Can Patents Prohibit Research?

On the Social Epistemology of Patenting

and Licensing in Science

Justin B. Biddle
Philosophy Program
School of Public Policy
Georgia Institute of Technology
Email: justin.biddle@pubpolicy.gatech.edu

Abstract. This paper examines one important aspect of the current organization of
scientific and technological research – namely, the system of patenting and licensing
and its role in structuring the production and dissemination of knowledge. The primary
justification of patenting in science and technology is consequentialist in nature. On this
account, patenting incentivizes research and thereby promotes the development of
scientific and technological knowledge, which in turn facilitates social progress. Some
have disputed this argument, on the grounds that patenting actually inhibits the
development of knowledge. In this paper, I make a stronger argument; in some areas of
research in the US – in particular, research on GM seeds – patents and patent licenses
can be, and are in fact being, used to prohibit some research. I discuss three potential
solutions to this problem: voluntary agreements, eliminating patents, and a research
exemption. I argue against eliminating patents, and I show that while voluntary
agreements and a research exemption could be helpful, they do not sufficiently address
the problems of access that are discussed here. More extensive changes in the
organization of research are necessary. In addition to developing an argument about
patenting and licensing, this paper illustrates a promising approach to social
epistemology.

Keywords

Social epistemology
Science and values
Patenting and licensing
Intellectual property rights in science
Genetically modified (GM) seeds
Commercialization of science

Acknowledgments. Earlier versions of this paper were presented at the Center for
Interdisciplinary Research at Bielefeld University, the Georgia Institute of Technology,
the Society for Philosophy of Science in Practice (SPSP) conference at the University of
Toronto, and the Society for Philosophy of Technology (SPT) conference at ISEG,
Technical University of Lisbon. I am grateful to the participants of these workshops and
conferences, and in particular to Rachel Ankeny, Hugh Lacey, John Walsh, and Torsten
Wilholt for their comments and suggestions.
1. Introduction. A topic of growing importance within the philosophy of science is the
epistemic implications of the social organization of research (e.g., Biddle 2007; 2012;
Goldman 1999; Kitcher 1993; 2001; 2011; Kukla 2012; Longino 1990, 2002; Reiss 2010;
Wilholt 2009). By ‘social organization of research,’ I mean to include such things as the
kinds of institutions in which research takes place; how research is funded, and the
incentive structures that encourage or discourage research and information sharing.
Some ways of organizing research are conducive to the production and dissemination of
knowledge, and others are not; the examination of which is which has become especially
important given the recent changes in how research is structured. Commercial
considerations are becoming more influential, and the organization of many areas of
research is changing accordingly.1
One way in which commercial interests are changing the organization of
research concerns patents and patent licenses. Patents provide the legal right to
exclude others from making, using, selling, offering to sell, or importing patented
inventions;2 licenses are contracts between patent holders and other parties, which give
certain rights to those parties (e.g., the right to use the patented invention). Since the
early 1980s, patenting and licensing activity in science has skyrocketed, especially in the
biomedical sciences and biotechnology. Between 1983 and 2003, the number of patents
issued to U.S. universities rose from 434 to 3,259 (Walsh, Cohen, and Cho 2007, p.
1184); patenting in biotechnology has also risen significantly, from 2,000 in 1985 to over
13,000 in 2000 (Walsh, Cohen, and Arora 2003, p. 293). Other countries have witnessed
similar trends (American Association for the Advancement of Science 2007). While there
are a number of potential justifications for patenting in science, including Lockean labor-
based justifications and Hegelian personality-based justifications, the most plausible
justification is consequentialist in nature (Biddle forthcoming).3 On this account,
patenting incentivizes research and thereby promotes the development of scientific and
technological knowledge, which in turn facilitates social progress. This justification has
both an epistemic and an ethical component: patenting is supposed to facilitate
knowledge production (the epistemic component) that leads to social benefits (the
ethical component). Recently, however, a number of commentators have criticized the
consequentialist justification, on the grounds that patenting and licensing activity is
actually inhibiting research in many areas of science (Biddle 2012; Eisenberg 2008;
Heller and Eisenberg 2006). In this paper, I will make a stronger argument. In some
areas of biotechnology, patenting and licensing are not only inhibiting research; they are
prohibiting it. In particular, I will argue that patent holders and licensors of genetically
modified (GM) seeds can use, and are using, patents and license agreements to prohibit
others from doing some types of research on GM seeds. This is the first major goal of
the paper. The second is to evaluate three possibilities for improving the situation, so as

1
See, for example, Bekelman, Li, and Gross (2003), Krimsky (2003), the Introduction to Mirowski
and Sent (2002), Mirowski and van Horn (2005), and Slaughter and Rhoades (1996). On the
ideology informing these changes, see Biddle (2011). For an excellent discussion of the effects of
patenting and licensing in the biomedical sciences, see Walsh, Cohen, and Cho (2007).
2
35 U.S.C. §§ 154, 271(a).
3
For a discussion of potential justifications of intellectual property in general, see Hughes (1988).

1
to better achieve our epistemic aims. An evaluation of these possibilities will
demonstrate that a significant reorganization of research will likely be required to solve
the problems that are raised here.
The philosophical framework of this paper is a particular strand of social
epistemology. Social epistemology covers a variety of different topics, including trust in
experts and testimony, reasonable disagreement, and systems design. There are a
variety of approaches to each of these; with regard to the latter, one category of
approaches might be termed ideal systems design, which when applied to science and
technology, takes as its primary or exclusive philosophical aim the articulation of an ideal
way of organizing research. The question of how we are to achieve that ideal is
acknowledged to be important and challenging – but not a matter for philosophers. Philip
Kitcher, in Science, Truth, and Democracy (2001), takes this approach in his proposal
for incorporating public participation into scientific decision making.4 (In his more recent
work (2011), he has begun to move in a somewhat different direction.)
The approach that I adopt is non-ideal systems design; instead of beginning by
articulating an ideal, one begins with the actual organization of research, in all of its
messiness, and attempts to improve how research is organized in a piecemeal, iterative,
and empirically-based manner. More specifically, this approach begins with an
examination of a specific aspect of the organization of an area of research, and the
implications of this form of organization for knowledge production. Of particular
importance are those features that impede the production or dissemination of
knowledge.5 One then predicts the effects of modifying particular organizational features,
and makes a corresponding proposal. Once the proposal is adopted, the iterative
process begins anew. This approach is similar to the strategy of ‘adaptive management,’
which has been defended by Mitchell (2009) in the context of policy-relevant research
and by Reiss (2010) in the context of biomedical research, but it can be traced back at
least as far as Dewey (1938). In this paper, I will use this approach to examine one
aspect of the organization of biotechnology – namely, the role of patents and licenses in
structuring the production and dissemination of knowledge – and to discuss specific
ways of redesigning this aspect so that we can better achieve our epistemic aims.
Given the intense controversy surrounding GM seeds, it is perhaps necessary to
make one final clarification before proceeding. This paper is not an evaluation of the
safety profile or environmental impact of GM crops. My own view is that there is nothing
intrinsically wrong with using recombinant DNA technologies to modify the genomes of
plants; some particular cases of genetic modification might prove to be problematic for
any number of reasons, but under the right circumstances, these techniques can be both
safe and environmentally unproblematic. However, there is no doubt that GM crops
should be subjected to extensive research; this is especially true given that seeds are
self-replicating and can be spread by wind, bees, and other means that are outside of
human control. This paper argues that the current system of patenting and licensing

4
Brown (2004, 82-83) makes a point similar to this.
5
This project, which one might call the critical project in systems design, has its roots in the work
of Marx, Mannheim, and others, and is similar to work in the emerging field of agnotology (e.g.,
Elliott 2013, Oreskes and Conway 2010, Proctor and Schiebinger 2008).

2
allows for the prohibition of some research on GM seeds, and it will suggest potential
ways of remedying this epistemic deficiency.

2. Patenting and Licensing of GM Seeds. GM seeds are seeds that have been
modified using recombinant DNA technologies. A target gene in one organism is
identified and isolated, and DNA sequences are inserted ahead of the gene (a promoter)
and behind it (a terminator) in order to regulate the gene’s expression. These are then
transferred into another organism via a vector such as a bacterium. While there are, in
principle, an almost infinite number of ways in which organisms might be modified
through the use of these techniques, almost all of the GM seeds currently on the market
are modified either to produce an insecticide (e.g., Bacillus thuringiensis) or tolerate an
herbicide.6 Of the latter, Monsanto’s Roundup Ready® seeds are the most common;
these seeds have been modified to tolerate N-phosphonomethylglycine- (or glyphosate-)
based herbicides, such as Monsanto’s Roundup®. Farmers who use this technology can
control weeds by spraying their entire crop with Roundup®, which kills all plants except
those modified to tolerate it. GM seeds are typically covered by numerous patents,
including patents on promoters for transforming the genetic material of target organisms;
patents on genes that are associated with desired phenotypic traits, and patents on plant
cells that contain those genes.
In order to obtain access to patent-protected GM seeds, one must sign a license
agreement with the patent holder. A seed company might wish to produce and sell a GM
seed that is patented by some other country; doing so requires a license agreement
between the two companies. (Pioneer, for example, is one of Monsanto’s licensed seed
producers.) Similarly, a farmer who wishes to use GM seeds for the purpose of growing
them, and a researcher who wishes to use GM seeds for the purpose of research, must
sign a license agreement, which stipulates the conditions under which they may use
them. For example, use of Monsanto’s patent-protected seeds for the purpose of
growing them is subject to a license called the “Monsanto Technology/Stewardship
Agreement,” according to which growers agree to a number of conditions, including:

• “To use Seed containing Monsanto Technologies solely for planting a single
commercial crop.”
• “Not to save any crop produced from Seed for planting and not to supply Seed
produced from Seed to anyone for planting other than to a Monsanto licensed
seed company.”
• “Not to save or clean any crop produced from Seed for planting and not to supply
Seed produced from Seed.”
• “…. Grower may not plant and may not transfer to others for planting any Seed
that the grower has produced containing patented Monsanto Technologies for
crop breeding, research, or generation of herbicide registration data” (Monsanto
Company 2008).

6
See, for example, the 2 May 2013 issue of Nature, entitled GM Crops: Promise and Reality.

3
Technology agreements such as this are standard in the industry; other seed companies
have agreements with similar restrictions. There are a number of important and
controversial elements of these agreement, two of which I will discuss here.
First, the agreement prohibits the planting of more than one commercial crop;
saving seed for use in another growing season is not allowed. This particular prohibition
has been the subject of a number of recent patent infringement cases in the U.S.,
including Monsanto v. McFarling (2004), Monsanto v. Scruggs (2006), and Monsanto v.
Bowman (2011). While there are a number of issues at stake in each of these cases,
one common argument against Monsanto is that its prohibition of seed saving violates
the doctrine of patent exhaustion, which states that a patent holder’s right to exclude
others from using or selling an invention is exhausted after an authorized sale. Suppose,
for example, that I purchase an alarm clock that contains a patented technology from an
authorized dealer. After this authorized sale, the patent holder’s right to exclude me from
using or selling this item is exhausted. Many have argued that owners of GM seeds have
no right to prohibit licensees from reusing seeds, because after the first purchase, their
patent rights are exhausted.
This argument was most recently discussed in Monsanto’s patent infringement
case against Vernon Bowman. In 2002, Bowman purchased Pioneer Hi-Bred® seeds,
which contain Monsanto’s Roundup Ready® technology, and signed the “Pioneer Hi-
Bred Technology Agreement,” which contains restrictions identical to those in the
Monsanto agreement. From 1999 to 2007, Bowman planted Roundup Ready® soybeans
and, consistent with the agreement, did not save seeds. However, beginning in 1999,
Bowman also purchased commodity seeds from a local grain elevator and planted these
seeds in a second, or late-season, planting; some of these seeds contained Monsanto’s
patent-protected genes, and some did not.7 The Technology Agreement that Bowman
signed did not explicitly extend to these commodity seeds, and purchase of commodity
seeds does not carry an obligation to sign a license agreement; because of this,
Bowman believed that it was legal for him to save seeds from his late-season plantings,
even seeds from crops that contained Monsanto’s patent-protected genes, and replant
them. He saved and replanted seeds from plants that exhibited a resistance to
Roundup®, and he was open about doing so, including with representatives from
Monsanto. Monsanto then sued Bowman for patent infringement. In a lower court case,
Bowman argued that he purchased the commodity seeds in an authorized sale, and that
once this occurred, Monsanto’s rights to exclude him from re-using the seeds were
exhausted (Monsanto v. Bowman 2011).
In 2011, a U.S. Federal Court of Appeals agreed with Monsanto. Bowman
appealed the decision, and the case was taken up by the U.S. Supreme Court. In 2013,
the court decided unanimously in favor Monsanto.8 In particular, the court upheld the
view that the prohibition of replanting GM seeds is not a violation of the doctrine of

7
Commodity seeds are undifferentiated seeds that are often used as feed for livestock. Bowman
®
used these seeds instead of the more expensive Roundup Ready seeds in the late-season
planting, because late-season plantings are at higher risk of being ruined by bad weather.
8
This decision is available at: http://www.supremecourt.gov/opinions/12pdf/11-796_c07d.pdf
(Accessed August 18, 2013).

4
exhaustion. The court argued that, in replanting seeds, Bowman was not using the
seeds but was making copies of them and was thus guilty of patent infringement. The
court’s rationale behind this argument is significant. It argued that, if farmers were
allowed to replant seeds, then a patent holder’s right to exclude others would extend
only to the first generation of seeds. If this were the case, then patent holders would
have no real incentive to invent these seeds in the first place. GM crops have at least the
potential to be beneficial, and given that they are time consuming and expensive to
develop, corporations that develop them should have a period of monopoly protection,
during which time they can recoup their investments. If patent rights on GM seeds were
exhausted after the first growing season, then patents would (according to the court) fail
to provide the incentives that they are designed to provide. The consequentialist
justification for patenting thus played an important role in the Supreme Court’s decision.

3. The Implications of Technology Agreements for Research. The second important

and controversial element of Monsanto’s Technology Agreement – and the one that is
the primary focus of this paper – is its prohibition of research, and its prohibition of the
distribution of seeds to others for the purpose of research. Technology agreements such
as this cannot prohibit all types of research on GM seeds. For example, they do not
prohibit observational studies that measure characteristics of the environment in areas in
which GM crops have been planted. They cannot prohibit such observational studies,
because one can perform them without having to purchase any seeds at all. One can
simply measure characteristics of an environment that contains GM crops – crops that
were purchased and planted by others, who themselves might be in compliance with the
relevant license agreements. However, technology agreements can, and do, prohibit
other types of research, such as controlled field trials; more generally, they can prohibit
any trial for which a researcher must obtain seeds from a patent holder or licensor. And
of course, many of the most sensitive and reliable trials fall into this latter category.
In order to obtain access to GM seeds for the purpose of research, scientists
must ask permission from the relevant seed companies, describing in detail the design
of experiment to be performed, and then negotiate the terms of the agreement, including
the conditions under which the results of the experiment can be publicized (Waltz 2009).
The process of requesting approval is itself a deterrent; it can be time consuming, and
there is no guarantee that seed companies will grant permission to perform research and
disseminate research results. Indeed, there are several cases – some to be discussed
later in this section – in which seed companies have either refused to comply with a
request for research or refused to allow research results to be publicized.
Prima facie, the ability of patent holders to prohibit others from doing some
research on their inventions is epistemically problematic. For example, seed companies
have the ability to prohibit others from performing many studies on the environmental
effects of their GM crops; this could significantly slow the development of our knowledge
of these effects. One might object to this being epistemically problematic in two different
ways. First, one might argue that patent holders perform all of the research on their
inventions that need to be done. Companies perform environmental studies of their GM
crops in the course of gaining market approval for them; one might argue that these

5
studies provide us with all of the knowledge that we need regarding the environmental
effects of these crops and that, as a result, the ability to prohibit others from performing
research has no detrimental epistemic consequence. This response, however, is
completely implausible. It is implausible to believe that any one entity could acquire all of
the knowledge that we could want about a particular technology; there is simply too
much that we might wish to know about a technology for any one entity to acquire it all,
especially in the limited time that is required to gain market approval. Moreover, even if
one company could, in principle, develop all of the knowledge that we need about a
particular technology, history has demonstrated that for-profit corporations have a
tendency to focus upon research that is likely to be beneficial to them financially and to
avoid research that is not. This is a well-known problem with industry-sponsored
pharmaceutical research, and it affects industry-sponsored agricultural biotechnology as
well (Elliott 2013). Additionally, for-profit corporations have a history of biasing studies to
increase the likelihood of obtaining desired results.9 One such example of this,
concerning the effects of a type of transgenic corn on the lady beetle, will be discussed
later in this section.
One might have a second reason for questioning whether the ability of patent
holders to prohibit some research on their inventions is epistemically detrimental. Simply
because patent holders have the legal right to prohibit others from doing research on
their inventions does not mean that they actually do so. For example, many patent
holders in many areas of academic research simply do not enforce their patents; they
allow others to infringe (Walsh, Cohen, and Arora 2003). Because of this, it is not
enough to show patent holders have the legal right to prohibit others from doing
research on their inventions; one must also show that they actually do so. Is there
evidence of this? While there has been no systematic study done on the extent to which
seed companies actually prohibit others from researching their inventions, there is
sufficient evidence to conclude that the prohibition of research from patents and licenses
is a serious problem. Some of this evidence comes in the form of individual cases, and
some comes in the form of corporate policy.
One of these cases concerns Paul Gepts, a plant geneticist at the University of
California, Davis, who wanted to test for the presence of GM maize in Mexican
households. (GM maize is prohibited in Mexico.) In 2002, Gepts requested seed
samples from three different seed companies for the purpose of comparing the genetic
material of the seeds with those found in Mexican households. Gepts’s request came
approximately a year after Ignacio Chapela and David Quist published their article in
Nature, which reported the cross-pollination of a type of transgenic corn with a local
Mexican variety (Quist and Ignacio 2001). Gepts requested samples from Monsanto,
Pioneer, and Syngenta. Monsanto responded that he should get a powder sample from
Europe (a sample that did not work well for his experiment); Pioneer responded that it
did not have anything appropriate to share, and Syngenta responded that he should ask
the Mexican government for a sample (Waltz 2009).

9
For systematic reviews on the effects of financial conflicts of interest in biomedical research, see
Bekelman, Li, and Gross (2003) and Lexchin et al. (2003). For discussions from the perspective
of social epistemology, see Biddle (2007), Brown (2008; 2008a), and Wilholt (2009).

6
 Another case concerns a group of twenty-four leading corn insect researchers
working at public research universities in the US who were unable to obtain access to
GM seeds for the purposes of research. In 2009, the Federal Insecticide, Fungicide, and
Rodenticide Act Scientific Advisory Panel (FIFRA SAP) met to discuss Environmental
Protection Agency (EPA) standards concerning the registration of Plant-Incorporated
Protectants (PIPs), which are pesticides produced by plants. The group of corn insect
researchers posted a public comment to the FIFRA SAP, claiming that seed companies
were inhibiting their ability to study biotech crops. Their statement is as follows:
Technology/stewardship agreements required for the purchase of genetically
modified seed explicitly prohibit research. These agreements inhibit public
scientists from pursuing their mandated role on behalf of the public good unless
the research is approved by industry. As a result of restricted access, no truly
independent research can be legally conducted on many critical questions
regarding the technology, its performance, its management implications, IRM
[insect resistance management], and its interactions with insect biology.
Consequently, data flowing to an EPA Scientific Advisory Panel from the public
sector is unduly limited. Given the importance of the FIFRA SAP process to an
effective and credible assessment of new PIPs on behalf of the American public,
we urge EPA to require registrants to remove the prohibition on research on their
products and specifically allow research by public-sector scientists.10
The group posted this comment anonymously out of fear that it might affect their ability
to obtain funding for future research. The comment provides evidence that the
prohibitions on research contained in technology agreements actually have the effect of
prohibiting research. Following the publication of this comment, Monsanto responded by
negotiating an umbrella agreement with ARS scientists, which allows them to perform
some types of research (“agronomic research”) without having to ask permission for
each study (Waltz 2010). Whether this agreement is sufficient will be discussed in
Section 5.
It might be argued that these are isolated cases that happen only rarely. The
frequency with which such cases occur, again, is unknown, as there has been no
systematic study investigating the issue. There are, however, examples of corporate
policies that suggest that the problem might be widespread. For example, in 2007,
Syngenta put into place a policy that prohibited outside scientists from performing
comparison studies on Syngenta’s GM seeds. According to a spokesperson from
Syngenta, many of its products contain parts that are licensed from other companies,
and agreements with those other companies do now allow for comparison testing (Waltz
2009). Comparison studies are an important type of experiment on GM crops, and the
fact that one prominent seed company has a blanket policy prohibiting them is significant
and epistemically worrisome.
A case involving the prohibition of publicizing results concerns Pioneer and its
development in 2001 of corn that was genetically modified to contain the toxin

10
The comment can be found at: http://www.regulations.gov/#!documentDetail;D=EPA-HQ-OPP-
2008-0835-0017. Accessed 9 July 2013.

7
Cry34Ab1/Cry35Ab1 for combating rootworms. Pioneer asked some university scientists
to study the effects of the corn on a type of ladybug; the result was that, after eight days,
almost 100% of the ladybugs that ate the corn died. This result was communicated to
Pioneer and, according to a participating scientist who asked to remain anonymous,
Pioneer responded: “you are under no circumstances able to publicize this data in any
way” (quoted in Waltz 2009, p. 882). Two years later, Pioneer received approval from
the EPA to market a GM corn with this same toxin. In the process of obtaining approval,
Pioneer performed its own studies on the corn and sent the results to the EPA; some of
these examined the effects of the corn on ladybugs. One of these studies stopped
collecting data after seven days (one day shorter than the study by the university
scientists). Arguably, what Pioneer did in this case was to draw on the results of the
earlier study, which Pioneer kept secret, in order to jerry-rig the results of its later studies
in order to increase the probability of obtaining a desired result. If this is the case, then
Pioneer did not merely use patents and licenses to keep others ignorant of a particular
effect; it used them to disseminate a falsehood. At least one scientist at Pioneer would
dispute this characterization. He noted that, while the corn approved by the EPA
contains the same toxin as the one studied by the university scientists, it is a different
construct; “key genes were integrated into a different place in the genome” (Waltz 2009,
882). The suggestion is that, because of the different construction, the approved corn is
not a threat to ladybugs. Unfortunately, whether or not the approved corn is a threat is
unknown; after the crop was put on the market, the same university scientists requested
permission to redo their previous study with the approved corn, and their request was
denied (Waltz 2009).
These examples demonstrate that not only do patent holders have the legal right
to prohibit others from conducting some studies on their inventions, but that they actually
do so. Because there have not been systematic studies of this problem, it is unclear how
frequent such prohibitions occur. But given that some seed companies have policies that
actually prohibit others from performing certain types of studies (e.g., Syngenta’s policy
on comparison studies), it is reasonable to think that such prohibitions occur frequently.
This is highly problematic from an epistemic point of view, and given that ignorance of
the effects of technologies can impact human health and the environment, it is highly
problematic from a moral perspective as well.

4. The Experimental Use Exemption in U.S. Patent Law. One might object to the
argument thus far by arguing that patent holders do not have the legal right to prohibit
research, because of the existence of a research, or “experimental use,” exemption. It is
true that there is an experimental use exemption in US law; however, this objection
misconstrues both the status and the scope of the exemption. There is no general
statutory exemption for research in the U.S.; there is a common law exemption that
dates back to an 1813 Massachusetts District Court case, Whittemore v. Cutter. In this
case, Justice Joseph Story stated: “it could never have been the intention of the
legislature to punish a man, who constructed such [a patented] machine merely for
philosophical experiments, or for the purpose of ascertaining the sufficiency of the
machine to produce its desired effects.” In that same year, in Sawin v. Guild, Story

8
attempted to clarify the exemption by arguing that it applies to users who have no profit
motive. Still, while judges since these cases have affirmed that there is an experimental
use exemption, there was for many years no clear specification of which activities are
included within it.
This changed in 2002 in a patent infringement case involving John Madey and
Duke University (Madey v. Duke 2002). Madey was a physicist who had developed a
free electron laser (FEL) laboratory at Stanford University in the mid-1980s. In the
course of his research, he had acquired a number of patents that covered various
aspects of FEL. In 1989, Madey moved to Duke, which built an addition to the physics
department to house his lab. Madey continued his FEL research at Duke for several
years, until a dispute arose between himself and the university over his management of
the lab. Following this, Madey left. Physicists at Duke continued to run the FEL lab, and
Madey sued Duke for patent infringement.
In response, Duke argued that it was not guilty of patent infringement, because
its activities were covered by the experimental use exemption. The case was taken up
by the U.S. District Court for the Middle District of North Carolina; at issue was whether
the activities of the physicists running the FEL lab were commercial in nature. Madey
argued that they were; on the basis of language taken from Duke’s patent policy, Madey
argued that Duke is in the business of “obtaining grants and developing possible
commercial applications for the fruits of its ‘academic research’” (Madey v. Duke 2002).
Duke, however, pointed to other language in the policy, which stated that Duke was
“dedicated to teaching, research, and the expansion of knowledge… [and] does not
undertake research or development work principally for the purpose of developing
patents and commercial applications” (Madey v. Duke 2002). The District Court sided
with Duke; it argued that Madey had the burden of establishing that Duke’s use of the
patented inventions had “definite, cognizable, and not insubstantial commercial
purposes” (Madey v. Duke 2002), and that Madey failed to do so.
Madey appealed to the U.S. Federal Circuit Court of Appeals, which overturned
the lower court’s ruling. One of the main arguments of the Court of Appeals was that the
District Court interpreted ‘experimental use’ too broadly. First, the Court of Appeals
argued that a commercial implication need not be substantial in order to be disqualified
from exemption; “use is disqualified from the defense if it has the ‘slightest commercial
implication’” (Madey v. Duke 2002). Duke’s use of the FEL lab arguably had some
commercial implications, insofar as it helped to bring in grants, increase the overall
prestige of the university, and so on; in that case, according to the ruling, the use is not
‘experimental.’ Second, and more significantly, the Court of Appeals argued that uses
that are “in keeping with the alleged infringer’s legitimate business, regardless of
commercial implications,” are not exempt (Madey v. Duke 2002, emphasis added). Thus,
the main issue is not whether an activity has a commercial implication, but rather
whether it is a part of alleged infringer’s “legitimate business.” The legitimate business
objectives of any university include educating students, supporting research, and
furthering knowledge; any activity that takes place within a university that furthers these
interests is thus not exempt. Given this very narrow interpretation of the concept of
experimental use, which activities are exempt? Only those that are performed “for

9
amusement, to satisfy idle curiosity, or for strictly philosophical inquiry” (Madey v. Duke
2002). Incidentally, it follows from the court’s ruling that – much to the chagrin of
philosophy departments around the world – ‘strictly philosophical inquiry’ is not the
‘legitimate business’ of anything. More importantly, it follows that very few activities
quality as ‘experimental,’ and that no experiments on patented inventions by university
scientists that are performed in their capacity as university scientists could be exempt
from patent infringement.
Thus far, I have argued that patents and licenses can be, and are in fact being,
used to prohibit some research on GM seeds. Given the narrow interpretation of
‘experimental use’ in Madey v. Duke, no research on GM seeds – even research on
environmental effects or safety profile – would qualify as exempt from patent
infringement. The epistemic implications of this are clear. Regardless of one’s views on
the risks and benefits of GM crops, it is uncontroversial that this technology – like any
other – should be subjected to critical scrutiny. Current laws and policies surround
patents and licenses do not allow for this – at least not to the extent that they should.
The primary justification for patenting in science is that it facilitates research and
development, which in turn leads to social benefits. In this area of biotechnology,
however, patenting and licensing allow for the prohibition of research. It is time now to
turn to the question of how this problem can best be solved.

5. On Voluntary Agreements. In the remainder of the paper, I will discuss three

potential solutions to this problem: voluntary agreements, eliminating patents, and
revising patent law to incorporate a substantial exemption for research. I will argue
against eliminating patents, and I will show that while voluntary agreements and a
research exemption could be helpful, they would not sufficiently address the problems of
access that I have just discussed. If we are to address these problems adequately, more
extensive changes in the organization of research will be necessary.
In response to the criticisms of restrictive license agreements by academic
scientists, Monsanto created a blanket license agreement – the Academic Research
License (ARL) – that allows academic scientists from many US universities to perform
some studies without asking permission for each study. Types of research that are
explicitly permitted under ARL are:
• agronomic and yield comparisons;
• testing for compositional profile such as oil content;
• studies relating to end-use such as animal feeding;
• comparative efficacy studies;
• studies on interactions of the trait of the trait with pest biology and pest
management practices including interactions related to resistance management;
• studies on interactions of introduced traits with the environment. (Monsanto
Company)
The ARL represents a significant step in the right direction; it allows many scientists to
perform many types of research on Monsanto’s patent-protected seeds – including
research on interactions with the environment – and it does so without introducing any
new burdens on scientists. Given this, one might argue that the best way to ensure that

10
scientists have the freedom to perform studies on GM seeds is to encourage seed
companies to create similar voluntary agreements.
Without denying the benefits of ARL, there are at least five respects in which
agreements such as this are problematic. First, ARL is limited in terms of the class of
scientists who benefit from the agreement. “ARLs are in place with all major
agriculturally-focused US universities – about 100 in total” (Monsanto Company). One
might argue that this is sufficient for the US, but it is a significant limitation that no
scientist outside of the US is covered by the agreement. Second, ARL is limited in terms
of areas of research that are allowed. The following areas of research are explicitly not
included in ARL:
• breeding with plants produced from the seed;
• reverse engineering or characterizing the genetic composition of patent-
protected traits in seed;
• development of methods for detecting the presence or absence of patent-
protected traits in seed;
• use of non-commercial methods to detect the presence or absence of patent-
protected traits in seed;
• research on modifications or improvements to the patent-protected traits
(Monsanto Company).
Some of these areas that are not included could have important implications for our
knowledge of the environmental effects of GM seeds. For example, a significant issue
concerning GM crops is their ability to spread beyond the boundaries within which they
were originally planted. Developing methods for detecting the presence of patent-
protected seeds could help us to understand this better, but this area of research is not
included in ARL. Additionally, research on improvements to patent-protected traits could
be beneficial in many respects, and yet this type of research is also not included in ARL.
Third, ARL is vague. There are some areas of research that are neither explicitly
permitted nor forbidden by the agreement, which makes it difficult for scientists to know
whether they could face legal action by performing studies in these areas. One important
and broad area of research that is neither explicitly permitted nor forbidden is research
on the health effects of GM crops (Monsanto Company). Fourth, voluntary agreements
are, obviously, voluntary; as a result, the seed company in question can revise or
completely abandon the agreement at any point. Finally, agreements such as ARL are
company specific. ARL is a license that allows some scientists to perform some studies
on Monsanto’s GM seeds, but as a Monsanto agreement, it does not cover seeds by
other companies, such as Dow AgroSciences, Syngenta, and Pioneer Hi-Bred. As of
October 2010, the latter three companies had “begun discussions with ARS over new
umbrella agreements,” but as of this writing, no agreement has been instituted.
The strategy of encouraging voluntary agreements between scientists and seed
companies is the most conservative of the strategies that I will discuss. It leaves the
current patent system intact, and it relies upon the good will of corporations to ease
sufficiently the restrictions that they have hitherto put into place. As Monsanto’s ARL
indicates, voluntary agreements can be beneficial. But they are also limited in a number

11
of respects, so much so that they are unlikely to allow academic researchers sufficient
freedom to study all significant hypotheses concerning GM seeds.

6. Eliminate Patents? The most radical of the strategies that I will discuss is to
eliminate patents in agricultural biotechnology.11 Adopting this strategy would have one
obvious benefit: it would eliminate all of the problems that have been discussed in this
paper. Despite this, there are a number of potential problems with the proposal.
The first is the obvious problem of practical feasibility. There are powerful,
entrenched interests that would forcefully resist the move to eliminate patents, so much
so it is safe to say that this strategy will not be implemented, at least not in the near
term. Thus, while eliminating patents in agricultural biotechnology might be the best
strategy in theory, it is unlikely to succeed in practice.
In addition to this practical problem, however, there are also reasons to question
the theoretical adequacy of the proposal. The proposal seems to assume that it is the
mere ability to hold patents that gives rise to the problems discussed in this paper, but
this is not the case. Rather, it is the current system of patenting and licensing that gives
rise to these problems; the ability to hold patents is only one component of this system
(Biddle forthcoming; Brody 2006, 2006a). More specifically, the current system in the US
can be analyzed into the following five components: (1) it allows patents on products as
well as processes; (2) it grants patents to the first to file a patent application, as opposed
to the first to invent; (3) it provides extensive rights of exclusion; (4) the requirements for
obtaining a patent – an invention must be statutory, novel, useful, and non-obvious – are
jointly sufficient for obtaining a patent, and (5) the requirements are interpreted broadly.
These components are discussed in detail elsewhere, and not all of them are relevant to
the issues of this paper.12 Important for our purposes is (3). Patents, again, provide the
legal right to exclude others from making, using, selling, offering to sell, or importing
patented inventions, but what it means to make or to use an invention can be interpreted
in different ways. In particular, there is considerable latitude that a system could have in
determining which kinds of uses can be excluded. The current system allows patent
holders to exclude almost any use, including almost all research.
The fact that it is the current system of patenting and licensing, and not the mere
ability to hold patents, that gives rise to the problems discussed in this paper has
important implications for the question of whether patents in agricultural biotechnology
should be eliminated. Analogous to the problem of underdetermination in confirmation
theory, there are a multiplicity of ways to solve the problem of patents and licenses
prohibiting research; one is to eliminate patents, but there are others as well.
In deciding which of these potential solutions to pursue, we ought to be guided by
what one might call the virtue of conservativism.13 More specifically, we ought to pursue
the strategy that will most will most likely solve the problems that we need solved, while

11
The activist Vandana Shiva has argued against patents in agriculture on many occasions (e.g.,
Shiva 2000). Brown (2008; 2008a) has made this argument in the context of biomedical research.
12
For discussion of these components, see Biddle (forthcoming) and Brody (2006; 2006a).
13
This is in certain respects similar to Quine and Ullian’s virtue of conservatism (Quine and Ullian
1978).

12
at the same time retaining as much of the rest of the system as possible. This is a
prudent strategy if one cares about the practical problem of implementing a proposal, but
it is also prudent in its acknowledgment that most systems have benefits as well as
costs. While the current patent system has significant costs, it also has some benefits.
Most important, patents provide incentives for research and development. The current
system might not do this as efficiently as it could – indeed, one of the main arguments of
this paper is that it does not – but it is still the case that patents provide incentives, and if
it is possible to revise the system in such a way as to eliminate its problems while
maintaining its benefits, then that is what we ought to do. Of course, it is possible that
multiple attempts to revise the system will lead to naught; if this turned out to be the
case, then the proposal to eliminate patents would become more plausible. But given the
benefits of patents, and given the risks associated with dismantling the patent system,
we should only consider eliminating the system after first attempting to revise it.

7. On the Implications of a Substantive Experimental Use Exemption. At least

initially, it would appear that an adequately formulated experimental use exemption
would have the potential to both solve the problems highlighted in this paper and ensure
that patents continue to fulfill their intended function of incentivizing research and
development. As we will see, this initial appearance is somewhat misleading. More
specifically, while an adequately formulated experimental use exemption could, in
principle, solve the problems discussed in this paper, it would not likely solve them in
practice.
As we have seen, there is no general statutory research exemption in the US;
rather, there is a common law exemption, the boundaries of which have been formed
over time. If one were to attempt to make explicit the formulation that has evolved, it
would be something like this:

• The U.S. Experimental Use Exemption: Research is covered by the experimental

use exemption if it is done for amusement, to satisfy idle curiosity, or for strictly
philosophical inquiry. Research is not covered if it is in keeping with the
researcher’s legitimate business, regardless of commercial implications.

As argued in Section 4, this formulation is highly problematic, primarily because it is so

narrow as to exempt almost no research. For example, research that is conducted by
university or government scientists on genetically modified crops would be “in keeping
with the researcher’s legitimate business;” as a result, it would not be exempt under this
formulation.
One can look to the European Union, and in particular to Germany, for a better
formulation. In 1975, the Community Patent Convention, or Luxemburg Convention
(CPC), was established in order to harmonize the disparate patent laws of European
nations. One of its more specific goals was to establish guidelines for determining which
uses of patented inventions constitute infringement. In Article 27(b) of the CPC, we find
the following formulation of an experimental use exemption:

13
 • CPC Experimental Use Exemption: The rights conferred by a European Union
Community patent do not extend to “acts done for experimental purposes relating
to subject matter of the patented invention.”

Despite the fact that the CPC has never been ratified, it has still exerted an influence on
the patent laws of various European countries. For example, France, the United
Kingdom, and Germany have all adopted experimental use exemptions that contain
language that is virtually identical to that of the CPC. Each country’s court system,
however, has interpreted the exemption differently. The exemption extends not to
experimental acts per se, but to experimental acts that relate to the subject matter of the
patented invention; this last clause is at the root of the different interpretations of the
exemption. An exhaustive comparison between these different interpretations is beyond
the scope of this paper.14 For the present purposes, it will suffice to discuss the German
interpretation, as it is the clearest and, in my view, the most effective at facilitating
knowledge production.
There are two issues in particular that call for discussion: how broadly one should
interpret ‘relating to the subject matter of a patented invention,’ and whether the
exemption can cover experiments with some commercial purpose. Paradigm examples
of experiments that relate to the subject matter of a patented invention are experiments
that validate whether the invention can be constructed (or isolated, or purified) according
to the methods specified in the patent application, and experiments that determine
whether the invention actually has the properties or effects that are specified in the
application. A matter of debate is whether experiments that involve the use of a patented
invention in order to intervene in a different process should be exempted. One of the
most important German cases on patent infringement, Clinical Trials I, examines this
issue (Klinische Versuche I 1995). The case involves patents on pharmaceutical
compositions of a polypeptide with human immune interferon (IFN-gamma) properties.
One of the defendants in the case was licensed to produce and sell this pharmaceutical
(“Polyferon”) for the treatment of classical rheumatoid arthritis; in addition to doing this,
however, they also conducted experiments to determine further indications of Polyferon
(including cancer, AIDS, leukemia, allergies, and others), with an eye toward acquiring
licenses for these as well (Klinische Versuche I 1995, 623). At issue in this case was
whether these experiments constitute patent infringement.
The court held that they do not. Important in reaching this decision is the fact that
the CPC, and the relevant section of German patent law (§ 11 No. 2 of the German
Patent Act), does not place any limitations on which experimental acts count as relating
to the subject matter of a patented invention. Because of this, experiments on the use of
an invention to discover new indications fall within the purview of the exemption, as do
experiments that have some commercial implications. According to the court:
the wording of the [German Patent] Act when examined naturally… indicates that
§ 11 No. 2 of the Patents Act in principle exempts all experimental acts as long

14
See Centre for Intellectual Property Policy & Health Law Institute for a comparative analysis of
different experimental use exemptions. http://www.cipp.mcgill.ca/data/newsletters/00000050.pdf
(Accessed August 18, 2013).

14
 as they serve to gain information and thus to carry out scientific research into the
subject-matter of the invention, including its use. There are then included, for
example, utilization acts for experimental purposes undertaken with the subject-
matter of the invention, in order to discover the effects of a substance or possible
new uses hitherto unknown. Since the provision makes no limit, either qualitative
or quantitative, on the experimental acts, it cannot matter whether the
experiments are used only to check the statements made in the patent or else to
obtain further research results, and whether they are employed for wider
purposes, such as commercial interests. (Klinische Versuche I 1995, 639,
emphasis added)
As is evident, the scope of the German experimental use exemption is broad – and
sufficiently broad to exempt in principle the research on GM seeds discussed in this
paper from patent infringement. (I will discuss the inclusion of the “in principle” shortly.)
There are strong philosophical reasons for interpreting the CPC experimental use
exemption in this way. The subject matter of a patented invention should include some
of the uses to which that invention can be put, because the subject matter of an entity
includes the ways in which it interacts with its environment. For example, the ‘subject
matter’ of a particular strand of DNA includes not just the structure of that strand, but
also the way in which it functions in its broader cellular environment. Because of this, it
would be arbitrary to exclude experiments on all uses of an invention from falling under
the exemption. This does not imply that research on just any use should be exempted.
For example, under the German experimental use exemption, research on the
marketability of an invention is not exempted, as it is not deemed to be ‘related to the
subject matter of the invention’ (Klinische Versuche II 1998). But some uses are
sufficiently related to the subject matter that experiments on them should be exempted.
Furthermore, there are strong grounds for allowing some research that has
commercial implications. Experiments into the subject matter of an invention – that is,
experiments that seek to increase our understanding of the subject matter in question –
can have commercial implications. This is especially true in areas such as agricultural
biotechnology and biomedical research, but if one interprets ‘commercial implication’
sufficiently broadly, then almost any area of research will be commercially relevant.
Because of this, any formulation of a research exemption must find a balance between
exempting research that leads to epistemic benefits, even if this research has
commercial implications, and providing protection for patent holders. Exempting
experiments that relate to the subject matter of the patented invention is, at least prima
facie, a reasonable way of doing this. The German experimental use exemption is one
that is conducive to the development of scientific knowledge; because of this, adopting it
could have significant epistemic benefits. Moreover, it is arguable that these benefits
come at little or no cost to the patent holder. The experimental use exemption allows
only research that is related to the subject matter of a patented invention; it does not
allow the researcher to then make or sell that invention. For example, if one were to
conduct experiments to find new uses for a patented pharmaceutical, one would not be
able to simply market this pharmaceutical for new uses. One would either need to
purchase a license from the patent holder to sell the pharmaceutical, or one would need

15
to wait until the patent life on the pharmaceutical had expired. Purchasing a license
would be financially beneficial for the patent holder; waiting until the patent life has
expired would arguably not affect the patent holder at all. In either case, there is no
significant cost to the patent holder.
The aim of this section has not been to argue that the German interpretation of
the CPC experimental use exemption is the best possible interpretation; I have not
discussed other interpretations, and there might be formulations that are better than the
CPC. At the very least, however, this formulation and interpretation is a significant
improvement over the one that is currently in effect in the US, and it highlights the
epistemically (and morally) important issue of instituting a broad research exemption.
As noted at the beginning of this section, however, it is misleading to think that a
research exemption – even one that is formulated broadly – would solve all of the
problems that are raised in this paper. More specifically, while the German interpretation
of the CPC exemption would, in principle, allow researchers to perform studies on GM
seeds, its benefits in practice might be limited. To see why this is so, we must pay close
attention to the distinction between a patent and a patent license.15 A patent, again,
provides the right to exclude others from making, using, selling, offering to sell, or
importing patented inventions; a patent license, again, is a contract between a patent
holder and another party, which gives certain rights to that party (e.g., the right to use
the patented invention). An experimental use exemption such as the CPC exemption
allows researchers to do many things that they could not otherwise do. They could, for
example, perform studies that test whether patented strains of bacteria provide the
herbicide tolerance that they are supposed to provide. They could construct those
strains of bacteria. They could even construct GM seeds for the purpose of studying
them further. In these respects, an experimental use exemption would be highly
beneficial. However, there is one important thing that researchers could not do, even
under such an exemption; they could not acquire GM seeds from seed companies for
the purpose of research. To acquire GM seeds from seed companies, researchers still
must sign license agreements, and as the owners of GM seeds, seed companies can
write virtually anything into license agreements that they want. Even in a legal system
with a general statutory research exemption, a seed company is free to write prohibitions
on research into their license agreements; if one obtains GM seeds from a company that
has done this, and if one does research on these seeds, then one has broken the law.
One might not be guilty of patent infringement, but one is guilty of breach of contract.
License agreements that are currently standard in the agricultural biotechnology industry
prohibit licensees from performing research, and they prohibit licensees from transferring
seeds to others for the purpose of research. Under these agreements, and assuming
that a research exemption is in place, the only way for a researcher to perform research
on GM seeds legally is to construct those seeds herself (which, again, is legal given the
research exemption).

15
This discussion is informed by Walsh, Cohen, and Cho (2007), which discusses in a careful
manner the differences in excludability brought by patents and by contracts (specifically material
transfer agreements).

16
 I have argued that an adequately formulated experimental use exemption would,
in principle, allow researchers to perform studies on patent-protected GM seeds. This
would represent an improvement over the current situation in the US. At the same time
such an exemption would, in practice, have a limited effect on research. The practical
barriers to research would be very high if scientists were forced to replicate GM seeds
before doing research on them. It would be far easier for scientists simply to acquire
those seeds from the companies that produce them, and yet this would not be possible
given current license agreements, even with a research exemption in place. In this
situation, the reason why scientists could not acquire GM seeds from patent holders has
nothing to do with patents, and it has everything to do with the contracts that they must
sign in order to gain access to the patented inventions.
Both voluntary agreements and a research exemption would improve the ability
of scientists to access patent-protected materials, but neither of them, either individually
or jointly, are sufficient to ensure that scientists can, in practice, perform the variety of
studies that are needed. For this, broader organizational changes are required. While a
discussion of these broader changes is beyond the scope of this paper, two important
ones concern the funding and institutional location of research. Studies that are
undertaken by university scientists and that are funded by the government can, of
course, lead to patents, but the licenses required to access these patented inventions
will be unlikely to include prohibitions on research. The problem of ensuring better
access to patent-protected materials will need to take into account such broader features
of the organization of research.

8. Conclusion. This paper has examined one important aspect of the organization of
scientific and technological research – namely, the current system of patenting and
licensing and its role in structuring the production and dissemination of knowledge. I
have argued that in some areas of research in the US – in particular, research on GM
seeds – patents and licenses can be, and are in fact being, used to prohibit some
research. This is a serious epistemic deficiency of the current organization of research,
one that philosophers of science, along with others in law and policy, should emphasize
and attempt to remedy. I have discussed three potential solutions to this problem:
voluntary agreements, eliminating patents, and a research exemption. I argued that
eliminating patents is, at best, premature, and I showed that while voluntary agreements
and a research exemption could be helpful, they do not fully address the problems of
access raised in this paper. If we are to address these problems adequately, more
extensive changes in the organization of research will be necessary.
The method that I have employed in this paper is a piecemeal, iterative, and
empirically-based method that is similar to the strategy of ‘adaptive management,’ which
has been defended by Mitchell (2009) in the context of policy-relevant research and by
Reiss (2010) in the context of biomedical research. This, in my view, is a promising
method for social epistemology, and one of the aims of this paper has been to illustrate
how it can be applied. One of the implications of this method is that the steps that I have
recommended – voluntary agreements and a research exemption – should be taken as

17
provisional and as one piece of a much larger project of reorganizing scientific and
technological research.

18
References

American Association for the Advancement of Science (2007) International Intellectual

Property Experiences - A Report of Four Countries, Washington, D.C.: Project on
Science and Intellectual Property in the Public Interest. Available:
http://sippi.aaas.org/Pubs/SIPPI_Four_Country_Report.pdf. Accessed 19 July 2012.

Bekelman, J. E., Y. Li, and C. P. Gross. 2003. “Scope and impact of financial conflicts of
interest in biomedical research: a systematic review.” Journal of the American Medical
Association 289: 454-465.

Biddle, Justin B. 2007. “Lessons from the Vioxx Debacle: What the Privatization of
Science Can Teach Us about Social Epistemology.” Social Epistemology 21(1): 21-39.

——— (2011). “Bringing the Marketplace into Science: On the Neoliberal Defense of the
Commercialization of Scientific Research.” In Science in the Context of Application, M.
Carrier and A. Nordmann, eds. Dordrecht: Springer, 245-269.

–––––. 2012. “Tragedy of the Anticommons? Intellectual Property and the Sharing of
Scientific Information. Philosophy of Science 79: 821-832.

Biddle, J. (forthcoming). “Intellectual Property in Biomedical Research.” In Routledge

Companion to Bioethics, R. Kukla, J. Arras, and E. Fenton, eds. London: Routledge.

Brody, B. (2006) “Intellectual Property and Biotechnology: The U.S. Internal Experience
– Part I,” Kennedy Institute of Ethics Journal 16: 1-37.

——— (2006a) “Intellectual Property and Biotechnology: The U.S. Internal Experience –
Part II,” Kennedy Institute of Ethics Journal 16: 105-128.

Brown, James R. 2008. “Politics, Method, and Medical Research.” Philosophy of

Science 75: 756-766.

——— 2008a. “The Community of Science®,” in M. Carrier, D. A. Howard, and J.

Kourany (eds), The Challenge of the Social and the Pressure of Practice, Pittsburgh:
University of Pittsburgh Press, pp. 189-216.

Brown, Mark. 2004. “The Political Philosophy of Science Policy.” Minerva 42: 77-95.

Centre for Intellectual Property Policy & Health Law Institute. “The Research or
Experimental Use Exception.” http://www.cipp.mcgill.ca/data/newsletters/00000050.pdf
(Accessed August 18, 2013.)

Dewey, John. 1938. Logic: The Theory of Inquiry. Holt, Rinehart, and Winston, New
York.

Eisenberg, Rebecca. 2008. “Noncompliance, Nonenforcement, Nonproblem? Rethinking

the Anticommons in Biomedical Research.” Houston Law Review 45: 1060-1099.

Elliott, Kevin. 2013. “Selective Ignorance and Agricultural Research.” Science,

Technology, and Human Values 38: 328-350.

19
Goldman, Alvin. 1999. Knowledge in a Social World. New York: Oxford University Press.

Heller, Michael and Rebecca Eisenberg. 1998. “Can Patents Deter Innovation? The
Anticommons in Biomedical Research?” Science 280: 698-701.

Hughes, Justin (1988) “The Philosophy of Intellectual Property,” Georgetown Law

Journal 77: 287-366.

Kitcher, Philip (1993). The Advancement of Science. New York: Oxford University Press.

–––––– (2001). Science, Truth, and Democracy. New York: Oxford University Press.

–––––– (2011). Science in a Democratic Society. Amherst, New York: Prometheus

Books.

Klinische Versuche Clinical Trails I (1995). Federal Supreme Court of Germany.

Klinische Versuche Clinical Trails II (1998). Federal Supreme Court of Germany.

Krimsky, Sheldon (2003). Science in the Private Interest: Has the Lure of Profits
Corrupted Biomedical Research? Lanham, MD: Rowman & Littlefield.

Kukla, Rebecca (2012). “’Author TBD’: Radical Collaboration in Contemporary

Biomedical Research.” Philosophy of Science 79: 845-858.

Lexchin, J., L. A. Bero, B. Djulbegovic, and O. Clark (2003). “Pharmaceutical Industry

Sponsorship and Research Outcome and Quality: Systematic Review.” BMJ 326: 1167-
1170.

Longino, Helen. (1990). Science as Social Knowledge. Princeton: Princeton University

Press.

–––––– (2002). The Fate of Knowledge. Princeton: Princeton University Press.

Madey v. Duke University (2002). No. 01-1567 (Court of Appeals for the Federal Circuit).
http://cyber.law.harvard.edu/people/tfisher/2002Madeyedit.html (Accessed August 18,
2013).

Mirowski, Philip and Esther-Mirjam Sent, eds. (2002). Science Bought and Sold: Essays
in the Economics of Science. Chicago: University of Chicago Press.

Mirowski, Philip and Robert van Horn (2005). “The Contract Research Organization and
the Commercialization of Scientific Research.” Social Studies of Science 35: 503-548.

Mitchell, S. (2009) Unsimple Truths: Science, Complexity, and Policy, Chicago:

University of Chicago Press.

20
Monsanto Company. “Academic Research Agreements.”
http://www.monsanto.com/newsviews/Pages/public-research-agreements.aspx
(Accessed July 20, 2013).

Monsanto Company. 2008. “2008 Monsanto Technology/Stewardship Agreement.”

Accessed November 16, 2012.
http://www.monsanto.com/SiteCollectionDocuments/tug_sample.pdf

Monsanto v. Bowman (2011). No. 07-CV-0283 (Court of Appeals for the Federal Circuit).
http://www.cafc.uscourts.gov/images/stories/opinions-orders/10-1068.pdf (Accessed
August 18, 2013).

Oreskes, Naomi and Erik Conway (2010). Merchants of Doubt. New York: Bloomsbury
Press.

Proctor, Robert and Londa Schiebinger (2008). Agnotology: The Making & Unmaking of
Ignorance. Stanford: Stanford University Press.

Quine, W. V. O. and J. S. Ullian (1970). The Web of Belief. New York: Random House.

Reiss, J. (2010) “In Favor of a Millian Proposal to Reform Biomedical Research,”

Synthese 177: 427-447.

Shiva, Vandana (2000). Stolen Harvest. Cambridge, MA: South End Press.

Slaughter, Sheila, and Gary Rhoades (1996). “The Emergence of a Competitiveness

Research and Development Policy Coalition and the Commercialization of Academic
Science and Technology.” Science, Technology, and Human Values 21: 303-339.

Walsh, J., Cohen, W., and Cho, C. (2007) “Where Excludability Matters,” Research
Policy 36: 1184-1203.

Walsh, John, Wesley Cohen and Ashish Arora. 2003. “Patenting and Licensing of
Research Tools and Biomedical Innovation.” In Patents in the Knowledge Based
Economy, ed. W. M. Cohen and S. Merrill, 285-340. Washington, DC: NAP.

Walz, Emily. 2009. “Under Wraps.” Nature Biotechnology 27: 880-882.

______. 2010. “Monsanto Relaxes Restrictions on Sharing Seeds for Research.” Nature
Biotechnology 28: 996.

Wilholt, Torsten. 2009. “Bias and Values in Scientific Research.” Studies in History and
Philosophy of Science 40: 92-101.

21