Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 256
Global Dimensions of Intellectual Property Rights in Science and Technology 11 Adapting the Intellectual Property System to New Technologies JOHN H. BARTON This chapter examines whether the intellectual property system is able to adapt to the current rate of change of technology.1 As the intellectual property system attempts to adapt, it must respond to increased system load as, for example, the number of patents increases. Yet, far more important, it must deal with changes in the character of technology itself. New technologies may require fundamentally new encouragement mechanisms and pose fundamentally new issues for the intellectual property system. Thus the question here is whether the system develops the appropriate new doctrines and mechanisms at a rate adequate to maintain incentives to innovation. This chapter uses as examples three of the fundamentally new technologies of our time: biotechnology, computer software, and computer-maintained and searched data bases. It describes the new issues posed by these technologies and reviews the approaches taken to adapt the intellectual property system in each case. It then evaluates the performance of this adaptation process, by looking at three levels: (1) the mechanisms for developing doctrine, (2) the systems (e.g., patent offices) that grant intellectual property rights, and (3) the formal systems (primarily courts) and informal systems (e.g., cross-licenses) that enforce intellectual property rights and shape 1 This is, in a sense, the inverse of the usual question—whether the intellectual property system encourages technological innovation.
OCR for page 257
Global Dimensions of Intellectual Property Rights in Science and Technology their practical economic implementation. In a number of the areas, the analysis is international and comparative; the conclusions are intended to be international as well. THREE PARADIGM TECHNOLOGIES The three paradigm technologies were chosen as those recent technologies that probably represent the most severe tests for the intellectual property system. Each requires new forms of expertise; for none can traditional legal principles be applied without substantial modification. Also, intellectual property protection is crucial for each of the three; for all, front-end costs are very large, copying of a marketed product is much easier than initial development, and product cycles are long enough that copying is a serious consideration.2 Biotechnology Biotechnology is defined here as genetic engineering and particularly recombinant DNA manipulation. Although it is different from more traditional areas such as pharmaceutical technology in that living organisms are involved and can often reproduce themselves, what has most troubled the early evolution of the law is the fact that many of the products and processes being patented derive directly from natural products and processes. Special Issues The early generation of biotechnology products consists of proteins found in nature in very small quantities, but available in volume through cloning. Typically, the researcher started with the protein and then identified the sequences of its amino acids and of the corresponding nucleic acids in the gene that codes for the protein's production. The question, then, is whether such a protein or sequence can be patented; in common language, it has been discovered rather than invented. This question is usually answered in the United States on the basis that a purified natural product can be patented (because it is not found in purified form in nature),3 but the 2 Thus, this chapter does not face the question of the irrelevance of intellectual property to many industries, such as those in which the product cycle is much shorter than the time required to obtain a patent. 3 In re Bergy, 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 (C.C.P.A.), 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).
OCR for page 258
Global Dimensions of Intellectual Property Rights in Science and Technology question is answered quite differently in the United Kingdom.4 Moreover, the question raises important international policy issues—if natural genes are patentable, it is very hard to maintain that countries (or individuals) hosting organisms containing potentially important genes should make them freely available for scientific purposes.5 The issue is not just theoretical or political. A number of firms seeking therapeutic products have often pursued exactly the same proteins. In a context in which there is no single "Aha" of invention, but rather a process of identification and sequencing, the priority choice is extremely difficult to resolve. A court ends up having to decide, for example, between two different proposed inventors, one of which may have isolated the protein first, while another has sequenced it first.6 The substantial delays that have marked the biotechnology patent area have intensified this problem, as competing firms have invested in research for a number of years without knowing which one will obtain the ultimate patent rights.7 This difficulty in identifying a specific point of invention is the fundamental problem underlying the current dispute over the National Institutes of Health patent application covering some 337 gene fragments sequenced as part of Craig Venter's cDNA approach to the human genome project.8 Traditionally (i.e., for the last decade or so), identification of the therapeutic value of a protein arose before its sequence was known; identification of the sequence became a particularly important step in defining priority for gaining patent rights. Now with the human genome project, the process is reversed. This project is developing and publishing the nucleic acid se- 4 Genentech Inc.'s Patent,  R.P.C. 147 (Ct. of App. 1988). Note, however, that law in the United Kingdom will permit patenting of an organism containing an inserted gene (e.g., a yeast containing the sequence coding for a protein to be produced) on the grounds that the product is novel. Thus, there is still a form of protection available for the genetic engineer-genes in the abstract have little economic value. 5 W. Tallent, Specific issues on proprietary rights, in Intellectual Property Rights Associated with Plants, Stuber et al., eds., Madison: Crop Science Society of America (1989), pp. 47-50 (discussion by U.S. Department of Agriculture official of impact of patenting on access to plant genetic resources); Moore v. Regents of the University of California, 51 Cal.3d 120 (1990). 6 The example cited in the text is Amgen, Inc. v. Chugai Pharmaceutical Co., 927 F.2d 1200 (CAFC 1991) (erythropoietin). For another example, involving a different factual situation, see Genentech Inc.'s Patent, supra (conflict with Wellcome), and Scripps Clinic and Research Foundation v. Genentech, 927 F.2d 1565 (CAFC 1991) (tissue plasminogen activase). 7 For genetic engineering, these delays were as long as 47.4 months in 1989. See General Accounting Office, Processing Delays Continue for Growing Backlog of Patent Applications (1990), cited in Office of Technology Assessment, Biotechnology in a Global Economy 212 (October 1991). 8 C. Anderson, U.S. patent application stirs up gene hunters, Nature 353:485-486 (October 10, 1991); L. Roberts, Genome patent fight erupts, Science 254:184-186 (October 11, 1991).
OCR for page 259
Global Dimensions of Intellectual Property Rights in Science and Technology quences of a variety of genes; the proteins coded for and produced by these genes, however, have not yet been isolated and their function is unknown without further research. Admittedly, the utility requirement for granting a patent is not satisfied. Nevertheless, a cautious lawyer would wonder exactly what additional steps would be necessary (by the sequencer or by someone else) before patentability becomes possible and, in the face of this legal uncertainty, might wonder whether it would be malpractice not to file an immediate application. The problems just identified are likely to change over the coming years. At some point, cloning and sequencing procedures will be regarded as obvious rather than novel, so that rights, if any, will be more likely to go to the discoverer of a pharmacologically active substance than to its cloner. The human genome project will move most human and some nonhuman research toward a pattern in which the researcher begins with a sequence and then looks for therapeutic activity (which may, in some cases, be suggested by the sequence's location on the gene). Because the sequence is known, it is obvious, and because the protein is directly coded, it may be obvious as well. Hence, the protein may be unpatentable (although its three-dimensional folding pattern may be difficult enough to reconstruct to provide a basis for nonobviousness). Sooner or later, the focus of biotechnology will move past natural proteins. Agricultural biotechnology has long been concentrating on transgenic plants and animals rather than therapeutic products; human biotechnology will probably look for new products not found in nature and for new ways to use them. The fact that living organisms are involved in biotechnology (and especially in agricultural biotechnology) poses a different group of questions. Some are ethical. Some are technical—for example, defining the appropriate scope of intellectual property protection. The obvious example of such a technical question is whether a patent should be regarded as reaching the progeny of a patented life form. By traditional law, the seller of a patented item exhausts his or her rights in the item, so that the buyer is entitled to use it as he or she sees fit. By definition, however, the reproduction of a patented article is an infringement of the patent. Clearly, this is a logical standoff between two traditional doctrines that can only be resolved by a policy analysis exploring the economic and innovation incentives of the two alternatives. That policy analysis might reach one result for a yeast, which has to be multiplied for many generations as part of a single fermentation application, and a different result for a chick bred for sale for meat production. Outline of Solutions, Thus Far The intellectual property community has worked extensively—but almost entirely autonomously—in an effort to face these issues. It was a U.S.
OCR for page 260
Global Dimensions of Intellectual Property Rights in Science and Technology Supreme Court decision, Diamond v. Chakrabarty, 477 U.S. 303 (1980), that launched the world's intellectual property system into the patenting of life forms, by authorizing the patenting of microorganisms. The U.S. Patent and Trademark Office (PTO) then, acting by internal decision, extended the Supreme Court's decision to plants and nonhuman animals.9 With the assistance of the Industrial Biotechnology Association, the PTO worked hard to build up its capabilities to deal with the backlog that emerged in the area.10 Congress has played only a passive and responsive role in this sequence. The PTO's extension of patent rights to plants has raised some concern in the agricultural community; its extension to animals raised concerns in both the agricultural and the animal rights communities. The result has been hearings, proposals for moratoriums, and suggested legislation to restrict these patents—but no actual legislation.11 Indeed the most relevant congressional action has been the passage of the 1983 Orphan Drug Act, which provides seven years of exclusivity, an equivalent of intellectual property protection, but enforced through the Food and Drug Administration's regulatory approval process.12 As of 1991, 9 of the 15 biotechnology-derived drugs on the market had such status, as did some 19 under development.13 At the international level, the process of adapting intellectual property law has been somewhat more thought out. During a large portion of the decade of the 1980s, there was an expert study organized by the World Intellectual Property Organization (WIPO)14 and, later on, a parallel study 9 Ex parte Hibberd, 227 USPQ 443 (PTO Bd. App & Int. 1985) (plants); Ex parte Allen, 2 USPQ 2d 1425 (PTO Bd. App & Int. 1987) (animals). Note that there had also been special coverage procedures for many agricultural plants under the Plant Patent Act of 1930 (35 U.S.C. sec. 161-164) for most species of asexual plants and the Plant Variety Protection Act of 1970 (7 U.S.C. sec. 3231 et seq.) for most species of sexually propagated plants. 10 Biotechnology in a Global Economy, supra at 210-214. 11 d. at 216. 12 Pub. L. No. 97-414, 96 Stat. 2049, Jan. 4, 1983. There is also a Drug Price Competition and Patent Term Restoration Act of 1984, which authorized the extension of patent terms for pharmaceuticals to compensate partially for time lost during the regulatory process, as part of an elaborate rebalancing of the relationships of generic and pioneer pharmaceutical firms, Pub. L. No. 98-417, 98 Stat. 1585, Sept 24, 1984. 13 Biotechnology in a Global Economy, supra at 92. 14 This was the Committee of Experts on Biotechnological Inventions and Industrial Property, whose First Session was November 5-9, 1984; Second Session, February 3-7, 1986; Third Session, June 29-July 3, 1987; and Fourth Session, October 24-28, 1988; Committee of Experts on Biotechnological Inventions and Industrial Property, various documents in series BioT/ CE. It is possible that there will be further meetings.
OCR for page 261
Global Dimensions of Intellectual Property Rights in Science and Technology organized by the Union for the Protection of New Varieties (UPOV).15 These are two international organizations that work with intellectual property; the former covers patents and a number of other forms of intellectual property, whereas the latter covers plant variety protection, a somewhat weakened form of intellectual property oriented toward the needs of the plant breeder.16 These working groups were very successful in defining the various hard issues that needed to be faced in applying intellectual property to biotechnology (e.g., questions of the application of a patent to progeny, questions of the interpretation of the experimental use exemption in the biotechnological context, and questions of the relative rights of two parties who make different kinds of improvements on the same organism). By almost any measure, although one might disagree with the conclusions they reached, their analysis identified most of the difficult problems earlier than anything in the domestic process in the United States. Probably the only hard issue that they missed (at least of the issues that have already appeared) is that posed by the current proposal to patent raw sequences arising from the genome project. The UPOV/WIPO work served as an intellectual (but not formal) basis for a European Community (EC) proposal for what amounts to sui generis protection for biotechnology.17 (The proposal would deal formally with patent law, but would establish a number of very specific rules.) This proposal, issued in 1988, responds in large part to the fact that the European Patent Convention prohibits the patenting of "plant or animal varieties or essentially biological processes for the production of plants and animals"18 and that there has been little uniformity in Europe in the national patent response to biotechnology. The EC proposal has met severe resistance from those who oppose extension of the patent system to living organisms on ethical or environmental grounds; there is not yet a fixed date for it to enter into force. Some of the critics argue that life should not be made subject to a property system. Others fear that the extension of intellectual property to living organisms will harm the economic position of small farmers or encourage the development of environmentally dangerous products. The European structure is not yet complete, and some of the technical issues that it faces will probably be solved in the United States through case 15 Committee of Experts on the Interface Between Patent Protection and Plant Breeders' Rights, Geneva, January 29-February 2, 1990, Report, WIPO/UPOV/CE/I/4, February 2, 1990. 16 This is the Plant Variety Protection Act of 1970, supra. 17 Proposal for a Council Directive on the Legal Protection of Biotechnological Inventions, COM(88) 496 final - SYN 159, Brussels, October 17, 1988. 18 Convention on the Grant of European Patents, Art. 53, signed October 5, 1973.
OCR for page 262
Global Dimensions of Intellectual Property Rights in Science and Technology law. Nevertheless, it seems clear that the European process has moved more effectively than the U.S. process to define the critical issues and subject them to debate. The key is probably the WIPO/UPOV pattern of repeated expert meetings producing substantive proposals that are subjected to external criticism. Computer Programs Computer programs (software) pose an extremely different set of issues. In the first instance, they appear to be (and are) text, and one is concerned about direct copy of this text. Yet, as Randall Davis of the MIT Artificial Intelligence Laboratory stated, "Programs are not only text; . . . they also behave."19 The type of protection to be provided that behavior is strongly debated, and given the extent to which computer and communications software (broadly conceived) is growing in market size and economic value compared to the corresponding hardware, the character of protection to be provided is extremely important economically. Special Issues The first special issue posed by software is its easy reproducibility. Discs can be copied cheaply and converted relatively easily from one computer language to another. This, of course, lends appeal to the copyright approach to software protection—and few would deny that copying or direct translation should in general be prohibited, because it fundamentally affects the incentives needed to develop software in the first place. However, after this point, the rights that should be given to a software producer become unclear. Should a user be entitled to decompile the program as part of a reverse engineering process? Should software be given protection against other software that uses the same code (in places) or program outline, when one considers that the outline and parts of the code may be defined by the problem and that independent "clean-room" development may lead to the same outline and in some cases the same code? Should there be protection for the program's appearance to the user (the "look and feel")? What about protection for novel algorithms? Also, should standards and interfaces associated with programs be protectable? These issues have arisen in an industrial structure that is, in general, much closer to that of engineering than to that of literature. Programs are 19 R. Davis, Intellectual property and software: The assumptions are broken, in World Intellectual Property Organization, WIPO Worldwide Symposium on the Intellectual Property Aspects of Artificial Intelligence, Stanford University (March 25-27, 1991).
OCR for page 263
Global Dimensions of Intellectual Property Rights in Science and Technology written by teams; they are constantly updated; consultants are used heavily; new programs build on previous programs. In a bizarre and wasteful response to the legal evolutions in the area, some are even written in clean rooms (i.e., written to specifications derived from an existing program but in a way designed to document that the writers of the new program had no knowledge of the detailed content of the first program). In an interesting new economic trend, programs are being written to common and evolving standards as part of "open systems." The complications of today's industry are probably simple compared to those of tomorrow's. Programs for parallel processing and for artificial intelligence may prove to be logically reducible to the kinds of programs being published today, but there may be practical differences that have intellectual property implications. For example, what property rights should be assigned to the expert information that is inserted into an artificial intelligence shell through a series of interviews? Then too, the embedding of software in products may create even more difficulty—consider, for example, the proposals to couple computer chips and biological sensors in ways that might allow "intelligent" management of the construction of individual biological polymers. Solutions, Thus Far In contrast to biotechnology, this area started out with an early expert advisory study by the National Commission on New Technological Uses of Copyrighted Works (CONTU), which issued its final report in 1978.20 This study explored the needs in the computer programming area and, without presenting any solid consideration of the implications, recommended use of the copyright pattern. Commissioner John Hersey's dissent warned of the point that programs act quite unlike traditional literary works and also foresaw the derivative works problem reflected in current look-and-feel debates.21 Several factors, however, appeared particularly to move CONTU. The copyright pattern obviously fit the need to avoid direct copying. It provided coverage that is relatively inexpensive for the right holder (in contrast to patent coverage), and it appeared much more appropriate than the patent and trade secret alternatives that the study considered.22 Although not explicit in the CONTU discussion, another important factor must 20 National Commission on New Technological Uses of Copyrighted Works, Final Report, Library of Congress (July 31, 1978). 21 Id. at 27-37. 22 Id. at 14-38. The relation between copyright protection and trade secret protection for software is not yet resolved; it is hard to understand how one can combine the two protections in light of the tradition that copyright applies to material that is published.
OCR for page 264
Global Dimensions of Intellectual Property Rights in Science and Technology have been that copyright permits use of the existing international copyright treaty network so that the right holder can obtain global coverage easily.23 CONTU went on to define the technical adaptations to copyright law needed to encompass computer programs, for example, to allow the ''copying" of a program into computer memory as an essential part of the use of the program. Congress followed the CONTU recommendations in amending the Copyright Act in 1980 to define "computer program" and to authorize such copies.24 Courts straightforwardly resolved a variety of similar, relatively technical issues (i.e., whether different principles should apply to software embedded in a microprocessor chip, on the grounds that such software is not published in the same sense as software sold on a disc.)25 However, the courts found it much more difficult to resolve the more fundamental questions noted above. In rather controversial decisions, the U.S. courts provided protection to structural features of a program26 and to its look and feel,27 but with at least one court taking a contrary opinion.28 From the viewpoint of many critics, who include a large portion of the academic specialists in the area,29 this extension of coverage provides the equivalent of patent protection (and for an irrelevantly long 75 years),30 without requiring that a patent-quality innovation be achieved and disclosed. Moreover, the logic of the cases involves significant stretching of the distinction between idea and expression. It is understandable to seek to interpret the relevant law to provide some form of protection to the intellectual logic and structure of the program—a creation that may involve significant expense and creativity. Yet, as long as one is within the copyright tradition, it is hard to say that this logic and structure are not in fact ideas and therefore unprotectable. The courts' efforts to describe these features as 23 Thus, the Berne Convention provides for automatic protection in all member countries, The Berne Convention for the Protection of Literary and Artistic Works, Art. 5 [Paris Act of July 24, 1971]. The Universal Copyright Convention, as revised at Paris, 1971, does have some formalities, but none requiring the types of filings that are typical of patent coverage. 24 Pub. L. 96-517, 94 Stat. 3015. 25 E.g., Apple Computer, Inc. v. Franklin Computer Corp., 714 F.2d 1240 (3d Cir. 1983), cert. dismissed, 464 U.S. 1033 (1984) (operating system); NEC Corp. v. Intel Corp., 10 USPQ 2d 1177 (N.D. Cal. 1989). 26 Whelan Associates, Inc. v. Jaslow Dental Laboratory, Inc., 797 F.2d 1222 (3d Cir. 1986), cert. denied, 479 U.S. 1031 (1987). 27 Lotus Development Corp. v. Paperback Software Int'l, 740 F. Supp. 37 (D. Mass. 1990); Broderbund Software, Inc. v. Unison World, Inc., 648 F. Supp. 1127 (N.D. Cal 1986). 28 Plains Cotton Cooperative Assn. v. Goodpasture Computer Service, Inc., 807 F.2d 1256 (5th Cir.), cert. denied, 484 U.S. 821 (1987). 29 See, e.g., P. Menell, An analysis of the scope of copyright protection for application programs, 41 Stan. L. Rev. 1045 (1989); P. Samuelson, CONTU revisited: The case against copyright protection for computer programs in machine-readable form, 1984 Duke L.J. 663. 30 17 U.S.C. sec. 302.
OCR for page 265
Global Dimensions of Intellectual Property Rights in Science and Technology expression have significantly distorted copyright's distinction between idea and expression. This difficulty is also shown by the question of decompilation of a program to understand its working. Were the program genuinely a literary work, such decompilation would be a normal form of study of the work. Were it genuinely a patented invention, such decompilation would be a normal step in reverse engineering or design improvement. Decompilation as a step toward design improvement is certainly a socially desirable activity (although there should be a reasonable way to sort out rights in the improved program). Yet the argument has been made (and accepted in Europe) that the reproduction of a program to study it is an infringement. Again, the problem is that the rights reasonably assigned to a software package do not match well with those defined by copyright law. Perhaps in response to the difficulties posed by copyright law in the software area, a number of firms have been seeking software patents. The area is governed by one of the most opaque series of Supreme Court cases that can be found in any body of law: Gottschalk v. Benson, 409 U.S. 63 (1972); Parker v. Flook, 437 U.S. 584 (1978); and Diamond v. Diehr, 450 U.S. 175 (1981). As a measure of the confusion, in November 1989, two different panels of the Court of Appeals for the Federal Circuit (CAFC) issued diametrically opposed opinions on the issue.31 Nevertheless, presumably in reflection of the fact that Diehr is relatively positive to patenting, the PTO has been granting a substantial number of software patents. The problem lies in whether software amounts to patentable subject matter under the current statute, which provides for coverage of a process, a machine, a manufacture, a composition of matter, or an improvement of one of these.32 The Court has long rejected the patenting of a "scientific truth or the mathematical expression of it."33 Not only do such discoveries or inventions not fit the statute; they also pose questions as to the potential overbreadth of a monopoly right. Although the explicit tests are much more complex, the basic outcome of the case law is that a program or algorithm is patentable provided it is adequately embodied in a machine or adequately restricted to a particular range of applications. The difficulty, of course, is that the relevant innovation resides precisely in the program or algorithm, not the way it is embodied or restricted to a specified range of applications.34 31 In re Grams, 12 USPQ 2d 1824 (CAFC 1989); In re Iwahashi, 12 USPQ 2d 1908 (CAFC 1989). For an effort to reconcile these opinions, see R. Laurie, The patentability of artificial intelligence under U.S. law, in WIPO Worldwide Symposium, supra at 121-150. 32 35 U.S.C. sec. 101. 33 Mackay Radio & Telegraph Co. v. Radio Corp., 306 U.S. 86 (1939). 34 See P. Samuelson, Benson revisited: The case against patent protection for algorithms and other computer program-related inventions, 39 Emory L. J. 1025 (1990).
OCR for page 266
Global Dimensions of Intellectual Property Rights in Science and Technology Other nations have gone through their own patterns, a number providing a U.S.-style mixture of patent and copyright law. A few have attempted sui generis proposals. Japan considered such an approach in the early 1980s and gave it up in the face of strong U.S. opposition.35 The EC, after much U.S. lobbying and international discussion of the details of the proposal, adopted its own sui generis approach in a 1991 directive.36 (As with the proposed EC biotechnology directive, this is technically an adaptation of existing law, copyright law in this case. The rules are so specific, however, that the directive can reasonably be called sui generis.) It is particularly interesting that each of these nation's laws have raised special issues that are both significant and different from those at issue in the United States. Thus, the Japanese law finally adopted as an amendment to the copyright law excludes "program languages, rules, or algorithms."37 The EC directive debate extensively considered whether interfaces should be protectable. Protection of interfaces would economically strengthen those with a large installed software network by enabling them to restrict the ability of others to market interoperating equipment. The directive basically declined to create such protection, but it was difficult to devise an appropriate exception to the directive's general prohibition on decompilation of a program. The result is that some fear that they cannot understand a program's interface without inadvertently violating the directive's general prohibition against decompilation. The legal difficulties in this area are an order of magnitude more severe than those in biotechnology. In biotechnology, one has a sense of new and difficult questions. In software, one has not only that sense, but also the more troublesome sense that the statutory models being used are fundamentally ill-adapted to the task and push the courts either to ignore important economic incentives or to twist the statutory language. The combination of troublesome questions and an ill-adapted statute suggests that CONTU was almost certainly wrong in its judgment that the copyright system should be used instead of a sui generis approach.38 35 D. Karjala, Protection of computer programs under Japanese copyright law,  4 E.I.P.R. 105. 36 Directive 91/250 on the legal protection of computer programs, May 14, 1991, OJ 1991 L 122/42. 37 Article 10(3) of Copyright Act, Law No. 48 (1970), as amended by Law No. 62, adopted June 7, 1985. 38 For similar judgments, see Office of Technology Assessment, Intellectual Property Rights in an Age of Electronics and Information 88-94 (April 1986); and Prepared Statement of P. Samuelson before the House Subcommittee on Courts, Intellectual Property, and the Administration of Justice, November 8, 1989. Note also that for each of the technical legal issues, there are relatively clear business winners and losers who have played an enormous role in the political debate. In general, up
OCR for page 273
Global Dimensions of Intellectual Property Rights in Science and Technology cess open to outside criticism and lobbying (particularly by the firm or interest group that can point out an important missed issue). These differences in timing and in opportunity for outside criticism are probably the important differences between the intellectually unsatisfying CONTU process and the more appealing WIPO expert pattern. CONTU was appointed in late July 1975 and first met in October of that year. Its first hearings were in May 1976, and its final report was filed in July 1978, three years after its creation.52 Certain of the important issues were raised in Commissioner Nimmer's concurrence and Commissioners Hersey's and Karpatkin's dissents;53 it is very plausible that the commission did not have adequate time to deal with the issues posed by these dissents. In contrast, the WIPO Committee of Experts, which published the results of its deliberations as it went along, met four times over the period November 1984 to October 1988 and may meet again.54 The critical transition, at which consideration went from analysis based heavily on the previous state of the law to discussion of the new hard issues, appears to have been in the staff work between the second session in February 1986 and the third session in July 1987.55 Thus, it took two years to identify these questions and to begin to deal with them; by that time, CONTU must have had to concentrate a significant portion of its attention on its final report.56 EVALUATION: THE RIGHTS-GRANTING PROCESS Doctrinal changes are only part of the system's response; the operational performance of the system in terms of providing mechanisms for obtaining and enforcing intellectual property rights is just as important. By way of introduction, one could hypothesize an intellectual property system's response to any technological perturbation. In the first instance, the system struggles with what rights to grant. If examination is needed as 52 CONTU, supra at 3-8. 53 Id. at 26-27, 27-37, and 37-38. 54 Committee of Experts, supra. 55 Compare Committee of Experts on Biotechnological Inventions and Industrial Property, Industrial Property Protection of Biotechnological Inventions; Report prepared by the International Bureau (prepared for the Second Session), WIPO BioT/CE/II/2 (November 5, 1985); with Committee of Experts on Biotechnological Inventions and Industrial Property, Industrial Property Protection of Biotechnological Inventions; Revised Report prepared by the International Bureau (prepared for the Third Session), WIPO BioT/CE/III/2 (April 8, 1987). The report of the Second Session reflects consideration of some of these issues. See Committee of Experts on Biotechnological Inventions and Industrial Property, Second Session, Report; Adopted by the Committee of Experts, WIPO BioT/CE/II/3 (February 7, 1986). 56 The current Advisory Committee on Patent Law Reform has only two years from creation to final report.
OCR for page 274
Global Dimensions of Intellectual Property Rights in Science and Technology it is for patents, the office hiring examiners will be in losing competition with industry for a small pool of capable specialists. Backlogs will build up and mistakes will be made. Ultimately, if the system can evolve, these problems will be solved. This process is unavoidable, no matter whether change occurs through statutory modification or through creation of a sui generis system; in either case there will be unanticipated issues to be faced. The same kind of evolution will mark interpretation of the rights. At first, courts will have doubts about the scope of the rights—there will be lots of decisions, many in conflict, while these issues are worked out. Depending on the structure of the industry, new patterns of patent licensing may evolve. Again, however, if the system is working well, these disputes will settle down after a while. As just suggested, the rights granting organization—e.g., the Patent and Trademark Office (PTO)—has to respond to new technologies by building new expertise. As an overall matter, the general increase in patent applications is taking place at a manageable rate. The number of patent applications filed increased from 106,295 in 1970 to 163,306 in 1989;57 this corresponds to an increase of 2.3 percent per year, slightly less than the 3.1 percent annual increase of real R&D expenditures over the same period.58 Also, if deviations in individual technology sectors are ignored, the PTO appears to have been quite successful in managing this case load. The backlog fell from 1.8 times the annual application rate in 1970 to 1.4 times that rate in 1989.59 There is, however, a difficulty in matching the personnel pattern to the needs of new areas of technology. As a new type of technology arises, the PTO must develop new expertise. As noted above, this requires education of existing employees or recruiting and holding of new employees in a market certain to be favorable to the employees. Data are available for the PTO backlog in the biotechnology sector; it is somewhat discouraging in that the average delay in the sector was three years in 1989—many years after the intellectual explosion in the area became obvious.60 In some subsectors, the delay was nearly four years.61 More recently, the biotechnology backlog has been brought down. The process of adapting to new technologies leaves behind not only a 57 Commissioner of Patents and Trademarks, Annual Report; Fiscal Year '89 (January 1990), Table 6. 58 R&D estimates calculated from U.S. Department of Commerce, Statistical Abstract of the United States 1988 (December 1987), Table 950 (data for 1970-1987). 59 Calculated from Commissioner of Patents and Trademarks, supra, Tables 6 and 8. 60 Biotechnology in a Global Economy, supra at 212. 61 Id.
OCR for page 275
Global Dimensions of Intellectual Property Rights in Science and Technology delay (which is especially significant in biotechnology where a number of firms have been pursuing the same product)62 but also errors. Almost every expert has a story of patents in new areas that were almost certainly granted only because the examiner was not really familiar with the state of the art in the technology or because there had not yet evolved a data base within which a search could be conducted.63 These patents can severely complicate the economic development of the relevant field, although it is possible for a sensitive court to choose not to enforce them. Nevertheless, the overall evaluation of the rights-granting process has to be relatively favorable. EVALUATION: THE RIGHTS-ENFORCING PROCESS Intellectual property rights are meaningless unless enforced (and for software copyright, where the grant of rights is essentially automatic, enforcement is the only context in which litigation comes to the surface). Enforcement is, in the first instance, a litigation issue, but litigation is so expensive that its economics shapes the effective scope of intellectual property rights. A patent that its holder cannot afford to defend is worthless; likewise, a patent claim can be significantly stretched against a firm unable to afford defensive litigation. Equally important, intellectual property licenses—whose pattern differs radically from industry to industry—dramatically shape the real-world impact of these rights. Judicial Enforcement Patent litigation is extremely expensive; rumor suggests a cost of $0.5 million per claim litigated per side. There is little reason to anticipate a significantly different number for copyright, save perhaps for the possibility that some of the disputes are less fact intensive and more law intensive. (The costs of fact discovery and proof are especially high.) This is a nearly absolute bar to use of the system by small firms: where they must use the system, as in the biotechnology industry whose pharmaceutical products are 62 There are sectors in which patentees would prefer delays in patent issuance because the market is growing faster than the interest rate so that a later monopoly is more valuable than an earlier one. This might even be true of biotechnology—except for the facts that development costs are so large, that development takes as long as the patent application processing time, and that these ongoing expenditures will be wasted if a competitor gains the patent rights. 63 ''Practically once a month, the nation's computer networks are abuzz with news of another patent issued on a fundamental concept that is widely used." S. Garfinkel, R. Stallman, and M. Kapor, Why patents are bad for software, 7 Issues in Science and Technology 50-55, 51 (1991).
OCR for page 276
Global Dimensions of Intellectual Property Rights in Science and Technology easily imitated, the resulting costs are likely to drain away research funds. Some biotechnology firms are said to be spending more on litigation than on research. The economics of litigation is thus likely to favor large firms at the expense of small ones. Large firms are more likely to be able to threaten litigation and to defend against litigation. There have been at least some cases of "strategic litigation" in which a large firm uses the threat of litigation costs to squash a start-up.64 The reality of such threats has evolved with recent strengthening of the intellectual property system. In the 1960s and 1970s, patents were generally considered nearly irrelevant, because they were so often found invalid.65 Moreover, even if they were found valid, there were a number of patent misuse doctrines and antitrust doctrines that restricted the effective scope of patents. During the 1980s, nearly all these factors changed. The Court of Appeals for the Federal Circuit was created66 which not only brought the body of patent law into uniformity but also substantively changed it. Congress, the courts, and the Reagan administration radically weakened the various defenses of an accused infringer and changed antitrust/patent perceptions.67 There are signs of a change back, however, for example, in Lasercomb America, Inc. v. Reynolds, 911 F.2d 970 (4th Cir. 1990), in which the circuit court created a new copyright misuse concept analogous to the old patent misuse doctrines. A separate group of enforcement issues arises at the international level and will probably be felt more strongly in the biotechnology sector than in other sectors discussed in this chapter. Many nations have hesitated to extend as wide a scope to intellectual property protection of biotechnology as does the United States, so that doctrines differ significantly from nation 64 A. Silverman, Symposium report: Intellectual property law and the venture capital process, 5 High Technology L. Journal 157 (1990). 65 Before the Court of Appeals for the Federal Circuit was created, appellate courts upheld only 30 to 40 percent of the patents found valid by trial courts, M. Adelman, The new world of patents created by the Court of appeals for the Federal Circuit, 20 U. Mich. J. L. Ref. 979 (1987). The new court upheld 89 percent of patents in such circumstances, D. Dunner, Special Comm'n. on CAFC, 1988 A.B.A. Sec. Pat., Trademark and Copyright L. 314, 325. Continuation of this trend would make it less likely that the courts would correct for patent office errors in new technological areas. 66 Federal Courts Improvement Act of 1982, Pub. L. No. 97-164, sec. 127(a), 96 Stat. 37 (codified at 28 U.S.C. sec. 1295). 67 For example, the CAFC rejected the idea that it was an antitrust violation for a patentee to claim infringement while knowing that there was not infringement. Loctite Corp. v. Ultraseal, Ltd., 781 F.2d 861 (Fed. Cir. 1985). The Supreme Court and Congress strengthened the right of a patentee not to grant licenses, Dawson Chemical v. Rohm and Haas Co., 448 U.S. 176 (1980); Pub. L. No. 100-703, sec. 201, 102 Stat. 4674, 4676 (1988) (codified at 35 U.S.C. sec. 271(d)).
OCR for page 277
Global Dimensions of Intellectual Property Rights in Science and Technology to nation. Because of these differences and because the area offers the opportunity for substantial trade in intermediate and finished products, border restrictions are likely to be particularly important in this area. The U.S. border restrictions are generally enforced through "section 337," which is administered by the International Trade Commission (ITC).68 This procedure provides a U.S. intellectual property holder with extremely strong rights to bar the import not only of infringing products but also of the direct and indirect products of processes that would infringe if practiced within the United States. The biotechnology industry would like to extend these rights further through making essentially automatic the granting of process claims in a number of situations in which product claims are already granted.69 In a world that has not achieved patent law unification and an industry that is and must be internationally based, this kind of trade barrier will also be a barrier to progress by decreasing the extent to which product intermediates made legitimately in one nation can be exported to another. Significant extension of the section 337 process will lead to global suboptimization and enormous headaches for managers who are attempting to organize on an international level.70 Nevertheless, current political trends in the U.S. Con- 68 This is section 337 of the Tariff Act of 1930, ch. 479, 46 Stat. 590, 703 (codified as amended at 19 U.S.C. sec. 1337, amended by Omnibus Trade and Competitiveness Act of 1988, Pub. L. No. 100-418, sec. 1342, 102 Stat. 1107, 1212 (1988). 69 E.g., S 654, the proposed Biotechnology Patent Protection Act of 1991. See Amended biotech process patent bill is cleared by Senate subcommittee, 42 BNA Patent, Trademark & Copyright J. 313 (August 1, 1991). The intended result would have been to change the outcome of the ITC phase of the Amgen-Chugai litigation, In re Certain Recombinant Erythropoietin, 10 USPQ 2d 1906 (ITC 1989). A Japanese firm held the product patent in the United States, while its U.S. competitor held a U.S. product patent on cells containing a gene inserted to manufacture the product. Because the foreign firm was able legally to use its cells with the inserted gene to produce the product abroad and sell it into the United States, the U.S. firm had little leverage to negotiate with its competitor. Had a process claim been available to the U.S. firm for its cells containing the gene, it would have been able to exclude the product under section 337 and would then have had a better bargaining position. The specific dispute was effectively resolved when, in parallel litigation, the Japanese firm's patent was unexpectedly found invalid, Amgen, Inc. v. Chugai Pharmaceutical Co., 927 F.2d 1200 (CAFC 1991). 70 See J. Barton, Testimony prepared for presentation before the Subcommittee on Intellectual Property and Judicial Administration of the House Judiciary Committee, May 16, 1991. It should also be noted that the existing section 337 provisions have been found to be in violation of the General Agreement on Tariffs and Trade (GATT), Re the DuPont de Nemours/AKZO Dispute: European Economic Community v. United States of America, reprinted at  1 C.M.L.R. 147, a problem that the administration suggested it would resolve as part of the implementing legislation for the Uruguay Round, [President Bush's] Memorandum for the United States Trade Representative; Subject Enforcement of Section 337 of the Tariff Act of 1930 (November 7, 1989). See generally, Barton, Section 337 and the international trading system, in Technology Trade and World Competition; Protecting Intellectual Property with
OCR for page 278
Global Dimensions of Intellectual Property Rights in Science and Technology gress are probably toward such strengthening, presumably based on the judgment that it will increase the rents accruing to U.S. firms. Given that the U.S. share of all intellectual property is declining as non-U.S. research increases, this is probably a shortsighted position. Thus, when one looks at the law as applied by courts, one is forced to retreat somewhat from the sense that the patent system is adapting relatively well in the biotechnology sector. The doctrine may be evolving rapidly enough to reduce the overall number of cases, but the expenses of these cases—and the still unsolved international issues—suggest a significant burden for the industry. It is at least possible that this burden falls disproportionately on the newer, smaller, more innovative firms. Informal Approaches Business has its ways of adapting to the difficulties of applying intellectual property systems. For example, the American Society for Composers, Authors, and Publishers is a privately created network that resolves the practical problems of collecting and distributing royalties in the musical area where there are so many individual performances of copyrighted works. Such systems are as much part of the effective body of law as are the formal statutes.71 On the basis of anecdotal evidence, the "normal" pattern of intellectual property enforcement is an industry-wide cross-license arrangement in which any payments by one firm to another are based on a very rough comparison of the relative value of the intellectual property contributed by each firm.72 The firm with more patents collects from the firm with fewer patents, with little attention to the value of specific claims. The system thus rewards innovation while avoiding expensive litigation (which is saved for the cases in which a firm makes a serious challenge of the balance). This pattern was, for example, typical of the electronics industry prior to Texas Instrument's challenge of Japanese firms in 1986,73 a challenge that triggered a round of litigation that may now be ending. The cross-license is a natural response Trade Sanctions 1 (1990). The GATT panel's concerns go to certain procedural differences between ITC and District Court enforcement of patent rights against import; the problem discussed in the text is not yet on the reform agenda (and was not posed to the GATT panel), but is probably much more important than the issues identified by the panel. 71 See Intellectual Property Rights in an Age of Electronics and Information, supra at 269-71. 72 Consider, for example, the arrangements in the auto and the aircraft industries considered in U.S. v. Automobile Manufacturers Assn., Inc., 1969 Trade Cases par. 72,907 (DC Cal. 1969), modified, 1982-83 Trade Cases par. 65,088; and U.S. v. Manufacturers Aircraft Assn.. Inc., 1976-1 Trade Cases par. 60,810 (DCNY 1975). 73 See Texas Instruments, Inc. v. United States International Trade Commission, 871 F.2d 1054 (CAFC 1989).
OCR for page 279
Global Dimensions of Intellectual Property Rights in Science and Technology to the limitations of litigation and a very effective response when each firm in the industry must use a variety of previously claimed inventions to produce a product. There are two major exceptions to this "normal" pattern. One arises in those industries in which individual patents are the basis of specific products and each product has substantial market power. The obvious examples are chemistry and pharmaceuticals. Different chemicals and pharmaceuticals substitute only exceptionally for one another. Each patent, in effect, represents a market monopoly that can be practiced independently of competitors. The precise claims are taken quite seriously because the patent monopoly defines the returns from the product very precisely, serves as the basic competitive protection, and faces few countervailing claims that could be the basis for a cross-license. There are some rare parallel cases, such as fundamentally new technologies in which competitors can be excluded for a while (e.g., the early days of instant photography). The medical biotechnology industry falls precisely within this exception.74 It is an industry that anticipates producing a relatively small number of products, each of which requires enormous front-end research and regulatory investment, and each of which is likely to have a substantial product lifetime and to be readily imitated. In short, it is an industry exactly adapted to the patent system, because a period of monopoly is nearly essential as a mechanism of covering the front-end fixed costs. What is at stake in a patent suit is the possibility of access to this period of monopoly returns and the precise (and relatively easily evaluated) scope of these returns. There is no wonder that firms expend great efforts to protect their patent position. It is likely that the current litigation pattern will continue in this industry. The other exception to the typical industry-wide cross-license is the occasional "flare-up" in an industry normally governed by a wide cross-license. This may typically be a response to an outsider who threatens to upset the competitive balance in an industry; this is certainly part of the explanation for Texas Instrument's use of patent law to protect its market against Japanese semiconductor competitors. (Texas Instruments has gone on, but without clearly improving its competitive position, to use patent royalties as a major source of income.) Given the international cross-flows of technology in the semiconductor sector, it is hard to imagine that the patent battle will not soon be settled. 74 In contrast, agricultural biotechnology may well turn out to involve proprietary genes with a variety of applications marketed through licensees with expertise in particular seed markets; the licensing structure may not be a traditional cross-license pattern, but it is likely to be relatively stable.
OCR for page 280
Global Dimensions of Intellectual Property Rights in Science and Technology One can anticipate, however, that the software and network areas will be within this exception for a long time. The economic structure of these areas is highly unsettled. Communications firms, computer firms, and software firms are all vying for additional influence and control, and are seeking to assert whatever intellectual property rights they can define or have defined as part of the jockeying. Along with the doctrinal confusion described above, this is undoubtedly a reason why litigation rates are so high in this industry. A cross-license may be the almost certain ultimate direction of software patents on algorithms and computational procedures; different claims are likely to be interlocked, and the task for a new competitor may prove to be to produce some useful algorithms of its own and then to enter the cross-license. On the broader software and interface issues, however, in the absence of a relatively stable industry structure and pecking order, continued intellectual property litigation is very likely. In summary, in all three of the new areas, continued and expensive intellectual property litigation is likely. Given that any new technology upsets existing market structures, the pattern is likely to prevail for other technologies as well. This implies continued expenditures on litigation rather than on research. Perhaps more important, it will favor firms with strong economic positions at the expense of their challengers. If their most innovative challengers are, as may be suspected, small firms, the costs of litigation and the imperfections of the litigation system will cut against new technologies.75 A full response to intellectual property litigation costs would go far beyond the scope of this chapter, and the costs may well derive more from the general character of U.S. litigation than from the specific character of intellectual property. Nevertheless, three points can be suggested: To the extent that legal uncertainty drives litigation, extensive use of expert studies and sui generis forms of coverage may help reduce costs. 75 Economic evidence on whether small firms are more innovative than large firms is inconclusive. Gellman Research Associates (1975) and Acs and Audretch (1990) found that small firms produce more innovations per unit sales and per employee, respectively, than do large firms. Freeman (1982), however, found that the share of industrial innovations in Great Britain contributed by small firms was smaller than their share of production and employment. Despite inconclusive quantitative evidence, there are many historical and recent examples of small firms playing an important role in the establishment of new branches of industry and the rejuvenation of old ones. Gellman Research Associates, Indicators of International Trends in Technological Innovation, report prepared for the National Science Foundation (1975); Z. Acs and D. Audretch, Innovation and Small Firms, Cambridge, Mass.: MIT Press (1990); C. Freeman, The Economics of Industrial Innovation, 2nd ed., Cambridge, Mass.: MIT Press (1982).
OCR for page 281
Global Dimensions of Intellectual Property Rights in Science and Technology Litigation can probably be discouraged somewhat if rights are defined less broadly.76 Again, if it is, as just suggested, desirable to favor smaller firms, the antitrust and intellectual property misuse doctrines may be helpful to assist them in their defense against litigation and thus to deter litigation against them. OVERALL IMPLICATIONS Ability of the System to Adapt to Increasing Innovation Rates Nothing developed in this analysis provides any reason to believe that increased filing rates have posed fundamental problems for the system. To the contrary, the PTO has, overall, successfully kept up with the rate of innovation. There are obvious qualifications. The PTO has not been able to hire new types of analytic capabilities rapidly enough; this has led to errors and delays in specific areas, and is likely to do so again and to cut against new technologies. In the biotechnology sector, in particular, the economic loss due to delay has been severe because firms are undertaking competing research projects and do not know which firm wins until a patent issues (and possibly until the patent is litigated). This is almost certainly a strong argument for switching to the global first-to-file system, with the applications made public a reasonable time after filing.77 Specific Doctrinal Implications of the Three Examples The biotechnology case shows the system to be relatively successful in dealing with change. The critical pervasive problem is defining the point at which a gene becomes patentable; this is an issue that should ideally be 76 Note that moderation is compelled by traditional economic analysis of intellectual property incentives. This analysis balances the incentive to innovation created by the monopoly right defined by intellectual property against the economic costs derived from the artificial prices created by the monopoly rent. An additional basis for moderation is suggested by the possibilities that smaller firms are better innovators and also victims of litigation costs. Put more broadly, under certain circumstances a first intellectual property monopoly right may in fact be exercised in a way that decreases incentives to subsequent innovation. For an argument in a similar direction, see Commissioner Hersey's dissent in CONTU, especially at pp. 35-36. 77 With a first-to-file system, one can publish patent applications without fear of complicating priority disputes and thus speed the flow of scientific information. The typical European pattern, for example, is to require filing before publication, but then to publish the applications 18 months after filing (European Patent Convention, Articles 54 and 93). The publication conveys a form of interim protection (Article 67).
OCR for page 282
Global Dimensions of Intellectual Property Rights in Science and Technology analyzed by a thoughtful study group before decisions are made on the patenting of genome sequences. The software sequence, in contrast, shows up very badly. The basic CONTU decision to rely primarily on a modified copyright concept appears to have been wrong, and complementary efforts to apply patent law have been, at best, doctrinally confusing. This sector cries out for a new sui generis approach. For the data bank context, it is too early to tell confidently, but if the Supreme Court's decision in Feist turns out to dominate the field, one can anticipate that there will be serious difficulties. This area also cries out for an early specific analysis. Broader Implications Technology is unpredictable; hence one must be careful about relying too heavily on the specific examples just discussed. Nevertheless, based on these examples (and on the other examples discussed in the analysis) one can suggest the following general conclusions: Sui generis approaches are far more likely to be successful than one might have expected and should be utilized far more often. The European biotechnology and software directives and the plant variety protection and chip mask work statutes are all relatively encouraging. It is essential to have routinized law reform processes to help deal with new technologies. The experience of the WIPO/UPOV panels is far more positive than that of CONTU. Although there may be a variety of reasons for this difference, one is that the WIPO/UPOV system was under much less time pressure, which enabled the expert panel and the staff to publish interim positions that could be broadly criticized and commented on. This is almost certainly better than the U.S. commission-style approach. It strongly suggests standing study groups, an approach that could be easily integrated into U.S. law. It might also suggest a pattern in which an expert panel is delegated the power actually to lay down and put in effect the rules for a sui generis system, leaving Congress the initiative to change the proposed rules if it wishes.78 The law reform process must be international. One of the reasons copyright has been pushed for software is that international coverage was relatively easy to achieve. In biotechnology, intellectual property protection is a possible trade barrier. The computer data bank issue is fundamen- 78 See R. Stern, The bundle of rights suited to new technology, 47 U. of Pittsburgh L. Rev. 1229 (1986).
OCR for page 283
Global Dimensions of Intellectual Property Rights in Science and Technology tally international. In short, future sui generis rights (and many proposals for interpretation of existing conventions) should be negotiated on an international basis,79 and standing reform and study groups should be international. It is crucial to take into account the role of small firms in innovation. Such firms may be prime sources of innovation in areas of new technology and there may be a serious risk that intellectual property rights can be used to stifle them. This means that such firms must be represented on study groups; it also means that intellectual property rights must be defined with consideration for their real-world impact on industrial structure. 79 Note that in the absence of international agreement, one can use statutory reciprocity in the pattern of the chip mask work arrangement. Even so, it would be best to coordinate such efforts with as many other nations as possible. Note also that the politicization of the international intellectual property system may make internationalization difficult. Yet, this politicization is primarily along a North-South axis, whereas the key negotiations in these new technology areas will generally be among the developed nations or with developing nations who have an interest in being included in a special regime.
Representative terms from entire chapter: