The promise and perils of strategic publication to create prior art: a response to Professor Parchomovsky.

AuthorEisenberg, Rebecca S.
PositionResponse to Gideon Parchomovsky, Michigan Law Review, vol. 98, p. 926, 2000

In a provocative recent article in the Michigan Law Review,(1) Professor Gideon Parchomovsky observes that a firm racing with a competitor to make a patentable invention might find it strategically advantageous to publish interim research results rather than risk losing a patent race. This strategy exploits legal rules limiting patent protection to technological advances that are new and "nonobvious" in light of the "prior art" or preexisting knowledge in the field.(2) By publishing research results, a firm adds to the prior art(3) and thereby limits what may be patented in the future. Parchomovsky posits that, before it is able to claim a patentable invention of its own, a firm might have sufficient information to publish results that would raise the threshold for patentability enough to preclude the issuance of a patent to its rival. A firm that believes a rival is likely to complete an invention first might therefore try to defeat the rival's patent prospects through publication, leaving both firms free to compete in the market for the unpatented invention. Otherwise, the firm that wins the race could use the patent to exclude the lagging firm from the market entirely. This "spoiler" strategy, according to Parchomovsky, "explains the otherwise peculiar practice of commercial firms that routinely publish research results in scientific and technological journals."(4)

One might question whether the widespread practice of publication can be taken as evidence that firms are pursuing a spoiler strategy of creating patent-defeating prior art in lieu of pursuing their own patent rights. Commercial firms might have other motivations for publication, such as gaining scientific recognition for their researchers and establishing credibility for their research results. These motivations can be particularly powerful in research-intensive industries. Access to top scientific talent may confer a competitive advantage on firms that permit their scientists to publish research results. Moreover, publication of promising results might make it easier for a firm to raise money for further research from investors or collaborators. Publication (typically after filing a patent application) might more typically mark a declaration of triumph in a race than an effort to contain the consequences of defeat.

But Parchomovsky does not attempt an empirical demonstration that firms, in fact, do publish research results to defeat the patent claims of their rivals. Rather, his mission is to illustrate how current doctrine might make this strategic maneuver attractive, and to suggest that patent policy should take this possibility into account. He attempts to illustrate the advantages of a preemptive publication strategy through a model that examines the possible payoffs confronting two pharmaceutical firms competing to invent a new drug. The plausibility of his analysis thus depends on the verisimilitude of the assumptions underlying the model.

Parchomovsky's model is vulnerable to two different lines of attack. First, it oversimplifies legal rules that, upon closer consideration, severely limit the circumstances in which a preemptive publication strategy would work. Second, the robustness of the model is further limited by its use of an optimistic hypothetical payoff matrix, in which preemptive publication leads to the sharing of only slightly reduced profits in a patent-free duopoly. In patent-dependent fields like drug discovery -- the ostensible setting for Parchomovsky's model -- firms would likely expect patent-free competition to be far more destructive of potential profits, minimizing its attractiveness. Preemptive publication might make sense for firms that would welcome lower competitive prices, but these are more likely to be prospective users of the invention than prospective sellers. Evidence of preemptive publication strategies at work in the field of genomics is consistent with this prediction.

Parchomovsky's analysis must contend with features of patent law that limit the circumstances in which a preemptive publication strategy would work. In order to understand these limitations, it is necessary to delve a bit more deeply into the role of "prior art" in patent doctrine. In evaluating an invention for patentability, U.S. law inquires into the state of knowledge in the field as of two different points in time: the invention date,(5) and one year prior to the application filing date.(6) An invention may not be patented in the U.S. if it was disclosed in, or made obvious by, a publication prior to the applicant's invention date.(7) The invention date, however, may be earlier than the patent filing date. Faced with a publication that might bar the issuance of a patent, a U.S. patent applicant can sometimes get around the problem by proving an earlier invention date through a practice known as "swearing behind the reference."(8) Even if, as Parchomovsky posits, the patent applicant had not yet completed the claimed invention as of the rival's publication date, the applicant may still avoid use of the reference as prior art by proving that, as of that date, the applicant had completed as much of the invention as the reference discloses.(9) This rule generally prevents a lagging firm from defeating the patent claims of a leading firm by publishing interim research results that merely duplicate what the leading firm has already done.

There is one important limitation on the ability of patent applicants to avoid prior art references by proving earlier invention dates: an applicant may only swear behind a reference that is dated within one year of the application filing date.(10) An applicant whose invention was disclosed in, or made obvious by, a publication dated more than one year before the filing date may not avoid the use of the reference as prior art, regardless of the invention date.(11)

These rules seriously limit the attractiveness of preemptive publication for the lagging firm in a race. If both firms are pursuing identical research paths, preemptive publication can only work if deployed at least a year prior to the application filing date of the leading firm. Otherwise, the leading firm could avoid the publication as prior art by proving that it had already progressed as far as the lagging firm as of the publication date. Put differently, the firm that is winning the race has one full year after the rival's publication date to file a patent application before forfeiting the ability to swear behind the publication. If it takes more than a year for the firm that is ahead to cross the finish line after seeing the publication, that fact alone might call into question whether the publication made the invention obvious.

What if the two firms are not pursuing identical research paths, but rather, are taking different approaches toward the same inventive goal? In that case, the patent applicant may be unable to show that he had already completed as much of the invention as the publication discloses, limiting the risk that the applicant could avoid the effect of the reference by swearing behind it. But other factors might make publication more risky in this setting. If the information disclosed in the publication is news to the leading firm, disclosure of that information might accelerate its research efforts, perhaps even hastening the day when it is ready to file a patent application. At the very least, the patent applicant, having seen the publication, will enjoy a strategic advantage in drafting claim language that focuses on aspects of its own work that are distinguishable from the disclosures of the reference, and therefore still qualify for patent protection.

Publication of inadequate research results by a lagging rival might actually help the patent applicant to establish that the invention was nonobvious. The patent statute provides that an invention may not be patented if it would have been obviousness at the time it was made to a hypothetical practitioner of ordinary skill in the same field in light of the prior art.(12) The strictness of this standard by its terms varies from one field to the next and changes over time, depending both on the scope and content of the prior art and on the prevailing level of skill among people working in the field. Rather than relying on hindsight reconstructions, courts routinely turn to evidence of actual events surrounding the making of an invention to shed light on determinations of nonobviousness, including evidence that the patent applicant succeeded where others had failed.(13) By publishing research results that fall short of achieving a successful invention, a lagging rival thus risks helping the leading rival to prove patentability. Although the publication adds to the prior art, its limitations...

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT