Patent Swords and Shields
Rebecca S. Eisenberg*
The recent rejection by the Court of Appeals for the Federal Circuit of an
"experimental use
defense" to a patent infringement lawsuit against Duke University (1) comes
as no surprise to
those who follow U.S. patent law, but it is an alarming wake-up call to the
academic community.
Ever since Congress affirmed the right of universities to patent the results
of
government-sponsored research in the Bayh-Dole Act of 1980 (2), academic
researchers
and university administrators have blithely assumed that they may enforce
patents on their
own inventions while avoiding liability for using the patented inventions of
others. This rests on
a belief, widespread in the scientific community, that patent infringement
requires use for
commercial purposes, and does not arise in "pure" academic research.
This view finds more doctrinal support outside the United States than within
it. The European
Commission's recently proposed Council Regulation on the Community Patent
excludes from
the effects of a European community patent "acts done privately and for
non-commercial
purposes" and "acts done for experimental purposes relating to the
subject-matter of the
patented invention." (3). The national patent laws of many E.U. member
states contain similar
provisions, as does Japanese law (4).
CREDIT: JOE SUTLIFF
The U.S. Patent Act has no statutory exemptions for noncommercial or
research uses of an invention, apart from a specific
provision on clinical testing of generic drugs (5). The academic community
can nonetheless claim a whiff of authority for its
vaunted experimental use defense in a line of cases going back to the 1813
opinion of Justice Story in Whittemore v. Cutter (6).
In approving a jury instruction that defined patent infringement as "the
making of a machine fit for use, and with a design to use it
for profit," Justice Story speculated that "it could never have been the
intention of the legislature to punish a man, who
constructed a [patented] machine merely for philosophical experiments, or
for the purpose of ascertaining the sufficiency of the
machine to produce its described effects."
As a formal matter, gratuitous statements of this sort (called dicta in
legalese) are not binding authority in subsequent cases. But
Justice Story's experimental use defense has enjoyed extraordinary
longevity. Subsequent courts have acknowledged the
availability of the defense, but have almost never applied it to the cases
before them. This is hardly surprising. Few patent
holders will notice or care if someone makes a patented invention "merely
for philosophical experiments." Most judicial decisions
considering the scope of the experimental use defense have resolved disputes
between commercial competitors. Within this
universe of cases, the defense is rarely sustained, leaving its boundaries
unclear.
Justice Story's early-19th-century picture of a gentleman scientist driven
by idle curiosity predates the rise of the modern
research university. Later courts seem to have had something closer to
contemporary academic research in mind. Yet only one
case, Ruth v. Stearns-Roger Manufacturing Co. (8), generated a published
opinion holding that use of a patented invention in a
university laboratory qualifies for the defense. The defendant in that case
was not a university but a commercial firm that sold
replacement parts for patented machines; consequently, the firm was
vicariously liable for infringement by its customers. In
calculating damages, the court followed a recommendation of a magistrate
that excluded sales of replacement parts to the
Colorado School of Mines on the grounds that these parts "were for use in
laboratory machines used for experimental purposes,
and consequently did not contribute to an infringing use."(9). Fifty years
later, in 1985, a commentator who was skeptical of the
experimental use defense nonetheless approved the Ruth decision, writing
that "few would deny the experimental use exception
for research on patented technology performed at a university in furtherance
of its educational function." (10).
But 1985 seems like ancient history when it comes to universities and
patents. Today, universities have become players in the
patent system in a way that could hardly have been imagined before the
Bayh-Dole Act. Universities owned 1.1% of U.S.
corporate-owned patents issued between 1969 and 1986; by 1999 that number
had risen to 4.8% (11). As their patent portfolios
have grown, universities have become more aggressive about enforcing their
patents in court. The University of California's
$200 million settlement with Genentech (12) and the University of
Minnesota's $300 million settlement with Glaxo-Wellcome (13)
have emboldened others to follow with their own lawsuits, including Baylor
College of Medicine, Cornell University, Columbia
University, University of Rochester, and the Massachusetts Institute of
Technology (14). Columbia University has further
emulated commercial patent holders by petitioning Congress to extend the
term of its patents (15). As universities shed their
noncommercial innocence to reach deeper into the pockets of commercial
firms, one might expect to see firms strike back with
their own infringement claims, urging courts to reject the experimental use
defense as a nostalgic fantasy.
But that is not what happened. Instead, the experimental use defense was
taken out in an inside job, a casualty of an
intraacademy squabble over control of resources. John Madey was recruited to
join the physics department at Duke as a
tenured professor in the late 1980s. In a previous position at Stanford,
Madey had designed equipment for his research on free
electron lasers. Stanford had no interest in patenting this technology
itself and yielded title to Madey, as the Bayh-Dole Act
permits (16). Duke constructed new laboratory space to house the equipment,
and the Office of Naval Research (ONR) awarded
Duke a grant with Madey as Principal Investigator. After some years the
relationship soured, Duke and ONR agreed to replace
Madey on the grant, and Madey resigned.
Madey sued Duke on a variety of legal theories, including patent
infringement. The trial court granted summary judgment for Duke
on the patent infringement claims, finding that Duke's use of the patented
equipment was protected by the experimental use
doctrine. Madey appealed to the Court of Appeals for the Federal Circuit
(17).
The Federal Circuit has been signaling its discomfort with the experimental
use defense for almost 20 years. That court takes
very seriously its mandate from Congress to bring about greater uniformity
and predictability in patent law decisions. The Federal
Circuit has also consistently taken a restrictive view of judge-made, common
law rules not incorporated into the language of the
statute.
In the 1984 case of Roche v. Bolar (18), the Federal Circuit rejected the
argument of a generic drug manufacturer that the
experimental use defense applied to its use of a patented drug to conduct
clinical trials, even as Congress was in the process of
enacting a statutory defense for exactly this use as part of the
Hatch-Waxman Act (19). The court observed that the use was
"no dilettante affair such as Justice Story envisioned" but had a commercial
purpose. Sixteen years later, the Federal Circuit
relied on Roche v. Bolar in rejecting the defense in Embrex v. Service
Engineering (20), noting that the defendant had "performed
the tests expressly for commercial purposes." One member of the panel would
have gone further to hold that "the Patent Act
leaves no room for any de minimis or experimental use excuses for
infringement." (21).
By the time the Madey case arrived before the Federal Circuit there were
warning signs that the experimental use defense was
not as robust as universities might hope, yet prior decisions had always
seemed to leave the defense intact for pure scientific
inquiry with no commercial motivation. The principal cause for concern for
universities under the language of prior opinions thus
appeared to be the growing difficulty in distinguishing commercially
motivated research from pure academic research in the
Bayh-Dole era. The most surprising move that the Federal Circuit made in
Madey was to sidestep this issue entirely.
Sweeping aside almost 200 years of opinions that had stressed the commercial
character of defendants' activities in refusing to
apply the defense (including its own prior opinions in Roche and Embrex),
the Federal Circuit declared the noncommercial
character of the research in Madey irrelevant to its analysis of the case.
What matters, in the Federal Circuit's revisionist
account of the defense, is whether the research "is in keeping with the
alleged infringer's legitimate business, regardless of
commercial implications." In the case of a major research university,
noncommercial research projects "unmistakably further the
institution's legitimate business objectives, including educating and
enlightening students and faculty participating in these
projects." Activities that further these "business objectives," including
research projects that "increase the status of the
institution and lure lucrative research grants, students and faculty," are
ipso facto ineligible for the experimental use defense.
Academic research is not "philosophical inquiry," in the court's
21st-century understanding of that term, but rather a means to
advance the "legitimate business objectives" of a university. In other
words, playing with the laboratory equipment might qualify
for the defense, but using it in the course of serious academic research
would require a license. Although the Madey decision
did not extinguish the experimental use defense entirely, it eviscerated it
to the point that it is essentially useless to research
universities.
In a footnote, the court revealed that it was fully aware that "Duke ...
like other major research institutions of higher learning, is not
shy in pursuing an aggressive patent licensing program from which it derives
a not insubstantial revenue stream." (22). It is hard
to believe that this awareness did not inform the court's assessment, and
regrettable that the court was unwilling to say so. The
result is a seemingly disingenuous opinion that neither conforms to the
implications of precedent nor explains the reasons for
steering the law in a different direction, but pretends that prior courts
never meant to give research science special treatment.
Universities have embraced the patent system as patent owners and have been
in the vanguard of claimants seeking patents
on "upstream" research discoveries that would have looked too far removed
from the commercial marketplace to qualify for
patent protection just a generation ago. Universities have barely begun to
contemplate the patent system's implications for their
interests as users of the patented technology of others. Generally, it is
only when scientists have sought access to materials
and data that they could not readily duplicate themselves that universities
have entered into negotiations. Although they have
haggled over the terms of material transfer agreements, they have largely
ignored the growing number of patents covering
technology that their scientists use without license and without apology.
As universities have become increasingly aggressive as patent owners, they
have compromised their claim to disinterested
stewardship of knowledge in the public interest, leaving themselves more
vulnerable to patent infringement claims as defendants.
With their large endowments and habits of documenting their activities in
scientific publications, universities would make easy
targets. Perhaps the experimental use defense could have evolved on a
case-by-case basis as a tool for mediating between
the private interests of patent owners and the public interest in unfettered
scientific progress, but the Federal Circuit has shown
no appetite for such a nuanced role. If universities are unhappy with the
current state of the law, they may need to go to
Congress to fix it.
References and Notes
1.Madey v. Duke University, 307 F.3d 1351 (3 October 2002).
2.Act of 12 December 1980, Pub. L. No. 96-517, Section 6(a), 94 Stat.
3015, 3019-28 (1980) [codified as amended at 35
U.S. Code � 200-212 (1994)].
3.Commission of the European Communities, Proposal for a Council
Regulation on the Community Patent, art. 9 (1 August
2000), Off. J. Eur. Communities 43 (C 337E), p. 278 (28 November 2000);
available at
europa.eu.int/eur-lex/pri/en/oj/dat/2000/ce337/ce33720001128en02780290.pdf.
4.J. Mueller, Wash. Law Rev. 76, 1 (2001).
5.35 U.S.C. � 271(e).
6.9 F. Cases 1120 (D. Mass. 1813).
7.For example, 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 apparently deemed necessary in the
manufacture and sale of their product").
8.13 F. Supp. 697 (D. Colo. 1935).
9.(8) at 703.
10.R. Hantman, Experimental Use as an Exception to Patent Infringement, J.
Pat. Off. Soc. 67, 617 (1985).
11.U.S. Patent and Trademark Office, U.S. Colleges and
Universities--Utility Patent Grants, Calendar Years 1969-2000 (U.S.
Patent and Trademark Office, Washington, DC, accessed 15 December
2002); available at
www.uspto.gov/web/offices/ac/ido/oeip/taf/univ/asgn/table_1.htm.
12.M. Barinaga, Science 286, 1655 (1999).
13.Univ. of Minnesota, "Fact sheet on Glaxo-Wellcome AIDS discovery
settlement" (5 October 1999); available at
www1.umn.edu/urelate/newsservice/newsreleases/99_10glaxofacts.html.
14.M. Fisk, National Law Journal, 26 August 2002, p. A1.
15."Some senators cry foul over inclusion of patent extension rider in
spending bills," Daily Report for Executives 23 June
2000 (no. 122), p. A-41 (Bureau of National Affairs, www.bna.com,
accessed 23 June 2000).
16.35 U.S.C. � 202(d).
17.Federal Courts Improvement Act of 1982, Pub. L. No. 97-164, 96 Stat. 25
[codified as amended at 28 U.S.C. sections 41 et
seq. (1982)].
18.733 F.2d 858 (Fed. Cir. 1984).
19.35 U.S.C. � 271(e).
20.216 F.3d 1343 (Fed. Cir. 2000).
21.216 F.3d at 1352-53 (J. Rader, concurring).
22.307 F. 3d 1351, at n.7.
23.I am grateful to A. Rai, M. Van Houweling, and M. Adelman for helpful
comments on an earlier draft.
_______________________________________________
www.e-laser.org
[EMAIL PROTECTED]
