope of patentable subject matter and that the combination of species failed to satisfy the nonobviousness or "invention" requirement. extended patent protection to asexually reproduced plants, and the Plant Variety Protection Act of 1970, Pub. L. No. 91-577, 84 Stat. 1542 (codified at 7 U.S.C. §§ 2321-2583 (1982)), allowed breed- ers of sexually reproduced plants to obtain patent-like protection from the Department of Agriculture. 53. Once the patent applicant has modified the product in any significant respect from its natu- rally-occurring state, this "product of nature" objection would seem to be overcome, since the claimed invention would then be a "manufacture" or "composition of matter" and thus within the statutory subject matter. I D. CHIsuM, supra note 7, § 1.02[7][a]. 54. In re Mancy, 499 F.2d 1289, 1294 (C.C.P.A. 1974) (distinction between new products and new processes); Guaranty Trust Co. of New York v. Union Solvents Corp., 54 F.2d 400, 411 (D. Del. 1931) (same). 55. Ex parte Latimer, 1889 Dec. Comm'r Pat. 123, 125 (1889). 56. 333 U.S. 127 (1948). 57. Different species of bacteria were effective for different crops, and multiple strains existed within each species. The patentee was the first to identify strains of bacteria that could be combined without inhibiting one another's effectiveness, permitting farmers to buy a single mixed culture inocu- lant usable for multiple crops. Id. at 129-30. 58. "The qualities of these bacteria, like the heat of the sun, electricity, or the qualities of metals, are part of the storehouse of knowledge of all men. They are manifestations of laws of nature, free to all men and reserved exclusively to none." Id. at 130. 59. [T]here is no invention here unless the discovery that certain strains of the several species of these bacteria are non-inhibitive and may thus be safely mixed is invention. But we cannot so hold without allowing a patent to issue on one of the ancient secrets of nature now disclosed. All that remains, therefore, are advantages of the mixed inoculants themselves. They are not enough. Id. at 132. In a concurring opinion, Justice Frankfurter suggested a more cogent basis for holding the patent invalid-that the patent claims were overly broad in extending to any mixture of compatible strains, rather than just the particular mixture derived by the patentee, and that the same failure to identify the particular compatible strains used by the patentee also made the description of the invention inadequate. Id. at 133-35. The description problem for inventions requiring the use of living organ- isms is now generally resolved by depositing cultures of the organisms in recognized depositories. See infra notes 167-78 and accompanying text. 1987] Biotechnology Research Thirty-two years later, in Diamond v. Chakrabarty, the Supreme Court held in a five to four decision that a living, genetically-engineered microorganism itself fell within the statutory categories of patentable sub- ject matter as either a "manufacture" or a "composition of matter." The majority held that living things are not categorically excluded from patent protection, noting that the relevant distinction is "not between living and inanimate things, but between products of nature, whether living or not, 8 1 and human-made inventions." While the Chakrabarty decision makes plain that the subject matter of patents extends to genetically-engineered microorganisms and that living things are not per se excluded from patent protection, the Supreme Court did not reach the issue of whether natu- rally-occurring microorganisms that have been newly isolated or purified also fall within the ambit of "manufactures" or "compositions of 6 2 matter." Since the Chakrabarty decision, the Patent and Trademark Office 6 3 (PTO) has extended patent protection to plants and indicated that non- naturally-occurring polyploid oysters would be patentable if the invention were nonobvious.11 The Commissioner of Patents recently issued a notice announcing that the PTO "now considers nonnaturally occurring non- human multicellular living organisms, including animals, to be patentable subject matter within the scope of 35 USC 1012"65 The Commissioner of 60. 447 U.S. 303 (1980). 61. Id. at 313. Chakrabarty modified an existing bacterial strain by introducing new de- oxyribonucleic acid (DNA) plasmids into a bacterial cell, thereby giving the organism the capacity to break down multiple components of crude oil. In so doing, he "produced a new bacterium with mark- edly different characteristics from any found in nature and one having the potential for significant utility." Id. at 310. The Court concluded that the resulting organism "is not nature's handiwork, but [Chakrabarty's] own." Id. 62. A companion case, In re Bergy, involved a claim to a biologically pure culture of an organism occurring in nature only in impure form. 563 F.2d 1031 (C.C.P.A. 1977), vacated sub nom. Parker v. Bergy, 438 U.S. 902 (1978), on remand, In re Bergy, 596 F.2d 952, cert. granted sub nom. Parker v. Bergy, 444 U.S. 924 (1979), vacated and remanded with instructions to dismiss as moot sub nom. Diamond v. Chakrabarty, 444 U.S. 1028 (1980). The Court of Customs and Patent Appeals held that the claims in both Chakrabarty and Bergy recited patentable subject matter. 596 F.2d at 952. Dicta from the Court of Customs and Patent Appeals in In re Bergy rejected the argument that a "biologi- cally pure culture" of an organism existing naturally only in impure form is an unpatentable product of nature, id. at 972-73, but so far no appellate court holding has resolved the issue. The Patent and Trademark Office (PTO) has been issuing patents on such organisms, and it seems likely that the Court of Appeals for the Federal Circuit would follow the Bergy dicta if the issue were to arise. See generally South Corp. v. United States, 690 F.2d 1368 (Fed. Cir. 1982) (Court of Appeals for Fed- eral Circuit will consider prior decisions of Court of Customs and Patent Appeals binding as precedent). 63. In re Hibberd, 227 U.S.P.Q. (BNA) 443, 444 (P.T.O. Bd. App. & Int. 1985). Plants may also qualify for more limited protection under the Plant Patent Act, 35 U.S.C. §§ 2321-2364 (1982), or the Plant Variety Protection Act, 7 U.S.C. §§ 2321-2583 (1982). The Plant Patent Act grants "the right to exclude others from asexually reproducing the [protected] plant or selling or using the plant so reproduced," 35 U.S.C. § 163, while the Plant Variety Protection Act grants the right "to exclude others from selling the [protected] variety, or offering it for sale, or using it in producing (as distin- guished from developing) a hybrid or different variety therefrom . . . ." 7 U.S.C. § 2483(a). 64. In re Allen, 2 U.S.P.Q.2d (BNA) 1425 (P.T.O. Bd. App. & Int. 1987). 65. Nonnaturally Occurring Non-Human Animals Are Patentable Under § 101, 33 Pat. Trade- mark & Copyright J. (BNA) No. 827, at 664 (Apr. 23, 1987). The Yale Law Journal [Vol. 97: 177 Patents cited Chakrabarty for the proposition that Congress intended the subject matter of patents to "include anything under the sun that is made by man," but reiterated that products found in nature may not be patented. In short, although patent protection has previously been unavailable for basic research discoveries in general and for those in the biomedical sci- ences in particular, the trend of recent authority is towards increasing availability of patent protection for biotechnology-related inventions. 2. Doctrinal, Practical, and Normative Impediments to Protection Through Secrecy Even in the absence of patent protection, however, inventors can some- times protect their research discoveries through some form of secrecy. The term "trade secrecy" is often used ambiguously to refer to two different types of protection: legal trade secrecy and actual secrecy. Legal trade se- crecy affords a remedy in tort to persons who disclose certain kinds of information in confidence against those who breach this confidence or who otherwise misappropriate the information. Actual secrecy is a practical, nonlegal strategy for protection that may be effective in circumstances where not all of the legal requirements for trade secrecy protection have been satisfied. Since in many jurisdictions legal trade secrecy doctrine might exclude from protection research discoveries not yet put to use in a 66. Id. It is not clear at this time whether Congress will permit the new policy of the PTO on the patentability of non-naturally-occurring, non-human animals to go into effect. Shortly after the Com- missioner's announcement, the Senate adopted an amendment to a supplemental appropriations bill, H.R. 1827, 100th Cong., 1st Sess. (1987), that would have barred the use of funds appropriated for fiscal year 1987 "for the purpose of granting any patent for vertebrate or invertebrate animals, modi- fied, altered, or in any way changed through engineering technology, including genetic engineering." Id., amendment no. 245, 133 CONG. REC. S7268 (daily ed. May 28, 1987); see Senate Votes Morato- rium on Patenting of Animals, 34 Pat. Trademark & Copyright J. (BNA) No. 833, at 124 (June 4, 1987). The Senate amendment was stricken, however, in conference, after the PTO indicated that it would not grant patents on animals during the remainder of fiscal year 1987. 100th Cong., 1st Sess., 133 CONG. REC. H5654 (daily ed. June 27, 1987) (Joint Explanatory Statement of the Committee of Conference); 34 Pat. Trademark & Copyright J. (BNA) No. 838, at 277 (July 16, 1987). A currently pending House bill would impose a two-year moratorium on the issuance of animal patents. H.R. 3119, 100th Cong., 1st Sess., 133 CONG. REC. H7206 (daily ed. Aug. 5, 1987); see House Bill Would Impose Two-Year Moratorium on Patenting Animals, 34 Pat. Trademark & Copyright J. (BNA) No. 841, at 351 (Aug. 6, 1987). In the meantime, the House Subcommittee on Courts, Civil Liberties and the Administration of Justice has held hearings on the animal patenting issue. See House Panel Holds Third Hearing on Animal Patents, 34 Pat. Trademark & Copyright J. (BNA) No. 845, at 460 (Sept. 3, 1987); House Panel Resumes Hearings on Animal Patenting Issue, 34 Pat. Trademark & Copyright J. (BNA) No. 840, at 321 (July 30, 1987); Hearing Ponders Patentability of Geneti- cally Altered Animals, 34 Pat. Trademark & Copyright J. (BNA) No. 835, at 191 (June 18, 1987). 67. For example, Roman Saliwanchik states that "[ain inventor has the right to hold any inven- tion in secret as a trade secret." R. SALIWANCHIK, LEGAL PROTECTION FOR MICROBIOLOGICAL AND GarIC ENGINEERING INVENTIONS 9 (1982) (emphasis added). This statement is clearly incorrect if he means to refer to legal trade secrecy, see infra text accompanying notes 68-88, yet his use of the terms "right" and "trade secret" suggests that it is indeed legal protection that he has in mind. 68. Actual secrets may also receive legal protection outside of trade secrecy law, through laws against trespass or theft. Biotechnology 19871 Research business, actual secrecy may offer the only feasible strategy for protecting such discoveries. Trade secrecy is usually governed by state common law, although an increasing number of jurisdictions have enacted statutory trade secret pro- tection, " and a handful of federal statutes protect trade secret rights once they have accrued under state law. Trade secret doctrine varies some- what among jurisdictions, but the basic requirements are similar in most states. State law generally limits the permissible subject matter of trade secret protection and further requires that the owner of a trade secret demonstrate some measure of actual secrecy, reasonable efforts to main- 7 3 tain secrecy, and misconduct by the defendant in acquiring, using, or secret. the trade disclosing In most jurisdictions, the permissible subject matter is quite broad, but it does have limits that potentially exclude basic research discoveries. The 7 5 definition in the first Restatement of Torts, which many jurisdictions follow, is susceptible to such a construction: A trade secret may consist of any formula, pattern, device or compi- lation of information which is used in one's business, and which gives him an opportunity to obtain an advantage over competitors who do not know or use it. It may be a formula for a chemical compound, a process of manufacturing, treating or preserving mater- ials, a pattern for a machine or other device, or a list of customers ... [I]t is not simply information as to single or ephemeral events in the conduct of the business . . . A trade secret is a process or device for continuous use in the operation of the business. Generally 69. See infra note 79; ARK. STAT. ANN. §§ 70-1001-1007 (1985); IDAHO CODE §§ 48-801 to -807 (1984); MASS. ANN. LAWS ch. 93, §§ 42, 42A (Law. Co-op. 1982); N.M. STAT. ANN. §§ 30-16 to -24 (1978); N.C. GEN. STAT. §§ 66-152 to -157 (1985). 70. See, e.g., Freedom of Information Act, 5 U.S.C. § 552b(c)(4) (1982) (exempting trade secrets from disclosure obligation of government agencies); Trade Secrets Act, 18 U.S.C. § 1905 (1982) (prohibiting disclosure of trade secrets by federal employees). 71. See, e.g., SI Handling Systems v. Heisley, 753 F.2d 1244 (3d Cir. 1985); American Alloy Steel Corp. v. Ross, 149 Cal. App. 2d 215, 308 P.2d 494 (1957); Revcor, Inc. v. Fame, Inc., 85 Il. App. 2d 350, 228 N.E.2d 742 (1967). 72. See, e.g., SI Handling Systems v. Heisley, 753 F.2d 1244, 1256 (3d Cir. 1985); K-2 Ski Co. v. Head Ski Co., 506 F.2d 471, 473-74 (9th Cir. 1974); Emery Co. v. Marcan Prod. Corp., 389 F.2d 11, 16 (2d Cir. 1968); Koehring Co. v. E.D. Etnyre & Co., 254 F. Supp. 334, 339 (N.D. I1. 1966). 73. See, e.g., Revcor, Inc. v. Fame, Inc., 85 Il. App. 2d 350, 355, 228 N.E.2d 742, 745 (1967). 74. See, e.g., E.I. duPont de Nemours & Co. v. Christopher, 431 F.2d 1012, 1014-15 (5th Cir. 1970), cert. denied, 400 U.S. 1024 (1971); Franke v. Wiltschek, 209 F.2d 493, 500 (2d Cir. 1953). 75. RESTATEMENT OF TORTS, § 757 comment b (1939). 76. See, e.g., Ruesch v. Ruesch Int'l Monetary Servs., 479 A.2d 295 (D.C. 1984); Koch Eng'g Co. v. Faulconer, 227 Kan. 813, 610 P.2d 1094 (1980); Engineered Mechanical Servs. v. Langlois, 464 So. 2d 329 (La. Ct. App. 1984); Jet Spray Cooler v. Crampton, 385 N.E.2d 1349 (Mass. 1979); Delta Filter Corp. v. Morin, 108 A.D.2d 991, 485 N.Y.S.2d 143 (N.Y. App. Div. 1985); Felmlee v. Lockett, 466 Pa. 1, 351 A.2d 273 (1976). The Restatement (Second) of Torts covers neither trade secrets nor the broader topic of "Miscella- neous Trade Practices" under which trade secrecy was subsumed in the first Restatement. See 4 RESTATEMENT (SECOND) OF TORTS 1-2 (1979). The Yale Law Journal [Vol. 97: 177 it relates to the production of goods, as, for example, a machine or formula for the production of an article." This definition could be read to exclude information about ongoing re- search on the ground that it has not yet ripened into "a process or device for continuous use in the operation of a business." Trade Secrets Act "trade secret" under the Uniform The definition of is considerably more amenable than the Restatement definition to a con- struction that permits protection of knowledge generated in the course of scientific research. Section 1(4) of the Act defines "trade secret" as follows: "Trade secret" means information, including a formula, pattern, compilation, program, device, method, technique, or process, that: (i) derives independent economic value, present or potential, from not being generally known to, and not being readily ascertainable by proper means by, other persons who can obtain economic value from its disclosure or use, and (ii) is the subject of efforts that are reasonable under the circum- stances to maintain its secrecy. As the Commissioners' Comment to this definition makes clear, the Uni- form Trade Secrets Act goes beyond the Restatement to extend trade se- cret protection both to information not yet put to use and to negative in- formation. Thus, in those jurisdictions that have enacted the Uniform Trade Secrets Act, trade secret protection may be more readily available 77. RESTATEMENT OF TORTS, supra note 75, at § 757 comment b. 78. Another type of research information that might be denied protection under a strict reading of the Restatement is "negative information," such as data on failed experiments. A researcher might well wish to keep such information secret, since knowing to avoid these dead ends might save competitor time and money. Yet this kind of information could be viewed as "simply information as to single or ephemeral events in the conduct of the business" and therefore not protectible as a trade secret. The Uniform Trade Secrets Act was approved by the National Conference of Commissioners on Uniform State Laws in 1979, and approved again with amendments in 1985. UNIF. TRADE SECRETS Acr, 14 U.L.A. 537-51 (1980 & Supp. 1986). It has been adopted by state legislatures in California, CAL. CIv. CODE §§ 3426-3426.10 (West 1984), CAL. CIV. PROc. CODE § 2019(e) (West 1987); Connecticut, CONN. GEN. STAT. ANN. §§ 35-50 to -58 (West 1983); Delaware, DEL. CODE ANN. tit. 6, §§ 2001-2007 (1986); Indiana, IND. CODE ANN. §§ 24-2-3-1 to -8 (West 1982); Kansas, KAN. STAT. ANN. §§ 60-3320 to -3330 (1983); Louisiana, LA. REV. STAT. ANN. §§ 51:1431-:1439 (West Supp. 1984); Minnesota, MINN. STAT. ANN. §§ 325C.01-.08 (West 1980); Montana, MONT. CODE ANN. §§ 30-14-401 to -409 (1985); North Dakota, N.D. CENT. CODE §§ 47-25.1-01 to -08 (1983); and Washington, WASH. REV. CODE §§ 19.108.010-.940 (1981). 80. The pertinent language of the Commissioners' Comment reads as follows: The definition of "trade secret" contains a reasonable departure from the Restatement of Torts (First) definition which required that a trade secret be "continuously used in one's business." The broader definition in the proposed Act extends protection to a plaintiff who has not yet had an opportunity or acquired the means to put a trade secret to use. The definition includes information that has commercial value from a negative viewpoint, for example the results of lengthy and expensive research which proves that a certain process will not work could be of great value to a competitor. UNIF. TRADE SECRETS AcT § 1 comment, 14 U.L.A., at 543 (emphasis in original). 1987] Biotechnology Research for research discoveries than it is in jurisdictions that follow the Restate- ment definition. The requirements of actual secrecy and reasonable efforts to maintain secrecy create considerable practical difficulties. Although absolute secrecy is not required, ' the claimed trade secret may not be generally known or 8 2 readily ascertainable by proper means. Efforts to maintain secrecy are costly and may be foiled. Microorganism cultures are especially difficult to maintain as trade secrets because they are easily stolen without detection and propagate rap- idly. A small sample of a culture can supply a competitor with bountiful quantities of commercially valuable organisms. The practical difficulty in- volved in protecting such materials as trade secrets is illustrated by a noto- rious early biotechnology trade secret case involving microorganism cul- tures used by American Cyanamid to produce antibiotics and steroids. Although the cultures were maintained as proprietary materials within the company, employees purloined sample cultures of the microorganisms from an American Cyanamid research facility and sold them to foreign pharmaceutical firms. The incident led to criminal convictions of two American Cyanamid employees" as well as civil trade secret actions for 8 5 damages. The final requirement of a trade secret claim-wrongdoing by the de- fendant in acquiring, using, or disclosing the trade secret-further limits the effectiveness of trade secrecy as a strategy for protection in competitive research fields. This requirement insulates from liability anyone who de- 8 6 rives the trade secret through independent research, reverse engineer- 81. Trade secrecy is not lost through disclosure in the course of a confidential relationship or pursuant to an agreement to keep the secret-indeed, the law of trade secrets protects the confidential- ity of such disclosures. 82. See, e.g., SI Handling Systems v. Heisley, 753 F.2d 1244 (3d Cir. 1985); K-2 Ski Co. v. Head Ski Co., 506 F.2d 471 (9th Cir. 1974); A.H. Emery Co. v. Marcan Prod. Corp., 389 F.2d 11 (2d Cir. 1968); Sandlin v. Johnson, 152 F.2d 8 (8th Cir. 1945); Juliano v. Hobart Mfg. Co., 200 F. Supp. 453 (D. Mass. 1961), affd, 303 F.2d 830 (1st Cir. 1962). 83. See, e.g., E.I. duPont de Nemours & Co. v. Christopher, 431 F.2d 1012 (5th Cir. 1970), cert. denied, 400 U.S. 1024 (1971). Typical measures include labeling documents or laboratory notebooks "Confidential," limiting disclosure on a "need to know" basis, and obtaining confidentiality agree- ments from employees and other persons to whom disclosure is made. See Sobel, Preserving Trade Secret Protection, in PRACTISING LAW INSTITUTE, PATENT, COPYRIGHTS, TRADEMARKS, & LITER- ARY PROPERTY COURSE HANDBOOK SERIES No. 224, PROTECTING TRADE SECRETS 567 (1986) [hereinafter PROTECTING TRADE SECRETS]. 84. United States v. Bottone, 365 F.2d 389 (2d Cir.), cert. denied, 385 U.S. 974 (1966) (criminal prosecution under National Stolen Property Act). 85. American Cyanamid Co. v. Sharff, 309 F.2d 790 (3d Cir. 1962) (civil contempt proceeding against defendant who refused to answer questions in deposition in civil action for misappropriation of trade secrets); American Cyanamid Co. v. Fox, 140 U.S.P.Q. (BNA) 199 (N.Y. Sup. Ct. 1964) (civil action for misappropriation of trade secrets and breach of employment agreement). 86. See, e.g., Speedry Chem. Prods. v. Carter's Ink Co., 306 F.2d 328, 330 (2d Cir. 1962) (right to independent discovery of secret has long history); Greenberg v. Croydon Plastics Co., 378 F. Supp. 806, 812 (E.D. Pa. 1974) (others in field who arrive at independent discovery may use plaintiff's trade secret). The Yale Law Journal [Vol. 97: 177 ing,8" or information obtained from publicly available sources." Once the "secret" becomes generally known to other scientists through independent discovery, the first discoverer loses protection. Thus, even where state law extends protection to research discoveries, the protection may not last for long. Trade secrecy presents a further problem for scientists who are condi- tioned by traditional scientific norms and rewards. The parsimonious hoarding of information necessary to maintain secrecy interferes with sci- entific communication and frustrates the pursuit of scientific recognition. Public disclosure, such as publication in a scientific journal or presenta- tion of a paper, ends actual secrecy and with it trade secret protection. Moreover, publication violates the requirement that the owner of the 9 0 maintain its secrecy. reasonable efforts to trade secret take While both trade secrecy and actual secrecy foreclose scientific recogni- tion in exchange for protection that may prove fragile and short-lived, trade secrecy may be more disruptive of scientific communication than ac- tual secrecy. Actual secrecy is only one of the elements of a trade secret claim; compliance with the other requirements, such as reasonable efforts to maintain secrecy, may also impede scientific communication. A public disclosure that facilitates reverse engineering may be enough to terminate 9 1 the right to enforce a trade secret claim even before it ends actual se- crecy. If researchers who wish to earn recognition in the scientific com- 87. See, e.g., Smith v. Dravo Corp., 203 F.2d 369, 375 (7th Cir. 1953) (person may legally obtain trade secret through inspection and analysis of product). 88. Sinclair v. Aquarius Elec., 42 Cal. App. 3d 216, 226, 116 Cal. Rptr. 654, 661 (1974) (by definition, trade secret is something which has not been placed in public domain). 89. See, e.g., Fishing Concepts, Inc. v. Ross, 226 U.S.P.Q. (BNA) 692, 695 (D. Minn. 1985) (disclosure of claimed trade secret in advertising textbook ends status as trade secret). The more typical avenue for terminating trade secrecy through public disclosure is the sale of a product from which the secret can be reverse engineered. See, e.g., Midland-Ross Corp. v. Sunbeam Equip. Corp., 316 F. Supp. 171, 177 (W.D. Pa.), affd, 435 F.2d 159 (3d Cir. 1970) (sale of product ends trade secrecy where nature of trade secret is ascertainable by inspection of product). But cf. Smith v. Dravo Corp., 203 F.2d 369 (7th Cir. 1953) (fact that trade secret could have been discovered through inspection of product does not preclude trade secret action against defendant who instead obtained and used information improperly); Tabor v. Hoffman, 118 N.Y. 30, 23 N.E. 12 (1889) (surreptitious duplication of patterns for manufacture of pump actionable notwithstanding fact that patterns could be derived from pumps themselves with sufficient time and effort); Classic Instruments v. VDO-Argo Instruments, 73 Or. App. 732, 700 P.2d 677 (Ct. App. 1985) (use of blueprints com- municated in confidence was actionable notwithstanding fact that trade secrets could be ascertained from product sold on open market). 90. The content of this requirement varies with the circumstances of each case. Disclosure to persons who have no need to know the secret, or who are not obligated to maintain its confidentiality, ends trade secret protection. Sandlin v. Johnson, 152 F.2d 8 (8th Cir. 1945); Crown Indus. v. Kaw- neer Co., 335 F. Supp. 749 (N.D. Il1. 1971). Requiring reasonable efforts to preserve secrecy serves to alert potential misappropriators to the owner's claim of trade secrecy and to weed out spurious after- the-fact claims of trade secrecy involving information that has been freely disclosed. 91. See Van Prods. Co. v. General Welding & Fabricating Co., 419 Pa. 248, 264-65, 213 A.2d 769, 778 (1965) (patent specification destroys trade secrecy after patent granted). 92. For example, suppose the scientists at American Cyanamid who discovered the tetracycline- producing microorganisms had wanted to gain recognition for their discovery in the scientific commu- nity through publication, but the company was unwilling to release the microorganism culture to the public. The company might have permitted publication of a description of the research, provided that Biotechnology 1987] Research munity can publish general descriptions of their research while keeping essential aspects of their inventions to themselves, they may find actual secrecy to be a more agreeable strategy than legal trade secrecy. II. CONJUNCTION OF INTELLECTUAL PROPERTY AND BASIC RESEARCH IN BIOTECHNOLOGY A. Developments Bringing About the Conjunction Scientists working in biotechnology-related fields are increasingly likely to be concerned simultaneously with the norms and rewards of research science and the rules and incentives created by intellectual property law. Research involving recombinant DNA and hybridoma technologies can create new organisms with the capacity to make desired products."' The obvious commercial potential of these technologies has attracted the inter- est of industrial sponsors. The products of biotechnology research are often readily commercialized-especially when the research is conducted in affiliation with commercial companies. The interest of industrial sponsors in biotechnology research is not con- fined to product development. In this field, the traditional dividing line between basic and applied research is blurred. Not only has the historical time lag between the two collapsed, but it has become difficult to charac- 9 8 terize given research problems as belonging in one category or the other. the microorganism itself were withheld from the public. In so doing, the company might have compro- mised its position in the event of a subsequent trade secret action. Arguably, the publication would have constituted a failure to use "reasonable efforts to maintain secrecy" or made the invention "read- ily ascertainable by proper means." Nonetheless, even without a legal trade secret remedy, American Cyanamid might have been able to maintain the microorganism culture as an actual secret for a while by keeping it within the company. 93. See generally Grobstein, supra note 3. 94. Recombinant DNA technology can alter the genetic information of cells, enabling the cells to make desired proteins. Every living cell contains genetic information in the form of deoxyribonucleic acid (DNA) inherited from the progenitors of the organism of which that cell is a part. When cells replicate, the new cells in turn inherit a copy of the parent cells' DNA. DNA is a long molecule in the form of a double helix composed of two chains of subunits called nucleotides. One of the compo- nents of each nucleotide is a base, of which there are four different kinds: adenine, guanine, cytosine, and thymine. DNA provides the cell with instructions on how to make the proteins it needs to survive. These instructions are coded in the sequence in which the four bases occur. Each group of instructions corresponding to an individual protein is called a gene and comprises a small portion of one chain of the DNA molecule. All organisms use the same genetic code for protein synthesis, although different organisms make different proteins in accordance with the different DNA sequences in their cells. Recombinant DNA technology involves taking genetic information from one cell and splicing it into the DNA of another cell-or even combining the DNA of two different organisms. In this manner, simple, single-celled organisms such as bacteria can be programmed to make human proteins. The recombinant organism can then be used as a living factory to make the protein encoded in the cloned gene. See W. KEETON, BIOLOGICAL SCIENCE 63-64, 665-66 (3d ed. 1980). Hybridoma technology involves the fusion of antibody-producing lymphoid cells with tumor cells to create new cell lines that can make desired antibodies. 95. A 1984 study reveals that industry may be funding as much as one-quarter of all biotechnol- ogy research in universities, with nearly one-half of all biotechnology companies funding such re- search. Blumenthal, Gluck, Louis & Wise, supra note 1. 96. For example, cells contain more genetic information than they actually use. Learning what causes cells to "express" certain genes and not others would not only answer major questions in The Yale Law Journal [Vol. 97: 177 Academic and industrial researchers are often working on the same or closely related problems, whether competitively or collaboratively. " Note- worthy scientific discoveries are made in industrial laboratories, and pat- entable inventions are made in university laboratories. In addition, the capital markets are placing a dollar value on intellec- tual property long before a product is ready for market. The prospectuses that high technology companies use to attract investment capital advertise the companies' affiliations with university facilities and researchers. These affiliations enhance the companies' credibility with investors. They also supply companies with a window on new scientific developments, 9 9 which further expedites commercialization. Sometimes the scientists themselves have equity interests in the companies sponsoring their work. Even if they do not, however, research scientists tend to be more aware of the legal and commercial rights arising out of their work when their sponsors take an interest in those rights. When industry funds basic research, the scientists and universities in- volved may be contractually bound to preserve patent rights in sponsored research. But even publicly funded research is increasingly likely to be protected as intellectual property. Many universities have in-house patent or licensing departments or use the services of outside patent development 1 0 2 organizations. "' The 1980 Patent & Trademark Act Amendments en- courage universities to patent and license inventions made in the course of government-sponsored research. Under this legislation, universities must 0 3 report any invention which "is or may be patentable" to the sponsoring agency within a reasonable time; if they do not, the goyernment may re- 0 4 ceive title to the invention. Universities may elect to retain patent rights, subject to a non-exclusive license to the agency for use of the inven- 0 5 tion, but only if they agree to file timely patent applications once their 1 06 research results have been published. Government "march-in rights," through which the sponsoring agency can license the university's inven- biology, but might also suggest ways of facilitating the manufacture of desired proteins or suppressing the expression of disease-causing genes. This knowledge would therefore be valuable to both scientists and industry. 97. D. DICKSON, supra note 6, at 74-77; Blumenthal, Gluck, Louis & Wise, supra note 1; Grob- stein, supra note 3. 98. Commercialization Hearings, supra note 3, at 7 (testimony of Dr. Donald Kennedy); see, e.g., CALGENE, INC., PRELIMINARY PROSPEcTus 25, 28-29 (May 23, 1986); INTEGRATED GENET- ICS, INC., AMENDMENT No. 1 To FORM S-1, at 17, 22, 25-26 (filed with SEC Apr. 24, 1986). 99. Peters & Fusfeld, supra note 6, at 34-36. 100. Blumenthal, Gluck, Louis, Stoto & Wise, supra note 3, at 1364. 101. M. KENNEY, supra note 1, at 74-77; Blumenthal, Epstein & Maxwell, supra note 9; Wil- liams, Business Aspects of Biotechnology 25, in GENETICALLY ENGINEERED ORGANISMS, supra note 102. Pub. L. No. 96-517, 94 Stat. 3019 (1980) (codified at 35 U.S.C. §§ 200-212 (1982)). 103. See 35 U.S.C. § 201(d); see also 35 U.S.C. § 202(c)(1). 104. See 35 U.S.C. § 202(c)(1). 105. See 35 U.S.C. § 202(c)(4). 106. See 35 U.S.C. § 202(a), (c)(2), (c)(3). Biotechnology Research 1987] tions itself if the university fails to do so, encourage universities to exploit 10 7 their patents. The statute also requires universities to share patent roy- 0 8 alties with inventors.' This personal financial stake gives researchers an incentive to be alert to patent rights. Finally, most biotechnology research makes use of patented inven- 10 9 tions. Thus, even researchers who have no interest in protecting their own intellectual property run up against the intellectual property system when they try to gain access to the inventions of others. In short, intellectual property law pervades contemporary biotechnology research, causing significant concern about the impact of secrecy and ex- clusive rights on the behavior of research scientists in fields where scien- tific norms call for publication and dedication of research results to the public. B. The Norms of Science in Contemporary Biotechnology Research Wholly apart from any conflict between the law of intellectual property and traditional scientific norms, contemporary biotechnology research presents scientists with a conflict between scientific norms and scientific rewards. As we have seen, in theory these norms and rewards are congru- ent: Publication of original research results both rewards the publishing scientist with professional recognition and satisfies the norm of permitting other scientists to replicate one's research results. But for research involv- ing the use of unique biological materials, such as bacterial strains and other types of self-replicating cells, publication in writing alone may not be sufficient to satisfy this replicability norm. To replicate the authors' results, subsequent investigators may need access to identical materials. By sharing access to unique materials, however, the publishing scientist not only enables other scientists to replicate her claims; she also allows them to compete with her more effectively in making new discoveries. The professional advantage that exclusive access to unique research materials gives her may motivate her to retain exclusivity for as long as possible in order to maximize the discoveries for which she can claim priority. Of course, scientists in other fields of research also run the risk of losing future claims to priority when they publish their current research results. But at some point reluctance to help the competition is likely to be over- come by desire to earn recognition for priority of discovery, which is es- tablished through prompt publication. Delaying publication may handicap the competition, but the dilatory scientist thereby runs the risk that some- 107. See 35 U.S.C. § 203. 108. See 35 U.S.C. § 202(c)(7)(C). 109. Cf Fox, Patents Encroaching on Research Freedom, 224 SCIENCE 1080 (1984) (discussing concern in biotechnology research community about impact of restrictions on research use of patented products and processes). 110. See infra notes 112-46 and accompanying text. The Yale Law Journal [Vol. 97: 177 one else will publish first and get all the credit. So long as scientists must publish their results in a form that permits replication in order to earn recognition for priority, the reward structure of science operates to rein- force scientific norms. In biotechnology, however, the requirement of publication is a less ef- fective mechanism for enforcing scientific norms than in some other fields. As noted above, replication of research using unique biological materials generally requires access to more than the published text. Since the mater- ials themselves cannot be made a part of the text, publishing scientists could conceivably gain recognition through publication without sharing their materials. Withholding materials is a relatively inconspicuous depar- ture from scientific norms. It occurs after publication and is not apparent from the written text."' Thus, publishing scientists with exclusive access to such materials have an opportunity to gain recognition while retaining a future advantage over their research competitors. This conflict between norms and incentives is aggravated when the materials (or the discoveries they facilitate) have potential commercial value. The records of a controversy that arose when a team of scientists from a commercial biotechnology firm submitted a manuscript to the Journal of Biological Chemistry (JBC) in the summer of 1980 dramatically illustrate 1 12 this conflict. The JBC, a scientific journal that frequently publishes reports of research using biological materials, had a policy requiring au- thors to make available to other researchers any biological materials re- 11 8 ferred to in published manuscripts. The manuscript submitted by the commercial scientists reported the development of a recombinant organism 1 1 4 containing a newly synthesized gene. The authors described in detail how they synthesized the DNA insert. But in response to a question from 111. A publication may designate expressly that materials are available from an independent source, such as a cell culture depository, but the absence of such a designation does not necessarily mean that the author would be unwilling to supply the materials. Some authors would rather supply the materials themselves in order to be sure of the identity and viability of the sample. See infra note 112. The subject of commercial biotechnology commanded considerable public attention at this time, which roughly coincided with the Supreme Court's Chakrabarty decision and the initial public offering of stock by Genentech, Inc. See Hamilton, Biotech's First Superstar, Bus. WEEK, Apr. 14, 1986, at 68. 113. This "availability" policy was set forth as Instruction #c.4 in the written instructions to authors, published in each new volume of the JBC, as follows: The MICROORGANISMS and TISSUE CULTURE STRAINS referred to in a manuscript should be identified by an American Type Culture Collection number, or that of a comparable collection. If this is not possible, it is understood that the authors will make the particular strains used available to interested investigators upon request. Well mapped strains of Escherichia coli should be sent to the E. coli Genetic Center at Yale University. 255 J. BIOLOGICAL CHEMISTRY 2-3 (1980). 114. The authors reported the transformation of a well-known strain of bacteria with a recombi- nant DNA plasmid containing a newly synthesized gene. There were thus a number of different biological materials "referred to" in the manuscript: the original bacterial strain, the original plasmid, the synthesized gene, the recombinant plasmid containing the synthesized gene, and the transformed bacterial strain containing the recombinant plasmid. In keeping with the JBC's policy, the original bacterial strain was identified by a collection number. 19871 Biotechnology Research an editor, they indicated that their company's policy prevented them from giving out the recombinant strain. This event prompted an extensive discussion of the journal's "availabil- 1 15 ity" policy at a meeting of the JBC editorial board, culminating in an informal vote of approximately seventy-five percent in favor of retaining percent opposed.1 ' and twenty-five the policy The editorial board decided to seek the advice of an ad hoc committee"" consisting of six academic scientists and two industry scientists on the fol- lowing question: "Should this Instruction to Authors be retained as is, 115. The content of this discussion is summarized in an enclosure to Letter from William F. Harrington to Nathan 0. Kaplan (July 11, 1980) [hereinafter Summary] (on file at Yale Law Jour- nal). Dr. Harrington is on the faculty of the McCollum-Pratt Institute, The Johns Hopkins Univer-" sity. Dr. Kaplan is a professor in the Department of Chemistry, University of California at San Diego. 116. Arguments raised in favor of retaining the availability policy emphasized three core norms: 1. Replicability. Scientists should publish their research claims in a form that permits replica- tion so that the scientific community can verify their claims: "Once an experiment is published anyone should be able to reproduce it, otherwise it is worthless and might as well just not be published." Summary, supra note 115. Replication requires access to the publishing scientist's unique materials. 2. Dedication. Publishing scientists should make their scientific advances available to the public: [T]he act of publishing is the giving of a scientific advance to the public in a form which permits other scientists to reproduce the experiments. . . . [To withhold any necessary ingre- dient which is not generally available is morally equivalent to withholding essential informa- tion from the text of the publication and authors should in general be required to make availa- ble to other scientists those rare materials that go into the published experiments. Id. This dedication norm is violated by withholding essential information or rare materials that the author is able to supply. 3. Progress. Scientists should publish their research in a form that facilitates the progress of science: "The work cannot be repeated or extended without the availability of the strain or culture." Id. (emphasis added). 117. Arguments raised against the policy included: 1. Inconsistency. The policy is inconsistent in that it requires authors to make living materials available but does not require authors to make non-living materials available: We don't require distribution of organic compounds. . . .[A] complicated organic synthesis is more difficult to repeat than a DNA recombination experiment. . . .[I]t would be enough to insist (as in all other experiments) that enough experimental detail be given to know what was done and to permit, in most cases, repetition by a knowledgeable investigator. Id. The editors acknowledged parenthetically that "this [repetition] could not be done if a rare cell line from a clinical case is involved." Id. 2. Free riders. The policy permits free riders to benefit from the work of authors without compensating them: Such a rule forces an investigator in a university to supply his strain to a profit-making organ- ization without any financial return to his laboratory or university, regardless of the amount of profit eventually received. A similar argument concerns a single investigator developing a spe- cial strain after years of work. (One suggestion . . . is that the rule have a time-delay of one or two years before a culture has to be deposited ... . Id. 3. Nonenforceability. The policy would be difficult or impossible to enforce. If the JBC were to enforce the policy and other journals did not, authors would choose to publish elsewhere: [W]ould some authors not publish (or at least not in this journal) if such a requirement were enforced? Thus, would we be hindering, rather than helping, communication by such a policy? If all the journals don't have the same policy, would the JBC be losing papers in some of the most active and exciting fields? [As for those authors who do publish in the JBC,] [t]he journal cannot act as a policeman. Id. 118. Id. The Yale Law Journal [Vol. 97: 177 or deleted in the light of recent advances in molecular biology[?]" modified Each of the scientists responded by letter." These eight letters, written as of proprietary rights in biotechnology inventions were just be- the effects ginning to be felt, illustrate a range of views within the biomedical re- as to what the norms of science are, what pur- search community itself poses they serve, and how they actually function in practice. viewpoints. Two The six academic scientists represented three distinct 2 1 two favored favored retaining the policy without modification.' Another rather than continuing the policy as a statement of the views of the JBC, 2 2 the as a condition of publication.' The remaining two favored modifying to policy to limit it to those materials that are strictly necessary in order scientists permit replication of the published work. ' The two industry the need for were the most hostile toward the policy, questioning even 2 4 publication in replicable form.' 1. Retention of the Policy The two scientists who favored retaining the policy without modifica- 2 "5 tion' came closest to embracing Merton's traditional norms of commu- of science," "the nism and universalism, emphasizing "the best interest freedom of science,' and "the general ethic of bacterial and phage ge- neticists."12' Both rejected the argument raised at the editorial board com- meeting that, in treating organisms differently from new chemical 2 8 are pounds, the policy is inconsistent.' They pointed out that organisms chemicals can be difficult to synthe- relatively easy to reproduce, whereas size and supply.' One of the two expressly opposed a different rule for scientists were briefed on the issue with a 119. At the time their views were solicited, the eight at the editorial board meeting. See written summary of arguments raised for and against the policy and accompanying text. The scientists referred to these arguments in their indi- supra notes 115-17 vidual letters. 120. I am grateful to the editors of the Journal of Biological Chemistry and to Paul Berg, Donald Brown, William Harrington, Nathan Kaplan, Philip Leder, Daniel Nathans, Jesse Rabinowitz, Or- rie Friedman, and I.S. Johnson for permission to use their letters. 121. See infra notes 125-31 and accompanying text. 122. See infra notes 132-36 and accompanying text. 123. See infra notes 137-39 and accompanying text. 124. See infra notes 140-45 and accompanying text. 125. These two scientists were Dr. Nathan 0. Kaplan, see supra note 115, and Dr. Philip Leder, Institutes of Health and is now Chair- who was then with the Public Health Service of the National of Genetics at Harvard Medical School. See Letter from Nathan 0. Kaplan man of the Department (Aug. 5, 1980) [hereinafter Kaplan letter] (on file at Yale Law Journal); to William F. Harrington Letter from Philip Leder to William F. Harrington (July 25, 1980) [hereinafter Leder letter] (on file at Yale Law Journal). 126. Kaplan letter, supra note 125. 127. Leder letter, supra note 125. 128. See supra note 117. for any investigator to regrow the organism and 129. Dr. Kaplan explains: "[O]ne cell is enough thereby have the organism available for his studies. Living systems can be kept and maintained but chemicals can be utilized and not replaced unless someone redoes the synthesis or preparation." Kaplan letter, supra note 125. Dr. Leder simply noted that the two situations involved "clearly differ- ent . . . logistics . . . " Leder letter, supra note 125. 1987] Biotechnology Research recombinant organisms, reasoning that such a rule would lead to different 1 3 standards for academic and commercial scientists. The other acknowl- edged the existence of competing incentives both to protect intellectual property and to earn scientific recognition, but nonetheless opposed com- promising scientific norms to accommodate either "proprietary interests" 1 3 1 or "the all-too-human foibles of individual scientists. 2. Policy as Statement of Principle The two scientists who favored retaining the policy as a statement of 13 2 principle rather than as a binding contractual term conceded the impor- tance of the traditional norms' but nonetheless felt that the JBC was 3 4 unable to enforce the policy. Both recognized that the policy is some- 3 5 times violated in practice, and suggested that it be reformulated as an exhortation, rather than a requirement for publication. 3. Modification of the Poicy The remaining two academic scientists favored a narrower policy re- quiring availability only to the extent necessary for replication of pub- 1 3 7 lished experiments. Both drew a distinction in this regard between or- 130. See Kaplan letter, supra note 125. Dr. Kaplan objected to scientists earning recognition without observing scientific norms: "[Slome of the people involved in commercial laboratories would like to . . .receive recognition for their research yet not follow the general ethics of scientists ... One cannot have it two ways." Id. 131. Leder letter, supra note 125. 132. These two scientists were Dr. Paul Berg, Department of Biochemistry, Stanford University Medical School, and Dr. Donald Brown, Department of Embryology, Carnegie Institution of Wash- ington. See Letter from Paul Berg to William F. Harrington (July 31, 1980) [hereinafter Berg letter] (on file at Yale Law Journal); Letter from Donald D. Brown to William F. Harrington (July 28, 1980) [hereinafter Brown letter] (on file at Yale Law Journal). 133. Dr. Berg praised the objective embodied in the policy as "the ideal we should be striving towards" and "a principle that is nearly as defensible as motherhood." Berg letter, supra note 132. 134. Dr. Berg thought that "without comparable admonitions by the other journals. . . it would be a useless gesture." Id. Dr. Brown concurred that the idea behind the availability policy was sound, but thought that "to make it a law is a little like Prohibition, it is unrealistic and unenforceable." Brown letter, supra note 132. 135. Dr. Berg noted that the policy "reaches for a kind of openness and sharing in science that many of us sould [sic] like to see practiced universally." Berg letter, supra note 132. Dr. Brown felt sure that previous JBC authors had "refused or delayed sending [organisms] . . .to their competi- tors," and added that "[t]he problem is not made new by recombinant DNA proliferation, nor by its increasing popularity with industry." Rather, he attributed the reluctance of scientists to share strains to the fact that they "want time to explore their discoveries." Brown letter, supra note 132. 136. Dr. Berg proposed that the JBC "construct a statement that sets forth clearly and forcefully the JBC's belief on [sic] the responsibilities of researchers and authors to make available, along with the data of their investigations and experiments, the biological material used in acquiring that infor- mation" and urge other journals to do the same. Berg letter, supra note 132. Dr. Brown favored refraining the policy to state that authors are "encouraged" rather than required to make their strains available. Brown letter, supra note 132. 137. These two scientists were Dr. Daniel Nathans, Department of Microbiology, The Johns Hopkins University School of Medicine, and Dr. Jesse C. Rabinowitz, Department of Biochemistry, University of California, Berkeley. See Letter from Daniel Nathans to William F. Harrington (Aug. 7, 1980) [hereinafter Nathans letter] (on file at Yale Law Journal); Letter from Jesse C. Rabinowitz to William F. Harrington (Aug. 28, 1980) [hereinafter Rabinowitz letter] (on file at Yale Law The Yale Law Journal [Vol. 97: 177 ganisms isolated from nature and organisms synthesized by the authors, arguing that authors should be required to make only the former availa- ble. They justified this distinction primarily on the basis of the repli- cability standard." Both also noted that departures from the policy were not uncommon in the case of newly-synthesized organisms, even among 3 9 noncommercial scientists.' 4. Nonenforcement The two industry scientists both asserted that if the policy were en- forced, the JBC would lose manuscripts from scientists in commercial lab- 4° oratories. One of them noted that enforcement of the policy could ulti- mately undermine the goal of disseminating new knowledge by deterring scientists in commercial laboratories from publishing,"" and added that 4 2 the problem is not confined to commercial scientists. He added that Journal). 138. Dr. Nathans thought that the JBC should also encourage authors to make available new mutants or recombinants described in published reports, as well as new isolates from nature, in order to facilitate research. But although it would be "desirable" to have such mutants or recombinants freely available once they have been reported in the literature, he felt that it is neither desirable nor possible "to coerce people into giving out their materials." Dr. Nathans nonetheless favored retaining the policy as a binding condition of publication for "new isolates from nature of microbes or tissue culture cells" where "the principle of verifiability of research results operates." Nathans letter, supra note 137. Dr. Rabinowitz felt that both mutants and new isolates from nature should be made available because "it is not possible to characterize such organisms exactly or completely or to be certain that two organisms isolated by identical experimental procedures are, in fact, genetically identical unless they can be compared by genetic analysis." Rabinowitz letter, supra note 137. He noted, however, that where the manuscript provided a description of the experiment sufficient to permit replication, as would be the case for products of chemical synthesis and some recombinant organisms, it would not violate scientific norms to withhold the materials. Id. 139. Dr. Nathans noted that scientists who have constructed a new organism "may want to limit distribution of strains that were difficult to construct or especially important for their future work." He added that commercial incentives were compounding the problem: "Add to this an increasing number of recombinants with potential commercial value and you have a widespread interest in con- trolling the availability of new strains." Nathans letter, supra note 137. Dr. Rabinowitz observed that investigators often avoid making available organisms that they have derived themselves, yet these transgressions have generally been overlooked. He attributed the reluc- tance of authors to share their strains not to competing commercial interests, but to the interests of the authors in retaining exclusive access to a new research tool: "[Miost investigators recognize that when mutants or organisms isolated by enrichment techniques are involved, . . . free exchange of these materials has frequently not occurred. The usual reason for this has probably been to protect the advantage that the original investigator possesses having isolated an organism." Rabinowitz letter, supra note 137. 140. These two scientists were Dr. Orrie M. Friedman, then President and Scientific Director and now Chairman of the Board of Collaborative Research, Inc., and Dr. I.S. Johnson, Vice Presi- dent, Lilly Research Laboratories, A Division of Eli Lilly and Company. See Letter from Orrie M. Friedman to William F. Harrington (Aug. 4, 1980) [hereinafter Friedman letter] (on file at Yale Law Journal); Letter from I.S. Johnson to William F. Harrington (Aug. 18, 1980) [hereinafter Johnson letter] (on file at Yale Law Journal). 141. I can say that under some circumstances commercial labs such as ours clearly would not be able to make certain of such strains available publicly. If as a result we would be barred from publication it would be regrettable from the point of view of both the individual scientist who did the work and of the others who would be deprived of the results of the work. Friedman letter, supra note 140. 142. Dr. Friedman asserted that "[t]he fact is that ... organisms . . . referred to in published Biotechnology Research 1987] necessary to ensure replicability of re- availability of strains may not be 1 3 scientist, going a step further, questioned sults. The other industry whether the JBC should even require that authors make available those materials necessary for replication. He noted that a policy requiring authors to enable readers to reproduce their work would "reduce publica- intellectual tion incentive," citing competing incentives to preserve property. of the policy "as a Ultimately, the JBC decided to retain the wording than as a contractual agreement between the statement of principle, rather Journal and an author." The chairman of the ad hoc committee concluded the statement, that a weakening of "understood" to "urged" or "en- such as changing the word couraged," would be interpreted as a compromise in values set by most agree should not occur. Moreover, since the the Journal, which in the past, we see no statement seems to have worked satisfactorily need to alter it at this time as long as the Journal makes no attempt an oral or written pledge or to enforce the statement by requiring authors. the ethics of policing otherwise a straightforward conflict A superficial view of this episode suggests norms and competing incentives arising from the desire between scientific The ultimate decision of the JBC to preserve intellectual property rights. of the policy the same but not to enforce it lends to keep the wording support to the view that while the traditional norms continue to hold some moral appeal for scientists, the research science community may be losing its ability to enforce its own norms against competing commercial incentives. papers often are not generally available upon request, often even from academic investigators." He defended this practice on the ground that "in particular instances this [complete access] would place an intolerable burden upon the investigator." Id. to make this point: "Moreover it may not be 143. Dr. Friedman used the language of patent law essential that they be available, provided the publication contains sufficient information so that one the organism etc. on truly skilled in the art (as opposed to a casual practitioner) is able to reproduce (patent specification must enable any person "skilled in the his own." Id.; cf. 35 U.S.C. § 112 (1982) and use invention). art to which it pertains" to make 144. I cannot accept the position, expressed by some, that a manuscript that is not wholly enabling is valueless. Many such manuscripts add to the fund of information, sometimes in a highly significant manner ... As a matter of fact, many manuscripts are readily accepted for publication even though they report work that, for one reason or another, cannot or will not be verified by others. Johnson letter, supra note 140. Dr. Johnson illustrated his point with the following example: "[A]n investigator collects thousands of sheep pituitaries and, by an arduous and lengthy procedure, succeeds in isolating a few milligrams of a previously unreported hormone. It is doubtful that the work will be verified independently. Nev- Id. ertheless, a manuscript reporting it would be unconditionally accepted for publication." 145. Dr. Johnson noted that such a policy would leave an author "with the options of disclosing His decision, all too often, will be to not publish." Id. the trade secret or not publishing. 146. Letter from William F. Harrington to Nathan 0. Kaplan (Feb. 3, 1981) (on file at Yale Law Journal) (discussing suggestions of ad hoc committee). The Yale Law Journal [Vol. 97: 177 But a closer reading of the record and a broader view of the context changes the picture somewhat. First, the letters illustrate that even where the norms are relatively clear, scientists often violate them for reasons un- related to commercial incentives. The most widely shared norm seems to be that publishing scientists should enable subsequent investigators to rep- licate their research results. All of the scientists invoked this replicability norm in their analysis of the functions served by the policy. Even the scientists who favored modifying or eliminating the availability require- ment nonetheless felt that the Journal should at least encourage availabil- ity of strains to promote replicability of results. Strikingly, however, most of the committee members observed that academic scientists as well as commercial scientists often depart from the norms-including even the 1 4 7 widely-shared replicability norm-in practice. This suggests that the problem does not originate with commercial interests or intellectual prop- erty incentives. Indeed, the arguments raised against the policy focused on tensions within the scientific community rather than conflicting incentives from the outside. No one adopted the argument raised at the editorial board meet- ing that the policy unfairly benefited "free riders" at the expense of pub- 14 8 lishing scientists, perhaps because of a sense that "free riding" on the published work of other scientists is an integral feature of scientific re- search. Nonetheless, several of the scientists cited the desire of individu- als to retain the exclusive benefits of a discovery for themselves as an ex- planation, if not a justification, for departures from the norms and for the 1 50 the policy. of enforcing difficulty consequent If the financial rewards of intellectual property protection tempt scien- 147. See, e.g., Brown letter, supra note 132; Friedman letter, supra note 140; Nathans letter, supra note 137; Rabinowitz letter, supra note 137. Even Dr. Kaplan, who came closest to claiming that the replicability norm is in fact descriptive of the behavior of academic scientists, went only so far as to state that "the academic people . . . are, from my experience, usually willing to give up the organism." Kaplan letter, supra note 125 (emphasis added). In a personal interview, Dr. Kaplan conceded that academic scientists sometimes fail to observe this norm in practice. He gave me a copy of a facetious form letter, then circulating as a joke among scientists, declining a request for a sample of a strain and setting forth a list of possible excuses to be selected or deleted as appropriate. The letter reads in part as follows: I regret to inform you that we are unable to send the strains you wish because* we have a graduate student who is studying this particular strain and it would be unfair to distribute it at this time/the problem you outline does not seem very significant/we have just concluded your particular study/we suspect you are lying about the purpose for which you want the strains/ we suspect you are the referee with whom we had considerable trouble recently. I might add that I, personally, am not in favour of withholding mutant strains. *Delete as appropriate The joke itself shows tension within the academic community over adherence to the norms. (This letter is on file at The Yale Law Journal.) 148. See supra note 117. 149. R. MERTON, supra note 18. 150. See, e.g., Brown letter, supra note 132; Johnson letter, supra note 140; Nathans letter, supra note 137; Rabinowitz letter, supra note 137. No one took up the suggestion raised at the editorial board meeting that it was particularly unfair to compel academic scientists to make their materials available to profit-making organizations. See supra note 117. 19871 Biotechnology Research tists to depart from scientific norms, the records of the JBC controversy suggest that the temptation may simply reinforce conflicting incentives within the reward structure of science itself. Even among purely academic scientists, the norms may never have had the force that the community would like them to have. Given that professional recognition rewards only the first researcher to arrive at a particular result, it is not surprising that scientists are reluctant to give their materials to their research 1 5 1 competitors. Moreover, when the JBC episode is considered in light of subsequent responses to the same problem by other scientific journals, it is not at all clear that scientific norms are losing ground. In the seven years since the JBC first confronted this issue, a number of prominent scientific journals have adopted policies similar to the JBC's. This spate of new policies addressing the issue may signify a number of things. It may indicate that the scientific community believes its norms are threatened and in need of fortification. Perhaps increasing violations of these norms have led jour- nals to respond with express policy statements. Another possibility is that there was no need for specialized policies for biological materials until the recent increase in recombinant DNA research. Thus, the biotechnology boom may have made it necessary to clarify the implications of longstand- 1 53 materials are involved. norms where biological ing Whatever the explanation, the recent record hardly suggests that tradi- tional scientific norms are crumbling in the face of intellectual property incentives as commercial interest in biotechnology research becomes more 1 5 4 pervasive. Although there are signs of tension over adherence to the norms, commercial pressures appear to be only a new aggravation of a familiar problem. 151. See supra text accompanying note 111. 152. The following journals now have express editorial policies requiring or encouraging the availability of biological materials referred to in manuscripts: 46 CELL No. 6 (Sept. 12, 1986) ("Infor- mation for Contributors"); 117 DEV. BIOLOGY No. 1 (Sept. 1986) ("Information for Authors"); 165 J. BACTERIOLOGY ii (Jan. 1986) ("Instructions to Authors"); 132 J. GEN. MICROBIOLOGY No. 7 (July 1986) ("Instructions to Authors"); 57 J. VIROLOGY ii (Jan. 1986) ("Instructions to Authors"); 153 VIROLOGY No. 2 (Sept. 1986) ("Notice to Authors"). 153. In fact, several journals have more general policies requiring authors to give sufficient infor- mation about their materials and methods to permit reproduction of their experiments, although they make no specific reference to biological materials. Such policies are set forth in, e.g., 156 ANALYTICAL BIOCHEMISTRY ix (1986) ("Instructions to Authors"); 56 BIOLOGY OF THE CELL 93 (1986) ("In- structions to Authors"); 32 CLINICAL CHEMISTRY 2 (Jan. 1986) ("Information for Authors"); 97 J. BIOCHEMISTRY iii (1985) ("Instructions to Authors"); 23 J. CLINICAL MICROBIOLOGY iii (Jan. 1986) ("Instructions to Authors"); 137 J. IMMUNOLOGY No. 6 (Sept. 15, 1986) ("Information for Contributors"); 187 J. MOLECULAR BIOLOGY No. 2 (Jan. 20, 1986) ("Instructions to Authors"); 232 SCIENCE xi (1986) ("Information for Contributors"). 154. See generally OFFICE OF TECHNOLOGY ASSESSMENT, COMMERCIAL BIOTECHNOLOGY: AN INTERNATIONAL ANALYSIS (1984). Indeed, a recent survey of over 1200 faculty members at 40 major universities in the United States found that biotechnology researchers with industrial support both publish at higher rates and patent more frequently than colleagues without such support. Blumenthal, Gluck, Louis, Stoto & Wise, supra note 3, at 1361. The Yale Law Journal [Vol. 97: 177 C. Patent Rights in Biotechnology Inventions Without patent protection, scientists who wanted to preserve intellec- tual property rights in their discoveries would have to resort to either ac- tual secrecy or trade secrecy. As we have seen, these two strategies conflict with scientific norms and rewards in that they preclude disclosure of the 5 5 discoveries for which protection is sought.' Patent law interferes with scientific communication less than either actual secrecy or trade secrecy because it is premised on disclosure rather than secrecy. Patent law pro- motes disclosure in two ways. First, in order to obtain a patent, an inven- tor must include in the patent application a full description of the inven- tion and how to make and use it. Although this application is confidential while pending,1 it becomes freely available to the public once a patent issues. Second, by granting a property right that survives disclosure, patent law removes an obstacle to disclosure of inventions that would otherwise have to be kept secret in order to preserve their commer- 1 58 value. cial On the other hand, patent law may be less consistent with scientific norms than secrecy protection in a different respect. Patent law grants a much more extensive right to exclude others from the use of a discov- ery,' thus potentially foreclosing further research to a greater degree than secrecy. A patent provides a remedy against all who make, use, or sell the invention without permission, however innocently, whereas trade secret law provides a remedy only against wrongdoers.' ' For exam- ple, independent discovery is not a defense to a patent infringement suit, although it is a defense to a claim for misappropriation of trade secrets. Because of the broad scope of the exclusive rights granted by a patent, patent law may offer considerably stronger protection than trade se- crecy-as well as greater potential disruption of ongoing research-in competitive research fields. By providing such broad exclusive rights, patent law may aggravate preexisting conflict between scientific norms and the reward structure of science. Scientific norms call for the dedication of discoveries to the re- search community without restrictions on subsequent use. But the reward structure of science grants recognition to the first scientist to make a dis- covery, creating an incentive to foreclose other scientists from doing com- petitive research. Ordinarily, the need to publish in order to gain recogni- 155. See supra text accompanying notes 89-92. 156. See 35 U.S.C §§ 111, 112 (1982). 157. See 35 U.S.C. § 122 (1982). 158. Indeed, once they are assured of an enforceable property right, patent owners may have an incentive to disseminate information about their inventions in order to attract potential licensees. See Kitch, The Nature and Function of the Patent System, 20 J.L. & ECoN. 265, 278 (1977). 159. See infra notes 207-42 and accompanying text. 160. See infra notes 211-13 and accompanying text. 161. See supra notes 86-88 and accompanying text. 1987] Biotechnology Research tion constrains the incentive to keep discoveries out of the hands of research competitors. Although publication may help competitors to make future discoveries, the reward structure impels scientists to make this con- tribution to the scientific community in order to gain current recognition for discoveries already made. But if patent law gives inventors the right to exclude others from using their discoveries in competitive research even after publication, it could potentially upset this balance of incentives. A system of exclusive rights that survives disclosure could thus undermine the mechanisms of the scientific community for gaining free access to new knowledge for use in subsequent research. I now turn to a more detailed consideration of the disclosure obligations and exclusive rights provided under the patent laws. 1. Disclosure A comparison of the patent disclosure requirements for microorganism- related inventions with the scientists' analysis of the functions of the JBC's availability policy shows considerable congruence between patent law policy and scientific norms concerning disclosure. If anything, patent law appears more rigorous than the scientific community with respect to the substance of its disclosure requirement. The two systems differ signifi- cantly, however, regarding the timing of disclosure. Patent disclosure is unlikely to occur until after a patent issues,"' which is often years later than disclosure to the scientific community would otherwise occur. a. Enablement Sections 111 and 112 of the Patent Act require an inventor to set forth in her patent application a written description of the invention, an expla- nation that will "enable" a skilled practitioner to "make and use" the invention, and the "best mode" contemplated by the inventor of "carrying out" her invention. This disclosure requirement serves two functions. First, it ensures that the public will inherit the full benefit of the inven- tion upon expiration of the patent term. Second, it permits the PTO to determine that the applicant has in fact developed an operative, useful embodiment of the invention. The first function parallels scientific norms by calling upon patent applicants, like publishing scientists, to dedicate 162. The statement in text assumes that the inventor has filed only a United States patent appli- cation. If the applicant also seeks foreign patent protection under the patent laws of other nations or international patent protection under the Patent Cooperation Treaty or the European Patent Conven- tion, the application will be published 18 months after the date of first filing. Patent Cooperation Treaty, done June 19, 1970, art. 21, 28 U.S.T. 7647, 7666-67, T.I.A.S. No. 8733, at 7666-67; European Patent Convention, October 5, 1973, art. 93, translated in K. HAERTEL, EUROPEAN PAT- ENT CONVENTION 19, 62-63 (1980), reprinted in 78 PAT. & TRADEMARK REv. 31, 39 (1980). See generally R. SALIWANCHIK, supra note 67, at 139-46. 163. 35 U.S.C. §§ 111, 112 (1982). [Vol. 97: 177 The Yale Law Journal their inventions to the public. The second function parallels the scien- calling upon patent applicants, like publish- tists' replicability standard by ing scientists, to supply sufficient information and materials to demon- strate to a knowledgeable audience that they have in fact achieved what claim. they materials, the patent dis- When an invention involves unique biological themselves if the written closure obligation extends to the materials called for expressly in the statute is insufficient to enable rep- description lication of the applicant's invention. Although microorganisms were not commonly considered patentable subject matter until the Supreme Court decided Diamond v. Chakrabarty, patent protection has long been available for processes using microorganisms and for the products of such 1 6 7 by processes. With the commercial development of antibiotics produced of bacterial strains, inventors sought a way to make enabling disclosures such inventions. Where the bacteria involved were generally available, the by explaining how to procure the necessary disclosure could be made 1 6 materials. ' But where a new or rare strain was involved, the only way to enable others to make or use (in patent law parlance, to "practice") the invention was to supply the strain. For these cases, patent attorneys de- vised the procedure of depositing samples of bacterial strains with recog- nized depositories from which the strains would become available to the public upon issuance of the patent.' 164. See infra notes 172-75 and accompanying text. 165. See infra note 175 and accompanying text. 166. 447 U.S. 303 (1980). The majority in Chakrabarty cited three prior instances in which patents had issued on microorganisms. The first was a patent issued to Louis Pasteur on yeast in and 1968. Id. at 314 1873, and the second and third were patents issued on microogranisms in 1967 n.9. 167. Note, Microorganisms and the Patent Office: To Deposit or Not to Deposit, That Is the Question, 52 FORDHAM L. REv. 592, 595-96 (1984). 168. See, e.g., Merck & Co. v. Chase Chem. Co., 273 F. Supp. 68 (D.N.J. 1967). 169. See, e.g., Ex parte Kropp, 143 U.S.P.Q. (BNA) 148 (P.T.O. Bd. App. 1959); Ex parte Davisson, 118 U.S.P.Q. (BNA) 520, 521 (P.T.O. Bd. App. 1957); Levy & Wendt, Microbiology and Applications, 37 J. PAr. a Standard Format for Infra-Red Absorption Spectra in Antibiotic Patent OFF. Soc'y 855, 855-59 (1955). It is interesting to compare the statutory disclosure requirements under the Plant Patent Act, 35 Act, 7 U.S.C. §§ 2321-2583 (1982), U.S.C. §§ 161-164 (1982), and the Plant Variety Protection those under the Patent Act. The Plant Patent Act, first promulgated in 1930, ch. 312, 46 Stat. with 376, extends patent protection to new varieties of asexually reproduced plants. The disclosure require- § 112, except for the following qualification: "No plant patent ment is the same as that in 35 U.S.C. section 112 of this title if the description is as com- shall be declared invalid for noncompliance with plete as is reasonably possible." 35 U.S.C. § 162. This provision has been construed to permit disclo- sures failing to satisfy the section 112 requirement of a description of the manner and process of making the invention. See In re Greer, 484 F.2d 488, 490-91 (C.C.P.A. 1973); cf. Funk Bros. Seed Co. v. Kalo Inoculant Co., 333 U.S. 127, 133-35 (1948) (Frankfurter, J., concurring) (arguing that patent for mixed culture of bacterial inoculants should be denied not because combination of bacterial strains was "work of nature," but because, inter alia, description of invention was inadequate for Patent Act allows patent protection for some failure to identify particular strains used). The Plant with section 112. inventions that it might not be possible to describe in accordance 1542, provides The Plant Variety Protection Act, enacted in 1970, Pub. L. No. 91-577, 84 Stat. by the Department of Agriculture for novel varieties of sexually patent-like protection administered reproduced plants. The disclosure provisions expressly call for both a written description and a Biotechnology Research 1987] The Court of Customs and Patent Appeals first addressed the adequacy 7 0 of microorganism deposits under section 112 in In re Argoudelis.1 The discussion of the function of deposits in this and subsequent cases parallels the analysis of the JBC's availability policy by the scientists. In re Argoudelis predates the JBC controversy by ten years. It involved a patent application on new antibiotic compounds and a process for pre- paring the compounds from microorganisms. The applicants deposited samples of their microorganisms in the permanent culture collection of the United States Department of Agriculture before the filing date of their patent application and referred to this deposit in the specification. At the request of the applicants, the depository agreed not to release the organ- ism without their written permission until a patent issued. Reversing the PTO, the Court of Customs and Patent Appeals ap- proved the use of a deposit to "disclose" a microorganism obtained from nature. The court held that section 112 did not require that the general public have access to the culture before the patent issued, reasoning that as long as the materials are available to the PTO during the pendency of the patent application, the disclosure is sufficient to demonstrate comple- tion of the invention as of the filing date. Since patent applicants in other fields need not make their patent disclosures public unless and until a patent issues, the court felt that no earlier public disclosure should be 1 7 required for microorganism-related inventions. In a concurring opinion, Judge Baldwin emphasized the dual role played by enablement in patent law. First, enablement ensures that the 1 7 2 public will receive its quid pro quo for the patent monopoly. Second, it demonstrates that as of the filing date the applicant is entitled to a pat- 1 7 3 ent. The first function does not call for public disclosure until the pat- deposit: An application for a certificate recognizing plant variety rights shall contain: (2) A description of the variety setting forth its novelty and a description of the genealogy and breeding procedure, when known .... (3) A declaration that a viable sample of basic seed necessary for propagation of the variety will be deposited and replenished periodically in a public repository .... 7 U.S.C. § 2422. In requiring a deposit in addition to a written description, this more recent statute follows the approach of the courts in interpreting section 112 of the Patent Act. 170. 434 F.2d 1390 (C.C.P.A. 1970). 171. Id. at 1393. 172. The first aspect of that role is to provide the assurance that the public will, in fact, receive something in return for the patent grant. This consideration is, of course, the full and complete disclosure of how to make and use the claimed invention. Thus, the patent adds a measure of worthwhile knowledge to the public storehouse . ... Id. at 1394. 173. Most, if not all, of the determinations involving the patentable merits of a claimed inven- tion vis-a-vis the prior art require that the application filing date be regarded as the prima facie date of invention . . . . Because the filing date is so important in determining patent rights, it is essential that there be no question that, at the time an application for patent is filed, the invention claimed therein is fully capable of being reduced to practice (i.e., that no technological problems, the resolution of which would require more than ordinary skill and reasonable time, remain in order to obtain an operative, useful embodiment). The Yale Law Journal [Vol. 97: 177 7 4 application contain ent issues." It does, however, require that the patent enough information for the PTO to determine whether the public will receive an enabling disclosure as of the issue date. The second function requires only that the disclosure demonstrate successful invention as of the filing date. Since Argoudelis' specification, including the deposited orga- nisms, was sufficient to show successful invention by the filing date, and the disclosure would become available to the public upon issuance of a 112.175 satisfied section specification patent, the Id. at 1395 (emphasis in original). 174. It should be apparent, however, that this first aspect of the enabling disclosure require- ment of section 112, requires only that the adequacy of the teaching disclosure be measured as of the issue date of the patent. There is no sense in making an applicant publicly disclose any part of his invention, much less its very essence, before he has been assured that he will obtain the protection he is seeking in return for that disclosure. Id. at 1394-95 (footnote omitted). 175. Judge Baldwin later reiterated this analysis in a majority opinion in In re Hawkins, 486 F.2d 569 (C.C.P.A. 1973), a case that did not involve microorganisms. The same approach has been followed in subsequent cases analyzing the use of microorganism deposits to satisfy the enablement requirement. In Feldman v. Aunstrup, 517 F.2d 1351 (C.C.P.A. 1975), the court extended the reasoning of In re Argoudelis to a case involving the deposit of a microorganism in a foreign depository. At the time the patent application was filed, the deposit was subject to a restriction that the organism could not be released without the depositor's consent. This restriction was removed after the filing date and before issuance of a patent. Id. at 1352-53. The Court held that the specification was sufficient to determine whether the inventor was entitled to a patent as of the filing date, and that release of the restrictions ensured that the specification would be fully enabling to the public upon issuance of a patent. Id. at The court also held that deposit in a foreign country could satisfy the enablement requirement. Id. at 1356. This issue is now moot because the United States has become a signatory to the Budapest Treaty on the International Recognition of the Deposit of Microorganisms for the Purposes of Patent Procedure, Apr. 28, 1977, 32 U.S.T. 1241, T.I.A.S. No. 9768. See 961 Official Gazette Pat. Off. 21-36 (Aug. 23, 1977). The Budapest Treaty establishes minimum requirements for maintaining an international depository for microorganisms and authorizes such depositories to receive and store de- posits and dispense samples in accordance with the Treaty and the patent laws of the signatory na- tions. 999 Official Gazette Pat. Off. 2 (Oct. 7, 1980). See generally Meyer, Problems and Issues in Depositing Microorganisms for Patent Purposes, 65 J. PAT. OFF. Soc'y 455 (1983) (discussing de- pository requirements under international treaties and section 112). The PTO responded to the Argoudelis decision by promulgating an administrative rule concerning the deposit of microorganisms. See MANUAL OF PATENT EXAMINING PROCEDURE, § 608.01 (c) (5th ed. 1983, rev. 1985). Under this rule, whenever an invention requires the use of a microorganism that is not known and readily available to the public, the applicant must deposit the organism not later than the filing date "in a depository affording permanence of the deposit and ready accessibility thereto by the public if a patent is granted." Id. The conditions of deposit must assure access to the deposit by the PTO during pendency of the patent application and unrestricted availability of the culture to the public upon issuance of the patent. The PTO has construed this rule strictly. For example, the PTO invoked the rule to invalidate an agreement by a depository to require that persons requesting samples of a microorganism identify the party making the request and undertake not to transmit the organism to third parties. Restrictions Voided on Availability of Deposited Yeast Strains to Public, 30 Pat. Trademark & Copyright J. (BNA) No. 732, at 104 (1985). In In re Lundak, 773 F.2d 1216 (Fed. Cir. 1985), the Court of Appeals for the Federal Circuit rejected the PTO's deposit rule insofar as it requires deposit before the filing date. That case involved a patent application on biological materials useful for producing immunoglobulins. The inventor ap- parently filed a patent application in the mistaken belief that the materials had already been placed in a depository. In fact, the materials were not deposited until one week after the filing date. The PTO rejected the application for failure to meet the enablement requirement of section 112. The Federal Circuit reversed. Following the reasoning of In re Argoudelis, the court noted that the materials were available from the inventor during the pendency of the patent application under a provision of the patent statute giving the PTO the right to request "specimens or ingredients" from the applicant. See 1987] Biotechnology Research This analysis appears congruent with scientific norms. Issuance of a patent brings about publication of the disclosure in the specification."' In patent law, as in science, a researcher may keep her discoveries secret until publication. Thereafter, she must surrender to the public an ena- 1 7 7 bling disclosure that permits replication of her claims. Indeed, in re- quiring access to materials through a source other than the inventor, pat- ent law goes beyond the requirements of the scientific community.' Two further respects in which patent law may impose tougher disclo- sure requirements than the scientific community deserve mention. The first concerns disclosures that permit replication only after further experi- mentation. The second concerns disclosures that, while fully enabling, conceal from the public the "best mode" contemplated by the inventor of practicing the invention. b. Undue Experimentation Recall that a number of the scientists consulted by the JBC thought that authors should not be required to supply newly-engineered organisms described in their publications, as long as the authors give enough infor- mation about the method of synthesis to permit other investigators to re- 35 U.S.C. § 114 (1982). The court deemed recourse to this statutory provision sufficient to satisfy any need for access by the PTO to verify successful invention. Although availability of materials from the inventor may be sufficient to allow the PTO to deter- mine constructive reduction to practice, it is plainly not a substitute for deposit in satisfying the en- ablement requirement once a patent issues. Otherwise, an inventor could effectively prolong the patent monopoly by restricting access to the materials after the patent expires. Cf White Consol. Indus. v. Vega Servo-Control, 713 F.2d 788 (Fed. Cir. 1983) (disclosure that incorporates trade secret available only by license from patentee fails to satisfy enablement requirement). Inventors could thereby cir- cumvent the terms of the bargain offered by the patent system: a monopoly on the invention for a limited term in exchange for dedicating the invention to the public thereafter. See Brulotte v. Thys Co., 379 U.S. 29, 31 (1964). The PTO recently issued an advance notice of proposed rulemaking relating to the deposit of bio- logical materials for patent purposes. 52 Fed. Reg. 34,080 (1987), reprinted in 34 Pat. Trademark & Copyright J. (BNA) No. 846, at 485 (Sept. 10, 1987). The proposed rule would update PTO practice in light of the decision of the Federal Circuit in In re Lundak; the Budapest Treaty on the Interna- tional Recognition of the Deposit of Microorganisms for the Purposes of Patent Protection, see supra; and the PTO's decision to extend patent protection to plants outside the Plant Patent Act in In re Hibberd, 227 U.S.P.Q. (BNA) 443, 444 (P.T.O. Bd. App. & Int. 1985). 176. In some countries, patent applications are published before a patent issues. See supra note and sources cited therein; see also Meyer, supra note 175. 177. The timing of disclosure may nonetheless be quite different under the two systems. See infra notes 201-06 and accompanying text. 178. Recall that a promise of availability from the author was an acceptable substitute for deposit under the JBC policy. See supra note 113 and accompanying text. Under some circumstances, scien- tists may be reluctant to make their organisms available through a depository. For example, scientists may not trust the depository to maintain and distribute viable culture samples that will continue to show the characteristics described in prior publications. See Hampar, Patenting of Recombinant DNA Technology: The Deposit Requirement, 67 J. PAT. OFF. Soc'y 569, 605-07 (1985). Of course, patent law does not prohibit patent holders from maintaining their own cultures in addition to those main- tained by the depository and making samples from their own laboratories available to corroborate their research claims. Thus, although patent law may go beyond scientific norms in requiring that organisms be made available through a depository, it should not deter scientists who are willing to make their organisms available from seeking patent protection. [Vol. 97: 177 The Yale Law Journal peat what was done. Patent law also permits a disclosure that demands as long as replica- some further effort in order to replicate the invention, ' 8 0 The specification is, tion does not require "undue experimentation." however, defective if replication requires "more than the skill of the art" ' ' to which the invention pertains or "an inordinate amount of work." The courts and the PTO have analyzed the "undue experimentation" standard in the context of microorganism-related inventions in Tabuchi v. 8 2 Jackson.2 Parte and Ex Nubel Tabuchi v. Nubel involved a priority contest between two inventors of a and isocitric acid using yeasts fermentation method for making citric acid The case turned on whether a specification that of the genus Candida. failed to supply a depository accession number for the particular Candida applicant satisfied the enabling disclosure requirement strain used by the of section 112.184 The applicant argued that it would have been possible to select a citric-acid-producing strain from publicly available Candida strains without undue experimentation by following the process disclosed in the specification. To prove his point, the applicant hired an expert in the field who read the application, selected a strain from Candida strains according to available to the public in depositories, and made citric acid the process described in the specification, all in a matter of fifteen working days. The court concluded that the disclosure "would not compel those of skill in the art to perform undue experimentation in selecting a ordinary '85 suitable strain" and that the enablement requirement was therefore satisfied. the discovery of three distinct microorganism Ex Parte Jackson involved same new antibiotic. The patent ap- strains, each of which produced the plicant classified the strains as belonging to a new species and deposited each of the three strains in a recognized depository. The application in- cluded claims to the process of producing the antibiotic from any microor- ganisms belonging to the same species as the deposited strains, ' and to the deposited strains or any mutations the same process employing 179. See, e.g., Friedman letter, supra note 140; Nathans letter, supra note 137; Rabinowitz letter, note 137; supra notes 137-45 and accompanying text. supra 180. W.L. Gore & Assocs. v. Garlock, Inc., 721 F.2d 1540, 1557 (Fed. Cir. 1983), cert. denied, 469 U.S. 851 (1984); White Consol. Indus., Inc. v. Vega Servo-Control, Inc., 713 F.2d 788, 791 (Fed. Cir. 1983). 181. 2 D. CHISUM, supra note 7, § 7.03141[a], [bi. 182. 559 F.2d 1183 (C.C.P.A. 1977). 183. 217 U.S.P.Q. (BNA) 804 (P.T.O. Bd. App. 1982). 184. Tabuchi claimed priority on the basis of a Japanese patent application. The issue before the court was whether this application met the enablement requirement of section 112 of the Patent Act, 35 U.S.C. § 112 (1982). Section 119 of the Patent Act, 35 U.S.C. § 119 (1982), allows inventors to provided that the claim United States patent priority on the basis of a filing date in a foreign country, application satisfies the disclosure requirements of section 112. Kawai v. Metlesics, 480 F.2d foreign 880 (C.C.P.A. 1973). 185. Tabuchi, 559 F.2d at 1189. 186. See 217 U.S.P.Q. at 805. Biotechnology Research 1 7 A divided Board of Appeals affirmed the rejection of the former thereof. claim, but reversed the rejection of the latter claim. The majority reasoned that the specification did not enable one skilled in the art to isolate other antibiotic-producing strains of the same species from nature without un- it was possible that the deposited strains due experimentation-indeed, the other were the only strains in existence suitable for the process."" On microorganisms are a common occurrence and "can be hand, mutations of 1 8 9 by a variety of known procedures." By deposit- intentionally produced ing the three strains, the patent applicant enabled skilled practitioners to derive mutant strains, and he was therefore entitled to extend his process mutants. claims to the use of such Here again patent law parallels and reinforces the scientific commu- nity's replicability norm. The "undue experimentation" cases on one hand reinforce the disclosure obligation and on the other hand limit the scope of invention. A patent applicant may what may be claimed as a patentable not, by withholding materials, leave too much work to be done by those invention. Conversely, the who would use the specification to practice the applicant is limited to claiming as her invention such advances as patent has fairly put within reach of the public. Patent applicants her disclosure to achieve who limit their disclosures so as to require inordinate effort replication thereby limit the scope of their monopoly. c. Best Mode Section 112 of the Patent Act further requires that the specification "set forth the best mode contemplated by the inventor" of carrying out the Unlike the enablement requirement, which is an objective invention."" focusing on the quality of the inventor's disclosure, the "best standard is a subjective standard focusing on the inventor's own mode" requirement knowledge and belief at the time the patent application is filed. ' The "best mode" requirement goes beyond the scientific community's repli- makes an enabling disclosure of an opera- cability norm. An inventor who tive version of her invention will not receive a valid patent if she knows of and conceals a better way to practice the invention at the time the applica- tion is filed.' Once again, patent law gives teeth to scientific norms through its tougher enforcement mechanisms. An applicant's failure to disclose the 187. See id. at 805-06. 188. Id. at 807-08. 189. Id. at 806. 190. 35 U.S.C. § 112 (1982). 191. 2 D. CHISUM, supra note 7, § 7.0511], [2]. 192. See, e.g., In re Gay, 309 F.2d 769, 772 (C.C.P.A. 1962) ("Manifestly, the sole purpose of [the best mode requirement] is to restrain inventors from applying for patents while at the same time concealing from the public preferred embodiments of their inventions which they have in fact conceived."). The Yale Law Journal [Vol. 97: 177 "best mode" of practicing the invention is unlikely to come to the attention of the PTO in the course of the ex parte patent examination procedure. But it may well surface in the course of discovery in a subsequent patent infringement suit. Should it appear at that time that the inventor con- cealed her "best mode" when she filed her patent application, the patent 1 9 invalid. be held will d. Patent Disclosure and Scientific Norms Aside from considerations of timing to be examined below,"' the disclo- sure requirements of patent law seem to fortify scientific norms to a de- gree that surpasses the willingness of the scientific community to enforce its own norms. Patent law reinforces the scientific community's interest in the replicability of claimed discoveries by requiring that a patent applicant make a disclosure of her invention that will enable persons skilled in the art to which the invention pertains to make and use the invention without undue experimentation. Patent law reinforces the scientific community's interest in having new discoveries dedicated to the public by requiring that the patent application set forth the best mode contemplated by the inven- tor of practicing the invention. Publication of the specification upon issu- ance of the patent and dedication of the claimed invention to the public upon expiration of the patent term further promote these interests. Scien- tists who protect their discoveries under the patent laws should therefore be at least as likely as scientists influenced solely by the rewards of science to disclose their inventions to the scientific community in a form that satis- fies scientific norms. In some circumstances, scientists may choose to forgo patent protection in order to circumvent enforcement of these norms. For example, scientists might want to restrict access to their organisms in order to keep the or- ganisms out of the hands of their research or commercial competitors. If the scientific community is willing to publish the research of these scien- tists while allowing them to keep their organisms to themselves, it may be possible for scientists to earn recognition in the scientific community while undermining the community's interest in replicability and public dedica- tion of research claims. Even if the publication is enough to forfeit legal trade secrecy protection, a strategy of publishing research results while maintaining the materials as actual secrets might permit scientists to by- pass the strictures of both patent law and scientific norms. They could thus earn scientific rewards and preserve the benefits of exclusivity with- out ever giving an enabling disclosure of their inventions. 193. See Biggart, supra note 52, at 2-73. 194. E.g., Engelhard Indus. v. Sel-Rex Corp., 253 F. Supp. 832, 836-37 (D.N.J. 1966), afl'd, 384 F.2d 877 (3d Cir. 1967). 195. See infra text accompanying notes 201-06. Biotechnology Research It might seem that the protection of a patent would be adequate to persuade inventors to forgo the narrower and more fragile protection of secrecy, since a patentee has the right to enjoin others from making, using, or selling the invention during the term of the patent even after disclo- sure. But the patent law exchange of disclosure for a right to exclude is not always attractive where secrecy is a feasible alternative. For one thing, an invention that can be exploited in secrecy by the inventor can often also be exploited in secret by a patent infringer. The unrestricted deposit of an organism necessary for replicating such an invention may "enable" infringements that cannot be detected. ' Cell lines and microor- ganisms can be propagated so easily that it is difficult under the best of circumstances to keep track of who has access to samples.' An inventor may choose instead to keep the organism to herself. Another reason why biotechnology inventors may be unwilling to de- posit their organisms in exchange for patent protection is that a patent may not be broad enough to protect inventors from research competition. For reasons more fully explored below, the use of patented inventions in competitive research may not constitute patent infringement. Scientists who make an enabling disclosure of their inventions for patent purposes may find that they have enabled their research competitors to gain scien- tific credit for subsequent discoveries to which the originator of the mater- ials might otherwise have eventually claimed priority. If the patenting sci- entist does not gain a corresponding right to exclude research competitors from using the materials, she might prefer to keep her materials to herself and forego patent protection. Of course, withholding research materials after publication to avoid being forestalled by research competitors violates scientific norms. But the reward structure of science sets up an incentive to maximize claims to priority of discovery that conflicts with these norms. And if patent law enforces its enablement requirement more zealously than the scientific community enforces its own norms, scientists may find 196. See Kewanee Oil Co. v. Bicron Corp., 416 U.S. 470, 489-90 (1974). 197. Of course, not all inventions may be exploited in secret. If an invention may be readily reverse-engineered, patent protection will be more attractive than futile efforts at secrecy. See R. SALIWANCHIK, supra note 67, at 10-11. 198. For example, suppose a company has created a new recombinant organism producing a pro- tein that is indistinguishable from human insulin. The recombinant product differs from previously- sold insulin solely in its lower cost of production. The company has a choice of either patent protec- tion or trade secrecy. If the company chooses patent protection, competitors will gain access to the organism once the patent issues. Since the recombinant insulin is identical to insulin produced by other means, it will be impossible to detect whether competitors are infringing the patent. The PTO's prohibition against agreements between inventors and depositories permitting inventors to monitor who gets access to deposits, see supra note 175, can only increase the likelihood of undetectable infringement. 199. The ease of propagation can also make it difficult to keep microorganisms as trade secrets or actual secrets. See supra notes 84-85 and accompanying text. 200. See infra notes 226-42 and accompanying text. Journal The Yale Law [Vol. 97: 177 that secrecy provides a more congenial strategy for maximizing their pro- fessional and commercial rewards. e. Timing of Disclosure There is one important respect in which the incentives arising from scientific rewards and those arising from the patent laws diverge: The timing of disclosure may be quite different for scientists seeking patent protection than for scientists seeking scientific recognition alone. Both the patent laws and the scientific community reward priority of invention, ' new claims. Nonetheless, thus placing a premium on prompt disclosure of scientists motivated solely patent applicants are likely to publish later than by scientific norms and rewards. One reason for this difference is that basic research discoveries might be ripe for recognition in the scientific 2 °2 community before they are ripe for patent protection. But if a scientist, guided by scientific norms and rewards, publishes results of research that has not yet yielded a patentable invention, she may be unable to obtain a 2 03 patent for inventions developing out of this research at a later date. of patent protection may therefore lead scientists to defer publica- Pursuit tion until their research has yielded patentable inventions, prolonging the they have filed patent applications and period of secrecy until after communication and progress. thereby slowing down scientific Once an inventor has a patent application on file, she may publish her for patent protection on research results without impairing her prospects 20 4 those inventions already claimed. Even then, however, patent applicants who are uncertain whether their inventions are patentable may choose to 2 0 5 defer publication until a patent actually issues. Publication prior to that thereby forfeits secrecy time is a risky proposition because the inventor 20 6 patent rights. A patent protection without any assurance of obtaining 201. R. MERTON, supra note 24, at 286-324; see 35 U.S.C. §§ 101, 102 (1982). 202. This disparity in timing is in part a consequence of patent law mechanisms for limiting patent protection to applied technology. See supra notes 38-47 and accompanying text. 203. There are two ways in which early publication could jeopardize the scientist's eligibility for a publication that discloses a subsequently claimed invention bars United patent protection. First, is filed within one year of the publication date. 35 States patent protection unless the application U.S.C. § 102. Foreign patent protection may be forfeited immediately upon publication. Second, pub- lication could have untoward consequences for subsequent patentability because of the nonobviousness requirement of 35 U.S.C. § 103. An invention is generally deemed to have been made on the date the inventor files a patent application, and the patentability of the invention turns on whether the inven- tion is obvious in light of the "prior art" as of that date. The prior art includes the applicant's own publication. See, e.g., Mass. Inst. of Technology v. AB Fortia, 227 U.S.P.Q. (BNA) 428, 431-32 publication discloses enough information to make the (Fed. Cir. 1985). If the inventor's own prior later-claimed invention obvious, the inventor will not be able to obtain a patent. See Chisum, Sources of Prior Art in Patent Law, 52 WASH. L. REv. 1 (1976). may impair the inventor's prospects for obtaining patent protection on 204. But such publication broader claims at a later date. See In re Ruscetta, 255 F.2d 687 (C.C.P.A. 1958). 205. If the inventor also applies for foreign patent protection, publication may occur 18 months after the application is first filed. See supra note 162. 206. In the United States patent system, secrecy may be preserved prior to the issuance of a patent. 35 U.S.C. § 122 (1982). 1987] Biotechnology Research application may languish in the PTO for several years before a patent ultimately issues. Thus, the combination of uncertainty as to the patent- able status of research discoveries and time lags in the PTO is likely to slow down the dissemination of information in the scientific community. In sum, notwithstanding the congruence between the substantive disclo- sure requirement of the patent laws and scientific norms, patent disclosure may occur considerably later than disclosure motivated solely by scientific norms and rewards. 2. Exclusive Rights Disclosure, of course, is only one side of the patent law bargain. Even if patent applicants make broader disclosures of their inventions than scien- tists motivated only by scientific norms and rewards, they do so in ex- change for a seventeen-year monopoly. Publication traditionally marks the end of exclusivity in science, but public disclosure through issuance of a patent marks the beginning of exclusive rights in the patent system. An "enabling" disclosure may be small consolation to a scientific community that is disabled from using it. Although the patent statute permits patent holders to exclude others from using their inventions, they do not always choose to do so. Depend- ing on the nature of the invention and its market, a patentee may find it advantageous to exploit the invention directly as a monopolist, to suppress the invention entirely, or to license others to exploit the invention on an 2 7 exclusive or non-exclusive basis in exchange for royalties. Licensing of patented inventions may pose special problems where the inventions are useful primarily for subsequent research rather than for commercial applications. The serendipitous nature of research discoveries may make it difficult to place a value on the right to use a patented inven- tion before the outcome of a research project is known. Moreover, most scientific research relies on many prior discoveries. If researchers needed to obtain licenses from each of the prior inventors on whose work they 20 8 build, the royalties and transaction costs could mount quickly. There are several reasons why a patentee might elect not to extend licenses to research competitors. The patentee might want to supress the 207. For an economic analysis of the exploitation of patent monopolies, see F. SCHERER, INDUS- TRIAL MARKET STRUCTURE AND ECONOMIC PERFORMANCE 379-99 (1970); see also Machlup, An Economic Review of the Patent System, Study No. 15 of the Subcomm. on Patents, Trademarks, and Copyrights of the Senate Comm. on the Judiciary, 85th Cong., 2d Sess. (1958) (economic overview of patent system as whole). Nonexclusive licensing of patented inventions minimizes interference with subsequent research. Patent rights on the basic process for recombinant DNA, owned jointly by Stanford and the Univer- sity of California, are licensed on a nonexclusive basis to all interested companies. See M. KENNEY, supra note 1, at 257-59. 208. See generally Gordon, Fair Use as Market Failure: A Structural and Economic Analysis of the Betamax Case and Its Predecessors, 82 COLUM. L. REV. 1600 (1982). The Yale Law Journal [Vol. 97: 177 invention in order to bolster its position in a related market, as happens when corporations build up extensive patent portfolios to prevent potential 2 9 licensees from using competitive technologies. " Or the patentee might fear that licensed use of the patented invention in further research will facilitate inventing around the patent, thereby undermining the future 2 1 ° value of the patent. Finally, the patentee might wish to preserve exclu- sivity in subsequent research in order to maximize future claims to prior- ity of discovery for purposes of both intellectual property and scientific credit. The extent to which patent law actually permits patent holders to en- join the use of their inventions by others in subsequent research is un- clear. On its face, the statute seems to forbid any unauthorized use of the patented invention. Section 154 of the Patent Act by its express terms grants to patentees "the right to exclude others from making, using, or selling the invention throughout the United States" during the patent term. " Section 271 of the Patent Act supplies a corresponding definition of patent infringement: "Except as otherwise provided in this title, who- ever without authority makes, uses or sells any patented invention, within the United States during the term of the patent therefor, infringes the patent. 112 According to the cases, this seventeen-year period of exclusive use is the consideration paid by the public for the disclosure of an inven- tion which it will be free to use once the patent expires. "' The enabling disclosure in the patent specification assures that the public will receive its 21 4 of the patent. the bargain upon expiration end of This commonplace formulation of the patent law bargain is indeed at odds with the traditional mechanisms of the scientific community for building upon prior research. Scientific research promotes the dissemina- tion of new knowledge to research competitors by awarding recognition to those who make their claims freely available to the community. Patent law, in contrast, promotes investment in innovation by providing exclusive rights in inventions. ' But other features of the patent statute indicate that this exclusivity is 209. F. SCHERER, supra note 207, at 392. Patents for inventions made by universities and small businesses in the course of federally-sponsored research cannot be held back from exploitation in this way. See supra notes 102-08 and accompanying text. 210. For an example of how this might occur, see infra note 242. 211. 35 U.S.C. § 154. 212. 35 U.S.C. § 271(a). A patentee may obtain an injunction against infringement "in accor- dance with the principles of equity," as well as damages "in no event less than a reasonable royalty" and up to three times the amount of actual damages in the discretion of the court. 35 U.S.C. §§ 283, 284. Reasonable attorney fees are also available to the prevailing party "in exceptional cases." 35 U.S.C. § 285. 213. See, e.g., Brulotte v. Thys Co., 379 U.S. 29 (1964); United States v. Dubilier Condenser Corp., 289 U.S. 178 (1933). 214. E.g., In re Argoudelis, 434 F.2d 1390, 1394-95 (C.C.P.A. 1970) (Baldwin, J., concurring). 215. See supra notes 17-28 and accompanying text. 216. Kewanee Oil Co. v. Bicron Corp., 416 U.S. 470, 480 (1974). Biotechnology Research not absolute. If the public had absolutely no right to make, use, or sell the patented invention until the end of the patent term, it would be somewhat puzzling to require that the patentee give the public an enabling disclo- sure of the invention at the beginning of the patent term. The require- ment of early disclosure suggests that certain uses of patented inventions during the patent term do not constitute patent infringement. Perhaps the most important function of early disclosure is to facilitate improvement of the patented invention or invention around the patent. Disclosure during the patent term also permits verification of the suffi- 2 18 ciency of the disclosure. Other practitioners in the field of the invention may be in a better position than the patent examiner to determine 2 19 whether a disclosure is enabling. Since an insufficient disclosure makes the patent invalid and unenforceable, those who have a use for the pat- ented technology will be motivated to uncover defects in the specification in order to avoid liability to the patentee. If an insufficient disclosure were not made public until the expiration of the patent term, the public would already have tolerated "the embarrassment of an exclusive pat- 2 2 ent" ' without receiving its quid pro quo. Disclosure at the outset of the patent term also reveals the scope of the patent claims, thereby giving notice to potential infringers (and to potential licensees) of the metes and bounds of the patent monopoly. Finally, early disclosure may facilitate "the Progress of Science and the useful Arts" by giving the public ac- 2 25 cess to those parts of the specification that the patentee does not claim. At least the first two of these functions-inventing around the patent and checking the sufficiency of the disclosure-seem to contemplate some 217. See generally Yarway Corp. v. Eur-Control USA, 775 F.2d 268, 277 (Fed. Cir. 1985) ("negative incentive" to "design around" other patented products is benefit of patent system); State Indus. v. A.0. Smith Corp., 751 F.2d 1226, 1236 (Fed. Cir. 1985) (same). The prevalence and social value of inventing around patented technologies are subject to dispute. Compare Turner, The Patent System and Competitive Policy, 44 N.Y.U. L. REv. 449, 455 (1969) (patent forces competitors to invest resources in finding noninfringing ways of achieving same result as patented invention, which may in many instances be wasteful if it diverts resources from other un- solved problems) with Adelman, The Supreme Court, Market Structure and Innovation: Chakrabarty, Rohm and Haas, 27 ANTrrusT BULL. 457, 464 (1982) (efforts to invent around pat- ent are unlikely to occur unless competitor and patentee have different views of cost of developing alternative technology and will not result in net loss to society because patented invention might never have been made without patent system). 218. See infra notes 232-34 and accompanying text. 219. Cf generally Loom Co. v. Higgins, 105 U.S. 580, 586 (1882) (patent specifications should be addressed to those skilled in art); Mowry v. Whitney, 81 U.S. (14 Wall.) 620, 644 (1871) (same). 220. See, e.g., Grant v. Raymond, 31 U.S. (6 Pet.) 218, 239 (1832). 221. Graham v. John Deere Co., 383 U.S. 1, 9 (1965) (quoting Thomas Jefferson). 222. United States v. Dubilier Condenser Corp., 289 U.S. 178, 186-87 (1933); Grant v. Ray- mond, 31 U.S. (6 Pet.) 218, 247 (1832). See Evans v. Eaton, 20 U.S. (7 Wheat.) 356, 434 (1822) (specification warns innocent user of patented invention against infringement and permits public to check whether patentee is entitled to patent). 224. U.S. CONsr. art. I, § 8, cl. 2; see supra text accompanying notes 38-39. 225. See generally Alexander Milburn Co. v. Davis-Bournonville Co., 270 U.S. 390 (1926) (dis- closure of unclaimed invention in specification of issued patent puts such invention in public domain). The Yale Law Journal [Vol. 97: 177 "use" of the disclosed invention during the term of the patent, if only in the laboratory. And while the statute itself does not expressly permit any use of patented technology during the term of the patent, the cases have recognized an exception to patent infringement for "experimental use" of an invention. It is difficult to discern the scope of this exception with any degree of precision. Since experimental use becomes an issue only in infringement actions, judicial pronouncements on its reach address situations where pat- entees have found a defendant's activities sufficiently annoying to be worth the trouble of pursuing a lawsuit. This factor has undoubtedly skewed the universe of experimental use decisions toward cases that impli- cate commercial interests. Within this universe, the experimental use de- 2 7 rarely sustained. raised and fense is frequently The experimental use doctrine was first expounded by Justice Story in dictum in an 1813 decision, Whittemore v. Cutter. " In that case, the trial court had entered judgment on a verdict for the plaintiff in a patent in- fringement suit. On appeal, the defendant argued that the jury had been erroneously instructed "that the making of a machine fit for use, and with a design to use it for profit, was an infringement of the patent right." Justice Story noted that the instruction was actually favorable to the de- fendant because it limited infringement liability to the making of a ma- chine with an intent to use it for profit. He supposed that this limitation "was adopted by the court from the consideration, that it could never have been the intention of the legislature to punish a man, who constructed such a machine merely for philosophical experiments, or for the purpose of ascertaining the sufficiency of the machine to produce its described 23 0 effects. 226. Roche Prods. v. Bolar Pharmaceutical Co., 733 F.2d 858, 862-63 (Fed. Cir.), cert. denied, 469 U.S. 856 (1984); Pfizer, Inc. v. International Rectifier Corp., 217 U.S.P.Q. (BNA) 157, 160-61 (C.D. Cal. 1982). See generally 4 D. CHISUM, supra note 7, at § 16.03(1); Bee, Experimental Use as an Act of Patent Infringement, 39 J. PAr. OFF. Soc'y 357 (1957); Hantman, Experimental Use as an Exception to Patent Infringement, 67 J. PAT. OFF. Soc'y 617 (1985). 227. The Court of Appeals for the Federal Circuit has called the defense "truly narrow," Roche Products v. Bolar Pharmaceutical Co., 733 F.2d at 863, and one commentator has argued that the defense is inconsistent with the language of the statute and the policies of patent law. See Bee, supra note 226, at 359, 377. 228. 29 F. Cas. 1120 (C.C.D. Mass. 1813) (No. 17,600); see also Sawin v. Guild, 21 F. Cas. 554 (C.C.D. Mass. 1813) (No. 12,391). The doctrine continued to appear solely in dicta for the next 45 years. See, e.g., Jones v. Pearce, Webster's Pat. Case 122, 124-25 (K.B. 1832); Byam v. Bullard, 4 F. Cas. 934, 935 (C.C.D. Mass. 1852) (No. 2,262). It finally found its way into a published holding in 1858, when the Circuit Court for the Southern District of New York approved a jury instruction that the use of a patented invention "merely for experiment, and not with a view to profit" would not give rise to infringement liability. Poppenhusen v. New York Gutta Percha Comb Co., 19 F. Cas. 1059, 1063 (C.C.S.D.N.Y. 1858) (No. 11,283). 229. 29 F. Cas. at 1121. The defendant objected to this instruction on two grounds: first, that as a matter of statutory construction the making of a machine without using it could not constitute patent infringement; and, second, that the mere making of a patented machine without actually using it causes no damage and thus could not be the basis for infringement liability. Justice Story rejected both of these arguments. 230. Id. at 1121 (emphasis added). Biotechnology Research 1987] Although subsequent cases elaborated the doctrine in confounding 23 1 ways, Justice Story's original articulation seems, at least in theory, to meet the needs of the research science community. The first exemption from liability for making a patented device-"philosophical experi- ments"-appears to protect "pure" research from patent infringement lia- bility, while the second-"ascertaining the sufficiency of the machine to produce its described effects"-appears to permit replication of research claims. Elsewhere in the opinion Justice Story elaborates a patent law correla- tive to the scientific community's interest in replicability. In objecting to a jury instruction that omissions from the patent specification rendering it nonenabling would not invalidate the patent unless the omissions were deliberate, the defendant argued that a materially defective specification should provide a defense to an infringement action even without proof of intentional deception. Story felt compelled to reject the argument under the language of the 1793 Patent Act, but was clearly sympathetic none- theless to the defendant's position: [T]he monopoly is granted upon the express condition, that the party shall make a full and explicit disclosure, so as to enable the public, at the expiration of his patent, to make and use the invention or improvement in as ample and beneficial a manner as the patentee himself. If therefore it be so obscure, loose, and imperfect, that this cannot be done, it is defrauding the public of all the consideration, upon which the monopoly is granted .... It was probably with a view to guard the public against the injury arising from defective specifications, that the statute requires the let- Later that year Justice Story reiterated his conception of the experimental use doctrine in the case of Sawin v. Guild, 21 F. Cas. 554 (C.C.D. Mass. 1813) (No. 12,391), as follows: [Tihe making of a patented machine to be an offence. . . must be the making with an intent to use for profit, and not for the mere purpose of philosophical experiment, or to ascertain the verity and exactness of the specification. In other words, that the making must be with an intent to infringe the patent-right, and deprive the owner of the lawful rewards of his discovery. Id. at 555 (citation omitted). 231. For example, some later decisions suggest that the rationale for the experimental use defense is that experimental use causes no harm. See, e.g., Byam v. Bullard, 4 F. Cas. 934, 935 (C.C.D. Mass. 1852) (No. 2,262); Albright v. Celluloid Harness-Trimming Co., I F. Cas. 320, 323 (C.C.D.N.J. 1877) (No. 147). Since Justice Story rejected the argument that making a patented ma- chine without causing harm cannot give rise to liability in Whittemore, see supra note 229, this cannot have been his rationale for the defense. 232. Section 6 of that Act reads in part as follows: Provided always,. . . That the defendant in such action shall be permitted to plead. . . that the specification, filed by the plaintiff, does not contain the whole truth relative to his discov- ery, or that it contains more than is necessary to produce the described effect, which conceal- ment or addition shall fully appear to have been made, for the purpose of deceiving the public. Patent Act of 1793, ch. 11, 1 Stat. 318 (1793), quoted in 6 D. CmisuM, supra note 7, at app. 10-3 (emphasis in original). Despite the language of the statute, the Supreme Court subsequently held in Grant v. Raymond, 31 U.S. (6 Pet.) 218 (1832), that a defendant in an infringement suit could avoid liability by showing that the specification was defective even without any intent on the part of the patentee to deceive the public. The Yale Law Journal [Vol. 97: 177 ters patent to be examined by the attorney general, and certified to be in conformity to the law, before the great seal is affixed to them. In point of practice, this must unavoidably be a very insufficient se- curity, and the policy of the provision, that has changed the common law, may be very doubtful. In recognizing an exception to patent infringement for use of the inven- tion "for the purpose of ascertaining the sufficiency of the machine to produce its described effects," Justice Story may have been trying to cor- rect for this deficiency in the 1793 statute. If, as Justice Story thought, examination by the attorney general prior to patent issuance is "a very insufficient security" against defective disclosures, giving the public a right to check the adequacy of the disclosure without liability for infringement provides additional assurance that the public has received its quid pro quo for the patent monopoly. The most effective scrutinizers of the patent dis- closure are likely to be potential infringers-or at least potential users of the patented invention. These people have the greatest knowledge of the relevant field and care the most whether the claimed invention works as described. The experimental use doctrine thus gives the patentee's compet- itors a right during the patent term to vindicate the public's interest in having an enabling disclosure. Although a few decisions have allowed the experimental use defense where the defendant simply wanted to see if the claimed invention worked, " hardly any cases have allowed the defense to excuse otherwise infringing activities that were conducted "merely for the purpose of philo- ' 23 5 sophical experiments. When commercial companies have claimed that their uses of patented inventions were for "philosophical experiments," 23 6 the defense has generally been denied. 233. 29 F. Cas. at 1122. 234. See, e.g., Dugan v. Lear Avia, Inc., 55 F. Supp. 223, 229 (S.D.N.Y. 1944); Akro Agate Co. v. Master Marble Co., 18 F. Supp. 305, 333 (N.D.W. Va. 1937). But cf. National Meter Co. v. Thomson Meter Co., 106 F. 531 (C.C.S.D.N.Y. 1900) (use of patented invention to see if it works constitutes threatened infringement by business competitor sufficient to support injunction). 235. These cases have recently been reviewed in Hantman, supra note 226. 236. See, e.g., United States Mitis Co. v. Carnegie Steel Co., 89 F. 343, 351 (C.C.W.D. Pa. 1898) (noting that defendant's use of patented invention "was a commercial use, extending over a period of several months, and involved a very large product"); Bonsack Machine Co. v. Underwood, 73 F. 206, 211 (C.C.E.D.N.C. 1896) (noting that machine "has not been made simply as an experi- ment, but has been used for profit, that is, for the purpose of selling the [defendant's] patent"); Al- bright v. Celluloid Harness-Trimming Co., 1 F. Gas. 320, 323 (1877) (No. 147) (holding use of patented invention in trial manufacture "is a technical infringement, and is sufficient to authorize an injunction restraining ... future use" but not sufficient for award of damages); Poppenhusen v. Falke, 19 F. Cas. 1048, 1049 (C.C.S.D.N.Y. 1861) (No. 11,279) (noting that defendants "are rivals of the complainant in the very business to which his patents relate"). The Robinson patent treatise, published in 1890, took the position that an experimental use of a patented invention in order to use or sell the products of the experiment or with a view to adapting the invention to the experimenter's business is "hostile to the interests of the patentee" and thus constitutes infringement. W. ROBINSON, THE LAW OF PATENTS FOR USEFUL INVENTIONS § 898 (1890). Consistent with this view, defendants who used patented inventions in the course of producing products for sale or performing services for customers have met with little success. E.g., Spray Refrig- 19871 Biotechnology Research A rare case allowing the experimental use defense for the use of a pat- ented invention in research is Ruth v. Stearns-Roger Manufacturing 2 37 Co. In that case, the defendant made unauthorized sales of patented machines and replacement parts to several customers, including the Colo- rado School of Mines. In awarding damages, the court adopted the recom- mendations of a special master, who had distinguished between direct in- fringement for the sale of whole machines and contributory infringement 2 3 8 for the sale of parts. The special master had calculated damages for direct infringement on all sales of infringing machines, including those sold to the Colorado School of Mines. In calculating damages for contrib- utory infringement, however, the special master had excluded sales of re- placement parts to the Colorado School of Mines on the ground that these parts "were for use in laboratory machines used for experimental pur- 23 9 poses, and consequently did not contribute to an infringing use."1 Be- eration Co. v. Sea Spray Fishing, Inc., 322 F.2d 34, 36-37 (9th Cir. 1963) (use of patented apparatus and method for freezing fish "for the purpose of experimentation as to the desirability of using this method" constituted patent infringement where use occurred in course of commercial fishing operation of kind which patent was designed to serve); Imperial Chem. Indus., PLC v. Henkel Corp., 545 F. Supp. 635, 656 (D. Del. 1982) (defendant's manufacture of 1200 pounds of patented chemicals for "pilot plant use" infringed patent where defendant supplied samples to potential customers and made it known in market that chemicals were commercially available); Northill Co. v. Danforth, 51 F. Supp. 928, 929 (N.D. Cal. 1942) ("Defendant's experiments were evidently not made for philosophi- cal or amusement purposes but were made in connection with his business as a manufacturer and salesman of anchors."); Sprout, Waldron & Co. v. Bauer Bros., 26 F. Supp. 162, 169 (S.D. Ohio 1938) ("The defendant cannot escape on the ground of experimental use where the machines were used to operate upon customers' products in the ordinary course of business."); Radio Corp. of Am. v. Andrea, 15 F. Supp. 685, 687 (E.D.N.Y. 1936), modified on other grounds, 90 F.2d 612 (2d Cir. 1937) (assembling parts of patented combination to test them was infringement where assembly was "not a scientific research or an engineering inquiry" but rather "a step which the defendants appar- ently deemed necessary in the manufacture and sale of their product"); Pairpearl Prods., Inc. v. Jo- seph H. Meyer Bros., 58 F.2d 802, 804 (D. Me. 1932) (use of patented process for extracting pearl essence infringed patent notwithstanding defendant's claim that use was experimental as incident to search for new agent, where pearl essence extracted through patented process was sold to customers); Cimiotti Unhairing Co. v. Derboklow, 87 F. 997, 999 (C.C.E.D.N.Y. 1898) (use of patented machine for customers "in the ordinary course of business" was patent infringement, notwithstanding defend- ant's claim that he was "experimenting" with machines in order to see if he could improve them). Two recent decisions have denied the experimental use defense to generic drug manufactuiers who performed clinical tests of patented drugs prior to expiration of the patents, reasoning that such testing relates to a commercial activity or the operation of a business. Roche Prods., Inc. v. Bolar Pharmaceu- tical Co., 733 F.2d 858, 862 (Fed. Cir. 1984), cert. denied, 469 U.S. 856 (1984); Pfizer, Inc. v. International Rectifier Corp., 217 U.S.P.Q. (BNA) 157, 160-62 (C.D. Cal. 1982). Congress responded to the decision in Roche v. Bolar by enacting the Drug Price Competition and Patent Term Restoration Act of 1984, Pub. L. 98-417, 98 Stat. 1585. That Act amends § 271 of the Patent Act to exempt from infringement liability the use of a patented drug solely to develop the information that must be submitted under federal law in order to obtain permission to sell the drug. See 35 U.S.C.A. § 271(e) (West Supp. 1987). 237. 13 F. Supp. 697 (D. Colo. 1935). 238. The unlicensed sale of a patented machine directly infringes the patent regardless of the use to which the machine is put by the buyer, but the sale of parts does not constitute direct infringement unless the parts are themselves the subject of patent claims. Selling unpatented parts with the inten- tion that they be used in a patented machine might nonetheless constitute contributory infringement, but only if the buyer's use of the parts itself constitutes direct infringement. See 35 U.S.C. § 271(a), (c) (1982); Aro Mfg. Co. v. Convertible Top Replacement Co., 365 U.S. 336, 341-42 (1961); 4 D. CHISUM, supra note 7, § 17.03. 13 F. Supp. at 703. The Yale Law Journal [Vol. 97: 177 yond this conclusory statement, the only information offered in the opin- ion as to the nature of the use is that "[tihese machines were all used in 24 0 the laboratory and were cut up and changed from day to day." The case offers no guidance for distinguishing a philosophical experiment from an infringing use."" The scope of the experimental use defense remains unclear, making it difficult to assess the degree to which the exclusive rights of patent holders might impede the progress of science. Justice Story's original formulation of the experimental use doctrine seems in principle to accommodate the interests of the scientific community in replicating and building upon new knowledge in basic research. The first prong of Justice Story's experimen- tal use privilege, permitting "philosophical experiments," is not well de- fined in the cases, but it seems to permit subsequent researchers to use the patented invention at least in traditional basic research with no commer- cial implications. The second prong, allowing the defense where the use is for the purpose of testing the adequacy of the specification, parallels the scientific community's interest in replication of published claims. The difficulty arises in trying to apply these principles when basic re- search has commercial value. As patent protection for basic research dis- coveries becomes more commonplace, and as the use of patented inven- tions in research becomes more of a threat to the interests of patent holders, the experimental use defense is likely to be litigated in new con- texts requiring careful analysis of its purposes and scope. The optimal extent of the experimental use defense cannot be deter- mined without attention to its likely effects on the scientific community. Too narrow a defense could stifle basic research and impair the commu- nity's mechanisms for validating and building upon new knowledge. Too broad a defense could cause industrial sponsors to lose interest in biotech- nology research or to rely on secrecy in lieu of patent protection. There may be no way to avoid both of these potential problems completely. The case for allowing the defense appears strongest where the subse- 24 2 quent user is attempting to devise alternatives to the patented invention. 240. Id. 241. In a recent review of experimental use cases, Ronald Hantman approved the Ruth decision while questioning its precedential value: This case is unlike any other experimental use case because it is the only one in which the exception is claimed for use at a university. Few would deny the experimental use exception for research on patented technology performed at a university in furtherance of its educational function. Because of the limited scope of the research in Ruth, the case probably has little precedential value for experimental use cases not involving university research. Hantman, supra note 226, at 633. 242. The facts of a pending lawsuit, Scripps Clinic & Research Found. v. Genentech, Inc., 3 U.S.P.Q.2d (BNA) 1481 (N.D. Cal. 1987), illustrate how the issue might arise. The lawsuit concerns technology for producing Factor VIII:C, a protein used in treating hemophilia. Plaintiffs hold patents on methods for recovering Factor VIII:C from human blood plasma in purified and concentrated form and on the purified and concentrated protein itself. The patented methods are costly, however, and involve a risk of transmitting disease. These difficulties make the protein a likely candidate for pro- 19871 Biotechnology Research In such a case, the interests of the research user are congruent with the interests of the public and the scientific community in advancing the state of human knowledge. The patent holder, by contrast, has an interest in prolonging the period in which the public is dependent on the patented technology. If the patentee sees the research user as a competitor rather than a customer, she may refuse to license the invention. Without an ex- perimental use defense, it is possible that no one would be able to build on the inventor's discovery until the patent expired. At the other end of the spectrum, the case for allowing the defense ap- pears weakest where the research user is essentially consuming a patented invention in an unrelated research effort-for example, by using a pat- ented laboratory machine. To allow such a user to avoid infringement liability on the ground that the machine was used in research would evis- cerate patent protection for technologies used primarily in research labo- ratories. The user falls squarely within the patentee's market, and there is no reason to suppose that the patentee would refuse to negotiate a license. If patent protection permits scientists to exclude the scientific commu- nity from making use of published research discoveries, it could under- mine a critical mechanism of the scientific community for facilitating the progress of science. Without the exclusive rights granted by the patent system, the reward structure of science encourages scientists to dedicate their results to the scientific community by requiring that they publish in order to gain recognition. This system promotes the interests of the com- munity in the validation and extension of new knowledge by overriding the individual scientist's contradictory incentive to hoard her findings. Dedication of new discoveries to the community necessarily entails facili- tating the work of one's competitors, but this is the price individual scien- tists must pay for current recognition. As the records of the JBC contro- versy show, the norms of dedication and replicability cease to be self- enforcing when it is possible for scientists to earn recognition without fully dedicating their results to the community-for example, when scien- tists can publish their research claims without supplying otherwise un- available materials to their competitors. By granting exclusive rights after publication, patent law may offer scientists another mechanism for earn- duction through recombinant DNA technology. Genentech is charged with infringing plaintiffs' patents in the course of cloning the gene for Factor VIII:O. Genentech used purified Factor VIII:C made by the plaintiffs' process to determine the se- quence of amino acids in the protein, and used this information to identify and clone in a recombinant organism the human gene responsible for manufacturing Factor VIII:C. The district court held that Genentech's use of the patented invention in its laboratories infringed the patent. Id. at 1486. Since Genentech evidently never asserted the experimental use defense, the court did not reach the question of whether the "philosophical experiments" prong of the experimen- tal use doctrine privileges research use of a patented invention by a commercial company. Had Genentech raised the issue, this might have been an appropriate case for allowing the experimental use defense (at least as to the method claims), since Genentech's goal was to invent around the patent, rather than to exploit the patented technology itself. The Yale Law Journal [Vol. 97: 177 dedication for sev- ing recognition while deferring payment of the price of enteen years. It remains to be seen whether the experimental use defense will develop into a means of reconciling scientific norms with the patent monopoly. D. An Illustration of the Conjunction A recently-settled dispute illustrates some of the problems that arise when scientists influenced by scientific norms and rewards seek to pre- serve intellectual property rights. The dispute erupted between competing teams of scientists at the National Cancer Institute (NCI) in Bethesda, Maryland, and the Pasteur Institute in Paris. Both teams of scientists had been doing research related to acquired immune deficiency syndrome (AIDS), separately and collaboratively, for several years. Their competi- tive efforts to secure intellectual property rights and scientific credit for invention of a test to detect the presence of antibodies to the AIDS virus in human blood became the subject of two legal proceedings: a breach of contract action in the United States Claims Court" and a patent interfer- 2 4 ence proceeding in the PTO. NCI scientists, under the direction of Dr. Robert Gallo, began studying 24 5 human retroviruses before the onslaught of the AIDS epidemic. In 1978, the scientists isolated the first human retrovirus, which they called human T-cell lymphotropic virus because of its affinity for T-lympho- cytes. The same group discovered a second human retrovirus, associated with leukemia, in 1982. They renamed the first human retrovirus 4 6 HTLV-I and called the new retrovirus HTLV-II. This early research laid the groundwork for the subsequent discovery that AIDS is caused by 2 4 7 yet another retrovirus. Scientists at the Pasteur Institute began studying AIDS under the di- rection of Dr. Luc Montagnier in late 1982. In February of 1983, they isolated a virus from the lymph nodes of a patient suffering from persis- tent lymphadenopathy. They called their virus lymphadenopathy- associated virus or LAV. This virus ultimately proved to be associated with AIDS. During this period, the NCI and Pasteur Institute scientists 243. Institut Pasteur v. United States, No. 730-85 (Cl. Ct. Dec. 12, 1985) (complaint reprinted as attachment C to Lipsey, Brezner & Majofis, Protecting Trade Secrets in Biotechnology, in PROTECT- ING TRADE SECRETS, supra note 83, at 807, 861); see Institut Pasteur v. United States, 10 Cl. Ct. 304 (1986), rev'd, 2 U.S.P.Q.2d (BNA) 1048 (Fed. Cir. 1987). 244. See Norman, AIDS Patent Negotiations Break Down, 232 SCIENCE 819 (1986); Altman, Ruling May Settle AIDS Test Dispute, N.Y. Times, Apr. 30, 1986, at 10 col. 3. its genetic information in the form of ribonucleic acid 245. A retrovirus is a virus that carries called (RNA) rather than deoxyribonucleic acid (DNA) and reproduces itself by using an enzyme reverse transcriptase to make a DNA strand corresponding to its own viral RNA. This DNA then integrates itself into the chromosomes of the host cell and uses the biological mechanisms of the host to replicate the virus. See Gallo, The First Human Retrovirus, SCi. AM., Dec. 1986, at 88. 246. Id. 247. See Gallo, The AIDS Virus, SCl. AM., Jan. 1987, at 47. 1987] Biotechnology Research 24 8 began communicating with each other and sharing biological materials. The NCI scientists supplied materials from their prior work with ret- roviruses, which allowed the Pasteur Institute scientists to identify LAV as a retrovirus distinct from either HTLV-I or HTLV-II. The Pasteur Institute published this finding in May of 1983.49 Despite the propensity of the LAV virus to kill the cells in which it was grown, the Pasteur Institute scientists were able to propagate the virus in small quantities. By August of 1983, they had developed a test that de- tected antibodies to LAV in twenty percent of serum samples taken from AIDS patients. They filed patent applications for their test in London in September of 1983 and in the United States in December of 1983, 25 1 priority. filing for purposes of date of the European claiming the In the meantime, the NCI scientists were also attempting to isolate virus from AIDS patients. While they found evidence of retrovirus in 25 2 AIDS patients, they were having difficulty growing the virus in culture. The Pasteur group twice sent samples of their virus to NCI, but the NCI scientists assert that they were unable to propagate the Pasteur vi- rus. Finally, in November of 1983, the NCI group found a cell line that could survive infection with the AIDS virus, making it possible to grow the virus in significant quantities. Now the NCI group was able to characterize the virus, which they called HTLV-III, and to develop a more sensitive test to detect antibodies to the virus in blood samples. NCI published its findings in May of 1984.255 A patent was issued on NCI's test kit in May of 1985.56 The terms under which these materials were shared, and the actual use made of the shared materials, are disputed. Institut Pasteur v. United States, No. 730-85 (Cl. Ct. Dec. 12, 1985). 249. BarrE-Sinoussi, Chermann, Rey, Nugeyre, Chamaret, Gruest, Dauguet, Axler-Blin, Vizinet-Brun, Rouzioux, Rozenbaum & Montagnier, Isolation ofa T-Lymphotropic Retrovirusfrom a Patient at Risk for Acquired Immune Deficiency Syndrome (AIDS), 220 SCIENCE 868 (1983). 250. Norman, Patent Dispute Divides AIDS Researchers, 230 SCIENCE 640, 642 (1985). 251. See supra note 184. 252. Norman, AIDS Priority Fight Goes to Court, 231 SCIENCE 11 (1986). 253. Complaint, Institut Pasteur v. United States, No. 730-85 (Cl. Ct. Dec. 12, 1985). 254. Norman, supra note 252, at 12. 255. Gallo, Salahuddin, Popovic, Shearer, Kaplan, Haynes, Palker, Redfield, Oleske, Safai, White, Foster & Markham, Frequent Detection and Isolation of Cytopathic Retroviruses (HTLV-III) From Patients With AIDS and at Risk for AIDS, 224 SCIENCE 500 (1984); Popovic, Sarnagadharan, Read & Gallo, Detection, Isolation, and Continuous Production of Cytopathic Retroviruses (HTLV- Ill)from Patients with AIDS and pre-AIDS, 224 SCIENCE 497 (1984); Shaw, Broder, Essex & Gallo, Human T-Cell Leukemia Virus: Its Discovery and Role in Leukemogenesis and Immunosuppression, 30 ADV. INTERN. MED. 1 (1984); Shupbach, Popovic, Gilden, Gonda, Sarnagadharan & Gallo, Sero- logical Analysis of a Subgroup of Human T-Lymphotropic Retroviruses (HTLV-IlI) Associated with AIDS, 224 SCIENCE 503 (1984). Almost two years later, controversy over priority of discovery was aggravated by the disclosure that electron micrograph pictures of the AIDS virus accompanying these publications depicted cells infected with LAV supplied by the Pasteur Institute rather than cells in- fected with HTLV-III isolated by the NCI scientists. See Gilden, Gonda, Sarnagadharan, Popovic & Gallo, HTLV-Il Legend Correction, 232 SCIENCE 307 (1986) (letter to editor); Norman, A New Twist in AIDS Patent Fight, 232 SCIENCE 308 (1986). 256. U.S. Patent 4,520,113. NCI also filed a patent application, Serial No. 602,946, on a method of continuous production of HTLV-III retrovirus from patients with AIDS and pre-AIDS. This application is still pending. See In re Gallo, 231 U.S.P.Q. (BNA) 496 (Comm'r Pat. 1986). The Yale Law Journal [Vol. 97: 177 The Pasteur Institute initiated legal proceedings to establish its priority of invention for purposes of both patent rights and scientific credit. In the patent interference proceeding, the Pasteur Institute asserted that its re- searchers were the first to isolate the AIDS virus and to develop an AIDS antibody test kit, and that it was therefore entitled to the patent rights on this invention. In the Court of Claims action, the Pasteur Institute alleged that NCI had violated express and implied contracts with Pasteur in pat- enting and licensing its own test kit and denying appropriate recognition to the Pasteur Institute scientists. Crucial to this latter action was the extent of NCI's actual use of the virus supplied by the Pasteur Institute in developing the NCI test kit. The Pasteur Institute asserted that it sup- plied the virus to NCI solely for "research purposes" and not for "indus- trial purposes," and that NCI's HTLV-III "is, or is identical to, the LAV strain first isolated by Pasteur." President Ronald Reagan and French Prime Minister Jacques Chirac announced the settlement of both pro- 2 55 by Chirac. during a visit to Washington on March 31, 1987, ceedings This dramatic conclusion to the conflict indicates the extraordinary degree of public interest in AIDS research. This controversy illustrates a number of the themes developed in the foregoing analysis of the interaction of scientific norms with intellectual property law. First, apart from intellectual property concerns, it shows the dilemma confronting scientists who seek current scientific recognition in the course of a competitive research effort. Each team of scientists bene- fited from work done by the other, and, conversely, each team facilitated the work of the other through its own disclosures. These disclosures un- doubtedly earned scientific recognition for the publishing scientists and advanced the progress of science, but they may also have given a boost to the competition. The dispute also highlights the difference in timing between a discovery that is ripe for scientific recognition and an invention that is ripe for pat- ent protection. The Pasteur Institute developed a more primitive test kit first; NCI developed a more sensitive test kit later. In the patent interfer- ence proceeding, NCI contended that Pasteur was not entitled to a patent because the test kit that was the basis for Pasteur's patent application was useless without a method for growing the virus in quantity, and that NCI rather than Pasteur was the first to develop such a method. This argu- ment may encourage biotechnology inventors to defer publication and dis- semination of their discoveries until they are confident that their inven- 257. Complaint, Institut Pasteur v. United States, No. 730-85 (Cl. Ct. Dec. 12, 1985). 258. Barnes, AIDS Patent Dispute Settled, 236 Sci. 17 (1987). Under the settlement agreement, the parties will share patent rights and jointly contribute 80% of their royalties on the antibody test kits to an AIDS research foundation. Both teams of scientists will be recognized as joint inventors of the test kits. As part of the settlement, Drs. Gallo and Montagnier signed a chronology of key discov- eries in AIDS research. Biotechnology Research 1987] tions are patentable. Had the Pasteur and NCI scientists followed such a strategy of interim secrecy, it might have taken additional months or years for either team to develop its AIDS screening test. The Pasteur Institute's efforts to restrict dissemination of LAV even after publication of its discovery demonstrate that the scientific community may enforce its disclosure norms less rigorously than the patent law en- forces its disclosure requirement. In this regard, it may be significant that Science magazine, where much of the research of both groups was pub- lished, did not at that time have a policy requiring publishing authors to make their materials available to other researchers, although it has since 2 59 adopted such a policy. Without such a requirement, the Pasteur Insti- tute was able to restrict access to LAV to its research collaborators and to prevent its use for "industrial purposes" or its dissemination to other scientists without their express permission. Restrictions on access to the virus would not have been possible if a patent had already issued on an invention requiring use of the virus." On the other hand, a patent might have given either team the right to enjoin the other from using the patented invention in subsequent research. Laboratory use of the virus by scientists seeking to develop an antibody test kit for ultimate sale in the market arguably falls outside the scope of 2 6 the experimental use defense. ' Fortunately, in this case neither team of scientists sought to enjoin the other's research. It is clear in retrospect, however, that the research of each team represented a threat to both the commercial and the professional interests of the other. Both of these inter- ests provide an incentive to impede the research efforts of competitors. But while the scientific community does not empower inventors to enjoin the use of prior discoveries in subsequent research, patent law may. CONCLUSION The commercialization of basic research discoveries in biotechnology and the availability of patent protection for such discoveries have gener- ated concern in the scientific community about the impact of intellectual 259. The Science policy now provides: "When a paper is accepted for publication in Science, it is understood by the editors that any materials necessary to verify the conclusions of the experiments reported will be made available to other investigators under appropriate conditions." 232 SCIENCE xi (1986) ("Information for Contributors"). The Pasteur Institute's restrictions on NCI's use of LAV might constitute "appropriate conditions" for making its materials available to other investigators within the meaning of the new Science policy. 260. See supra notes 175-78 and accompanying text. The dispute also illustrates the risk of unde- tectable infringement of patents involving the use of microorganisms. The Pasteur Institute contended that NOI used LAV to develop its test kit, while NCI insisted that it isolated its own virus. Subse- quent studies indicated that the two viruses differ in their genetic sequences by approximately 1.5%, compared to a difference of approximately 6% relative to an AIDS virus isolated by another labora- tory. Norman, HTLV-Ill and LAV: Similar, or Identical?, 230 SCIENCE 643 (1985). Since AIDS viruses can undergo spontaneous mutation, this genetic similarity is consistent with but does not prove use of LAV by NCI. 261. See supra notes 234-41 and accompanying text. Journal The Yale Law [Vol. 97: 177 in biotechnology- property law on the behavior of research scientists of intellectual property rules in light of scientific related fields. Analysis reveals substantial parallels between the two sys- norms and incentives requirements of patent law fortify scientific norms tems. The disclosure research results in replicable form. To the extent promoting publication of go beyond scientific norms, mandating broader that these requirements necessary to earn recognition in the scientific commu- disclosure than is in favor of se- nity, some inventors may choose to forgo patent protection strategies crecy. Trade secrecy and actual secrecy, while offering feasible considera- for protection of some biotechnology-related inventions, conflict they bly more with scientific norms and rewards than patent law in that patent entail substantial nondisclosure. Even for inventors who choose dis- protection over secrecy, patent disclosure is likely to occur later than and rewards. Moreover, if the closure motivated solely by scientific norms the patent statute forbids anyone but the patentee from making or using may be prevented from invention for any purpose, subsequent researchers inventions. Insofar as patent law permits inventors to building on patented keep research competitors from using their discoveries, it could aggravate existing conflict between scientific norms and the incentives created by sci- entific rewards. The experimental use doctrine offers a potential mecha- nism for reconciling the patent monopoly with the interest of the research community in building upon prior discoveries through subsequent research. Of course, it is open to question whether the patent system should do anything at all to avoid aggravating existing conflict between the norms structure of science. The patent system formulates rules to and the reward agenda, which may not coincide with the aims of the advance its own If patent rules and incentives conflict with the norms scientific community. perhaps the scientific community should either and rewards of science, adjust its own norms and rewards, or do without patent protection. are several reasons why it might make sense for the patent sys- There tem to make some adjustments as well. First, the two systems are not far apart. Although patent law has tradi- tionally operated in the realm of applied technology, whereas scientific and rewards have governed basic research, both systems promote norms the creation and dissemination of new knowledge by rewarding priority of discovery in exchange for disclosure. As patent law moves into areas tradi- tionally defined as basic research, it will be able to achieve its ends more of the scientific community in effectively if it draws from the experience to achieve similar ends. developing norms and incentives the incentives of private research sponsors with Second, harmonizing the norms and incentives of research scientists can help attract vitally- research projects. While private sponsor- needed private dollars to basic ship may have some undesirable side effects, such as distorting the re- Biotechnology Research search agenda of the scientific community, it nonetheless offers some pro- tection against the vicissitudes of political support for federal research funding. Third, the extension of patent protection to living organisms demands a fresh look at traditional patent doctrine. As the PTO and the courts ex- plore the special challenges posed by this new kind of intellectual prop- erty, sensitivity to the interaction of patent rules with scientific norms and rewards can facilitate the progress of science in these fields while preserv- ing incentives for private sector funding of research. Finally, the patent system will influence the behavior of research scien- tists more effectively if it takes into account the norms and incentives that guide behavior in the scientific community. Although patent law has avoided interference with basic research in the past, the two systems of rules and incentives are beginning to converge. Legal rules are more likely to succeed in influencing the behavior of scientists if they resonate with scientists' conception of appropriate behavior than if they call for counter- intuitive departures from the norms of science.

Journal

The Yale Law JournalUnpaywall

Published: Dec 1, 1987

There are no references for this article.