Which Science? Whose Science? How Scientific Disciplines Can Shape Environmental Law

[أحمد أبو زنط]

The University of Chicago
Law Review
Volume 79 Spring 2012 Number 2
© 2012 by The University of Chicago

ARTICLES


Which Science? Whose Science? How Scientific
Disciplines Can Shape Environmental Law

Eric Biber




Environmental law fundamentally depends on the production of information by
environmental science. However, the inevitable uncertainty in environmental science
means that policy will necessarily intertwine with science, allowing for the concealment of
value choices and trade-offs by agencies, interest groups, and politicians under a patina of
science. Scholars and decision makers have struggled over how to resolve this “science
charade.” This Article proposes a new approach to the underlying challenge by closely
examining the diversity of scientific disciplines in environmental law. Scientific disciplines
each develop their own perspectives that are based in part on values and that shape the
information they produce. Understood this way, the intersection of law and science can be
greatly simplified. Instead of attempting to separate science and policy out for every
significant individual decision, we can make generalizations about how science and
policy will interact depending on the discipline that produces the relevant information. We
can also understand scientific disciplines as essential components to the designing of any
regulatory or management system. For instance, particular disciplines might be privileged
in the legal or institutional structure to help advance specific policy goals; we might
† Assistant Professor of Law, University of California at Berkeley School of Law;
Visiting Assistant Professor of Law, University of Chicago Law School, Fall 2011.
Thanks to Ty Alper, Michelle Wilde Anderson, Robert Bartlett, Holly Doremus, Dan
Farber, Prasad Krishnamurthy, Brian Leiter, Katerina Linos, Prasad Krishnamurthy, Anup
Malani, Emily Hammond Meazell, Martha Nussbaum, Dave Owen, Eric Posner, Bertrall Ross,
Adam Samaha, Joseph Sax, Eleanor Swift, David Takacs, David Weisbach, David Winickoff,
and Katrina Wyman, and participants at workshops at UC Berkeley School of Law, the
University of Chicago Law School, the University of Maine School of Law, the Law and
Society Association 2011 Annual Meeting, and the Colloquium on Environmental Scholarship
at Vermont Law School for helpful comments. Thanks to Santosh Sagar, Jill Jaffe, Jennifer
Aengst, Zachary Markarian, and Jessica Cheng for research assistance.
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
472 The University of Chicago Law Review [79:471
insulate the decision-making process from disciplines that interfere with the
accomplishment of those goals; or, we might balance multiple disciplines in order to
reduce the risk of disciplinary blind spots that interfere with policy making.
INTRODUCTION .................................................. .................................................. ............... 472
I. SCIENCE AND ENVIRONMENTAL LAW: A PROBLEMATIC, BUT INEVITABLE,
PARTNERSHIP....................................... .................................................. ..................... 476
II. SCIENTIFIC DISCIPLINES .................................................. .......................................... 487
A. The State of Global Fisheries: An Example of How Disciplines
Matter in Environmental Science .................................................. ................. 488
B. Defining Scientific Disciplines .................................................. ....................... 493
C. How Disciplinary Structures Are Created and Shaped ............................... 495
D. How Disciplinary Structures Shape the Production of Scientific
Information .................................................. .................................................. .... 501
1. The importance of disciplinary perspectives ......................................... 501
2. How disciplines can shape the nature and content of their own
perspectives .................................................. .............................................. 506
3. How outside institutions can shape disciplinary perspectives ............ 510
III. EXPLORING THE ROLE OF DISCIPLINES IN SHAPING ENVIRONMENTAL LAW .... 512
A. Using Scientific Disciplines as Legal- or Institutional-Design Tools ......... 513
1. Option one: constraining .................................................. ........................ 514
2. Option two: legitimation .................................................. ........................ 521
3. Option three: insulation .................................................. ......................... 525
4. Option four: balancing .................................................. ............................ 528
5. Limits to using disciplines as institutional- or legal- design tools ....... 532
6. Assessing the use of disciplines as legal- and institutionaldesign
tools .................................................. ............................................... 544
B. Seeing Science and Law in a New Way .................................................. ........ 545
CONCLUSION .................................................. .................................................. ................... 550
INTRODUCTION
The debate over climate change is, in part, about whether and
how our planet has been growing warmer over the past century and a
half of global industrial development. One group of scientists—
generally called “climate scientists”—has developed complicated
computer models that attempt to predict how the global climate will
respond over the next ten to five-hundred years to a range of
different human actions, such as increased concentrations of carbon
dioxide, changes in deforestation rates, and increasing urbanization.
These scientists have been the leaders of the international
organizations that have sought to summarize and present the
relevant science to policy makers through bodies such as the United
Nations’ Intergovernmental Panel on Climate Change (IPCC).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 473

But there is another group of scientists (or at least, professionals
who rely heavily on science) who often speak about matters related
to climate, albeit on a very different scale: meteorologists, who try to
forecast weather conditions in particular geographic locations in
relatively short time frames, such as days to months. In the United
States, meteorologists have a tremendous amount of public influence
because they host many of the television weather forecasts.1 Yet
meteorologists are far more skeptical of climate change than climate
scientists, and far more likely to believe that humans are not causing
any climate change that is occurring, perhaps because of their own
travails in dealing with the problems of forecasting weather in short
time periods.2 A recent study found that over one-quarter of
meteorologists believe climate change is a “scam,” and there are a
number of high-profile meteorologists who have aggressively
questioned climate change in public.3
The divide between climate scientists and meteorologists is an
example of a much broader pattern in environmental law and policy:
different scientific disciplines have very different perspectives and
often reach very different conclusions about the state of the world
and the need for policy intervention based on similar or identical
information. While environmental law scholars have touched upon
these issues, this Article builds on those initial insights to carefully
examine the implications of the many disciplines and perspectives
within environmental science.
There are substantial benefits to taking into account the
diversity of disciplines and perspectives in environmental science.
One of the intractable problems in environmental law has been the
troubled boundary between science and policy. Scholars have
criticized agencies, interest groups, and scientists for pursuing a
science “charade” in which policy conclusions and value choices are
hidden in complicated, technical models and analyses, primarily via
1 See Leslie Kaufman, Consensus on Climate Change Ends at the TV Weather Desk, NY
Times A1 (Mar 30, 2010) (noting a study that found that 56 percent of Americans trusted
weathercasters to inform them about climate change).
2 See Bill Dawson, Why Are So Many TV Meterologists and Weathercasters Climate
“Skeptics”?, The Yale Forum on Climate Change & the Media (June 12, 2008), online at
http://www.yaleclimatemediaforum.org/2008/06/why-are-so-many-tv-meteorologists-andweathercasters-
climate-skeptics/ (visited Dec 15, 2011).
3 See Kaufman, Consensus on Climate Change, NY Times at A1 (cited in note 1). In
contrast, climatologists have developed a stong consensus that climate change is real, with
90 percent believing that climate change is real, and more than 80 percent agreeing that human
activity is a significant contributing factor. See id, citing Peter T. Doran and Maggie Kendall
Zimmerman, Examining the Scientific Consensus on Climate Change, 90 Eos AGU 20, 22 (2009).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
474 The University of Chicago Law Review [79:471
assumptions and inferences.4 The frequent response has been a call
for greater transparency through a more explicit discussion of the
policy conclusions and value choices necessarily part of regulatory
decision making.5 But given the complexity and dynamism of natural
systems and therefore the large number of assumptions and
simplifications necessary for tractable and informative analyses, it
may be impossible (or at least extremely difficult) to ever fully
identify all of the policy conclusions and value choices in regulatory
science.
An alternative approach is to take into account the diverse
range of values and perspectives already embedded in the scientific
disciplines that are relevant for environmental law. In doing so, we
can see that disciplines in fact often come as a “package” combining
a range of perspectives that may be both methodological and value
based. For instance, conservation biologists believe that biodiversity
protection is an essential social goal, and the information they
produce is shaped accordingly. It will often be much more feasible to
identify the relevant discipline, its perspectives and values, and the
likely way in which those perspectives and values might shape the
ultimate conclusions of the relevant scientists, than it will be to
painstakingly identify every assumption in every study and the policy
implications of each assumption.
But there are larger benefits than simply making it easier to
understand how policy and science intersect in environmental law.
Once we understand scientific disciplines as bringing perspectives to
the table that are relevant to policy making, we can also see them as
legal- and institutional-design tools.6 If we wish to accomplish
particular policy goals in our regulatory- or management-design
structures, then we may choose to embrace a particular discipline
because its perspective will help advance those goals. Alternatively,
we might seek to insulate the regulatory- or management-decision
makers from a discipline that we believe will undermine those policy
goals. Scientific disciplines may have particular promise as a precom-
4 See, for example, Ian Fein, Comment, Reassessing the Role of the National Research
Council: Peer Review, Political Tool, or Science Court?, 99 Cal L Rev 465, 471–73 (2011). See
also Alvin M. Weinberg, Science and Trans-science, 10 Minerva 209, 209 (1972).
5 See Fein, Comment, 99 Cal L Rev at 474 (cited in note 4).
6 For an overview of major issues in the design of administrative agencies, see generally
Jacob E. Gersen, Designing Agencies, in Daniel A. Farber and Anne Joseph O’Connell, eds,
Research Handbook on Public Choice and Public Law 333 (Edward Elgar 2010). One
“designer” might be the legislature that establishes a statutory structure for regulatory or
management decision making, though other parties (the President, the head of an agency)
might play similar roles through other institutional mechanisms, such as the issuance of agency
regulations.
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 475

mitment device by which environmental policy implementation can
be protected against slippage as a result of pressure by powerful
special interests or political myopia. Indeed, particular legal and
administrative choices—such as statutory mandates that
environmental agencies use the “best available science” in making
regulatory decision making, or the embrace of particular disciplines
by management agencies in terms of staffing and research—make
more sense when understood in this way.
Understanding the importance of scientific disciplinary perspectives
for environmental law can also lead us to consciously rely on a
diversity of disciplinary perspectives when we are uncertain what our
policy goal is. That diversity of perspectives might reduce the risk of
“blind spots” in policy making that could cause serious political,
technical, or economic problems in the course of policy
implementation.
The exact choices regarding whether and how to use disciplines
in these ways will depend on the particular goals that policy makers
have for their regulatory or management programs. Given that
contingency, my goal in this Article is not primarily to provide a
normative assessment of the roles that scientific disciplines play in
environmental science, or of the use of those disciplines in policy
making. Instead, it is to provide a description of the ways in which
disciplines do shape environmental science, and therefore
environmental law today, and explore the implications of that fact
for environmental policy making.
Thinking about scientific disciplines as including a package of
science and values does not mean, however, that we give up hope on
the ability of science to gain useful information from the natural
world or the ability of the natural world to constrain the information
that science produces. Nor does it mean that we give up on the ideal
and goal of science as an endeavor that seeks to objectively
determine the state of the natural world. Instead, building on the
concept of “strong objectivity” developed by Sandra Harding,7 it
simply recognizes the limited but inevitable contribution of social
factors to the development of science and information, and the
limited but inevitable ways in which the perspectives of scientists and
others shape the production of information from the natural world.
In Part I, I develop the basic interaction of science and policy in
environmental law, the problems that environmental law has faced in
7 Sandra Harding, Whose Science? Whose Knowledge? Thinking from Women’s Lives
138–63 (Cornell 1991); Sandra Harding, After the Neutrality Ideal: Science, Politics, and “Strong
Objectivity,” 59 Soc Rsrch 567, 569–75 (1992).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
476 The University of Chicago Law Review [79:471
balancing the two, and the weaknesses in existing proposals to
“separate” science and policy. In Part II, I draw on science and
technology studies literature to develop the concept of scientific
disciplines and identify the primary factors that shape disciplines and
their perspectives. I also present examples of the importance of
values and perspectives in environmental science, focusing on
fisheries science, marine ecology, conservation biology, and wildlife
management. In Part III, I develop two main implications of the
understanding of the diversity of scientific disciplinary perspectives
in environmental law: the possible use of scientific disciplines as
legal- or institutional-design tools, and the simplicity of relying on
disciplinary perspectives as a proxy for understanding how science
and policy interact in environmental law. I conclude by developing
the implications of my analysis for other areas of law, such as
forensic science in criminal law.

[أحمد أبو زنط]

I. SCIENCE AND ENVIRONMENTAL LAW: A PROBLEMATIC, BUT
INEVITABLE, PARTNERSHIP
Science is an essential input for environmental decision making.
Regulatory standards for pollution depend on a number of
questions: How much harm pollution will impose on human or
natural systems? How effective are pollution-reduction systems?
How much pollution actually is present in the natural environment?8
Management decisions for natural resources depend on questions as
to the status of the natural resources that are to be exploited or
protected, the potential impacts on those resources from various
management options, and the dynamic interaction of resources in
response to human management decisions and changes in natural
systems.9 Answers to all of these questions will in turn depend on
information produced by fields as diverse as epidemiology,
toxicology, ecology, hydrology, and chemistry.
8 See, for example, 42 USC § 7408(a)(1)(A) (requiring regulation to control air
pollutants that “may reasonably be anticipated to endanger public health or welfare”); 33 USC
§ 1316(a)(1), (b)(1)(B) (requiring regulation to control water pollutants “through application
of the best available demonstrated control technology”); 33 USC § 1312(a) (requiring
regulation to control pollutants where necessary to “assure protection of public health, public
water supplies, agricultural and industrial uses, and the protection and propagation of a
balanced population of shellfish, fish and wildlife” in waterways).
9 See, for example, 43 USC § 1732(b) (requiring the federal land management agencies
to “take any action necessary to prevent unnecessary or undue degradation” of public lands);
16 USC § 1604(g)(3)(B) (requiring the Forest Service to issue regulations that will “provide for
diversity of plant and animal communities” on national forests).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 477

Environmental law often explicitly requires agencies to draw
upon science. The Endangered Species Act of 197310 (ESA) and the
Marine Mammal Protection Act of 197211 (MMPA) require
administrative agencies to rely upon the “best available science.”12
The Clean Air Act13 (CAA) requires the EPA to draw upon science
advisory boards when making regulatory decisions.14
While science is essential to environmental law and policy,
science often cannot provide clear answers to the questions that
environmental decision makers ask.15 This is in large part because of
the significant uncertainties that surround environmental science.16
Because of the complexity of the interaction of the large number of
variables in environmental systems, the high rates of dynamic change
in those variables, and the multiple levels of temporal and
geographic scale at which change can happen, it is quite difficult to
get high-quality information about environmental systems.17 Simple
measurement error and variability in the environment can lead to
wildly different outcomes in the information produced from studies
of environmental systems.18 Statistical analysis of calculations
estimating how much global temperatures will respond to changes in
carbon dioxide concentrations in the atmosphere (“climate
10 Pub L No 93-205, 87 Stat 884, codified at 16 USC §§ 1531–43.
11 Pub L No 92-522, 86 Stat 1027, codified at 16 USC §§ 1361–1423.
12 See 16 USC § 1533(b)(1)(A); 16 USC § 1536(a)(2); 16 USC § 1373(a).
13 Pub L No 88-206, 77 Stat 392 (1963), codified as amended at 42 USC § 7401 et seq.
14 Clean Air Act § 1, 42 USC § 7417.
15 See, for example, Holly Doremus, The Purposes, Effects, and Future of the Endangered
Species Act’s Best Available Science Mandate, 34 Envir L 397, 438 (2004) (noting the problem
in the context of the ESA).
16 See, for example, C.S. Holling, Two Cultures of Ecology, 2 Ecol & Socy (1998), online
at http://www.ecologyandsociety.org/vol2/iss2/art4/ (visited Dec 15, 2011) (“In principle, . . .
there is an inherent unknowability, as well as unpredictability, concerning ecosystems and the
societies with which they are linked.”); Holly Doremus and A. Dan Tarlock, Science,
Judgment, and Controversy in Natural Resource Regulation, 26 Pub Land & Res L Rev 1, 6
(2005) (“The hard reality is that the scientific information available to support environmental
and natural resource policy decisions is frequently incomplete, ambiguous, and contested.”).
17 See, for example, Eric Biber, The Problem of Environmental Monitoring, 83 U Colo L
Rev *3 (forthcoming 2012), online at http://ssrn.com/abstract=1680000 (visited Dec 15, 2011);
Silvio O. Funtowicz and Jerome R. Ravetz, Three Types of Risk Assessment and the Emergence
of Post-normal Science, in Sheldon Krimsky and Dominic Golding, eds, Social Theories of
Risk 251, 253 (Praeger 1992); Naomi Oreskes, Evaluation (Not Validation) of Quantitative
Models, 106 Envir Health Persp 1453, 1453 (1998) (noting that inherent uncertainty in natural
systems makes it impossible to “demonstrate the predictive reliability of any model of a
complex natural system in advance of its actual use”).
18 Reinette Biggs, Stephen R. Carpenter, and William A. Brock, Spurious Certainty:
How Ignoring Measurement Error and Environmental Heterogeneity May Contribute to
Environmental Controversies, 59 BioSci 65, 69–73 (2009).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
478 The University of Chicago Law Review [79:471
sensitivity”) shows that reduction of the uncertainty range for those
estimates is difficult or impossible, even with improved data inputs.19
There is therefore tremendous room, even need, for
assumptions, inferences, and interpretations in the process of
converting incomplete environmental data into conclusions on which
environmental law and policy decisions can be based.20 Those
assumptions, inferences, and interpretations21 will in turn depend on
a range of value choices and positions, explicitly or implicitly.22 Even
when scientists can develop some consensus on which factors might
be used to address the problems of underlying uncertainty,23 there is
little consensus on how to prioritize those factors, retaining ample
space for the use of subjective preferences and value choices to affect
the judgments drawn from ambiguous data.24
A simple but real example of this problem comes from the field
of risk assessment, in which scientists and regulators attempt to
determine whether and to what extent chemicals pose a health risk
(particularly the risk of cancer) to humans. A common form of
conducting risk assessment is to expose lab animals to high doses of
the chemical over a relatively short period of time and then to
observe whether any adverse health effects occur.25 The results from
high-dose testing on lab animals must then be extrapolated to the
19 Gerard H. Roe and Marcia B. Baker, Why Is Climate Sensitivity So Unpredictable?,
318 Sci 629, 629 (2007).
20 See, for example, Biber, 83 U Colo L Rev at *46–47 (cited in note 17); Steven Yearley,
The Green Case: A Sociology of Environmental Issues, Arguments and Politics 119–32
(HarperCollins 1991).
21 Doremus and Tarlock, 26 Pub Land & Res L Rev at 9 (cited in note 16) (calling these
the exercise of “judgment”). For other examples of the concept, see Elizabeth Fisher,
Drowning by Numbers: Standard Setting in Risk Regulation and the Pursuit of Accountable
Public Administration, 20 Oxford J Legal Stud 109, 110, 116–17 (2000); Steven Yearley, Bog
Standards: Science and Conservation at a Public Inquiry, 19 Soc Stud Sci 421, 432–35 (1989).
22 See Holly Doremus, Scientific and Political Integrity in Environmental Policy, 86 Tex L
Rev 1601, 1624–29 (2008) (noting that “scientific judgments and value judgments are often
closely intertwined”); Holly Doremus, Science Plays Defense: Natural Resources Management
in the Bush Administration, 32 Ecol L Q 249, 253 (2005); Sheila Jasanoff, The Fifth Branch:
Science Advisers as Policymakers 7, 16–17, 230–31 (Harvard 1990).
23 See, for example, Douglas L. Weed, Underdetermination and Incommensurability in
Contemporary Epidemiology, 7 Kennedy Inst Ethics J 107, 112–17 (1997) (listing factors used
in epidemiology to evaluate whether ambiguous data has supported or refuted particular
hypotheses).
24 See id at 117–19 (showing how different epidemiological review studies that relied on
essentially identical evidence reached contrary outcomes, in large part because of different
weighting of factors and noting possibility of “wish bias” in which “the results of the individual
reviewers’ studies match the conclusions reached in their respective reviews”).
25 High doses are used in order to reduce the time and expense of the studies; ethical
constraints prevent this sort of experimentation with humans. See Mark E. Rushefsky, Making
Cancer Policy 21–58 (SUNY 1986).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 479

impacts on humans from more realistic low-dose, chronic exposures.
A key inference in this extrapolation is the nature of the “doseresponse”
curve, or how responses to the chemical in humans change
as the dose levels decrease. The relationship might be linear, in
which case if 1 gram of exposure causes a 1-in-100 risk of cancer,
0.01 grams of exposure should cause a 1-in-10,000 risk of cancer; or,
the relationship might be non-linear, such that if 1 gram of exposure
causes a 1-in-100 risk of cancer, 0.01 grams of exposure causes a 1-in-
1,000,000 risk of cancer. Both curves are plausible given what we
know about biology and chemistry.26 But the different assumptions
result in dramatic policy differences; the first relationship produces a
much higher potential risk of cancer from low-dose exposures, and
therefore a much stronger basis for regulatory intervention.27 And of
course, the underlying values and policy positions of the relevant risk
assessors may, consciously or not, shape their choices among these
various options.28
This inevitable intertangling of science and value judgments in
environmental law (what some scholars have called “trans-science”
problems29) leads to a series of problems. On the one hand, the
public, scientists, and policy makers often fail to understand the
importance of the value choices and preferences hidden in the
assumptions, inferences, and interpretations needed to translate
incomplete science into policy decisions. Instead, they assume a
“linear” model of science and policy in which science determines
facts that in turn determine policy options.30 Each side identifies the
information that best supports their position and accuses the other
side of “bad science” by highlighting and undermining the key
assumptions and uncertainties in the other side’s chosen
information.31 Science becomes “a proxy for political battle,” as each
side tries to get the facts right so they can win the debate.32
26 See id at 48–49.
27 See Wendy E. Wagner, The Science Charade in Toxic Risk Regulation, 95 Colum L
Rev 1613, 1626 (1995).
28 See Rushefsky, Making Cancer Policy at 22–27 (cited in note 25).
29 “[Q]uestions which can be asked of science and yet which cannot be answered by
science.” Weinberg, 10 Minerva at 209 (cited in note 5).
30 Roger A. Pielke Jr, When Scientists Politicize Science: Making Sense of Controversy
over The Skeptical Environmentalist, 7 Envir Sci & Pol 405, 406 (2004).
31 Sheila Jasanoff, (No?) Accounting for Expertise, 30 Sci & Pub Pol 157, 158–60 (2003)
(“[T]hose wishing to question a given scientific interpretation can generally find errors, hidden
biases or subjective judgments that undercut their opponents’ claims to truth and objectivity.”).
32 Pielke, 7 Envir Sci & Pol at 412 (cited in note 30).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
480 The University of Chicago Law Review [79:471
Of course, savvy policy makers and political actors sooner or
later understand this dynamic.33 If they are outside an administrative
agency, they make take advantage of the inevitable uncertainty to
raise doubts about the scientific justification for a particular
government action or inaction, hiding their value choices in a
critique of “bad agency science.”34 If they are decision makers within
the agency, then they may take advantage of the uncertainty to hide
value choices by contending that “science” mandates the outcome
reached by the agency, without fully disclosing the assumptions,
interpretations, and inferences that underpin the agency’s decision
and how they might advance particular political positions.35 For
instance, agencies may construct models to support decision making
that depend on important, but unstated, assumptions for their
functioning.36 In this dynamic, which Wendy Wagner called the
“science charade,”37 in the context of toxic chemical regulation,
“[s]cience . . . becomes a convenient and necessary means for
removing certain options from a debate without explicitly dealing
with disputes over values.”38
One possible solution is for “better science” to reduce the
uncertainty and the need for assumptions, inferences, and
interpretations, and therefore to let the natural world resolve the
political dispute.39 Unfortunately, modern science often cannot
successfully reduce uncertainty in a way that will fully resolve
33 Michael S. Carolan, The Politics in Environmental Science: The Endangered Species
Act and the Preble’s Mouse Controversy, 17 Envir Polit 449, 449 (2008) (“Unfortunately, the
banner of objective science is often waved by politicians to mask some very subjective beliefs
and assumptions.”).
34 See notes 268–71 and accompanying text.
35 Cary Coglianese and Gary E. Marchant, Shifting Sands: The Limits of Science in
Setting Risk Standards, 152 U Pa L Rev 1255, 1262–73 (2004); Thomas O. McGarity,
Substantive and Procedural Discretion in Administrative Resolution of Science Policy
Questions: Regulating Carcinogens in EPA and OSHA, 67 Georgetown L J 729, 782 (1979)
(proposing that “a regulator’s open acknowledgement that result-oriented policy
considerations will guide his resolution of science policy issues will simply ensure that the
regulator will make the subjective decision instead of a scientist or low-level bureaucrat”).
36 See, for example, Biber, 83 U Colo L Rev at *47 (cited in note 17); Orrin H. Pilkey
and Linda Pilkey-Jarvis, Useless Arithmetic: Why Environmental Scientists Can’t Predict the
Future xii-xiv, 10, 20–21, 115–16 (Columbia 2007).
37 Wagner, 95 Colum L Rev at 1617 (cited in note 27).
38 Pielke, 7 Envir Sci & Pol at 409 (cited in note 30).
39 See, for example, Simon Shackley and Brian Wynne, Representing Uncertainty in
Global Climate Change Science and Policy: Boundary-Ordering Devices and Authority, 21 Sci,
Tech, & Hum Values 275, 282 (1996) (describing how “[a]ppeals and pledges to reduce
uncertainties . . . are common” in climate science).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 481

environmental disputes.40 An underlying challenge in science is the
underdetermination of theories from data: “There will always be
more than one theory supported by the evidence, because for any
piece of recalcitrant data, we always have two options: abandoning
the . . . hypothesis being tested, or preserving the hypothesis, but
abandoning” the claim that the test was informative as to the truth
of the hypothesis.41 As a result, decisions as to the implications of
experimental results or investigations into the natural world will
require “extraevidential considerations like symmetry, simplicity,
and elegance, or personal, political, or metaphysical preferences.”42
In the context of environmental science, this means that political
actors can often develop new arguments to “explain away”
inconvenient new factual data through claims about, for example, the
exceptional circumstances of the time period in which the data were
collected, alternative causal mechanisms, claims as to bias or error in
the collection of data, and so forth. This allows the maintenance of
the original claim as to the state of the world despite the new
information, a claim that also (conveniently enough) supports that
party’s political position. The problem is particularly troublesome in
the context of environmental decision making because the
complexity, dynamism, and diversity of relevant scales in
environmental science ensure that data will be especially incomplete
and therefore that theories will be especially underdetermined.43
40 See Dorothy Nelkin, Science, Technology, and Political Conflict: Analyzing the Issues, in
Dorothy Nelkin, ed, Controversy: Politics of Technical Decisions 9, 11–14 (SAGE 1979); Sheila
Jasanoff, The Dilemma of Environmental Democracy, 13 Issues Sci & Tech 63, 69 (1996).
41 Brian Leiter, Why Quine Is Not a Postmodernist, 50 SMU L Rev 1739, 1747 (1997)
(describing the “so-called ‘Duhem-Quine’ thesis”). See also Naomi Oreskes, Kristin Shrader-
Frechette, and Kenneth Belitz, Verification, Validation, and Confirmation of Numerical Models
in the Earth Sciences, 263 Sci 641, 642 (1994); H.M. Collins and T.J. Pinch, Frames of Meaning:
The Social Construction of Extraordinary Science 177–85 (Routledge & Kegan Paul 1982);
Andrew Pickering, Constraints on Controversy: The Case of the Magnetic Monopole, 11 Soc
Stud Sci 63, 66 (1981) (noting an “infinite number of potential interpretations” of experimental
results). Underdetermination does not mean that useful scientific information cannot be
produced based on observations and experiments. See Larry Laudan, Science and Relativism:
Some Key Controversies in the Philosophy of Science 48–68, 76–86 (Chicago 1990) (discussing
the problem of underdetermination and noting that the concept in some form is widely
accepted in the philosophy of science but that it does not necessarily preclude the exercise of
judgment in selecting among competing theories nor does it inevitably foreclose the
elimination of hypotheses as a result of experiment in all circumstances). See also Leiter,
50 SMU L Rev at 1749–50 (cited in note 41) (rejecting the conclusion that the Duhem-Quine
thesis forecloses the possibility that science can produce reliable information based on the
natural world).
42 Oreskes, Shrader-Frechette, and Belitz, 263 Sci at 642 (cited in note 41); Pickering,
11 Soc Stud Sci at 65–66 (cited in note 41).
43 See Roger A. Pielke Jr, The Honest Broker: Making Sense of Science in Policy and
Politics 67–70 (Cambridge 2007).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
482 The University of Chicago Law Review [79:471
Indeed, given the high stakes around science in environmental
decision making, and the problem of underdetermination, increasing
the role of science in environmental decision making can make
political conflicts worse, not better.44 As the political stakes over an
environmental decision and the supporting science rise, there are
more incentives for those within and outside the scientific
community to closely investigate the research, discover the inevitable
ambiguities and uncertainties, and emphasize how the policy
conclusions are underdetermined by the facts.45 Thus, the
unproblematic use of science in decision making may be easiest
where there is consensus about the relevant policy goals.46
An alternative is to attempt to separate the science from the
policy in environmental decision making—to try, as clearly as possible,
to distinguish between the information or facts that we do know and
the policy-laden assumptions, inferences, and interpretations that have
been made.47 Different methods have been proposed to accomplish
this goal. J.B. Ruhl and James Salzman have argued for the use of
“regulatory peer review” to increase transparency by identifying
which parts of an agency’s decision are science and which parts are
44 Michael S. Carolan, The Bright- and Blind-Spots of Science: Why Objective Knowledge
Is Not Enough to Resolve Environmental Controversies, 34 Critic Sociology 725, 731 (2008);
Daniel Sarewitz, How Science Makes Environmental Controversies Worse, 7 Envir Sci &
Pol 385, 397 (2004).
45 Sarewitz, 7 Envir Sci & Pol at 396 (cited in note 44) (“When political stakes associated
with a controversy are relatively low, high certainty is more permissible than when the stakes
are high.”). See also Carolan, 34 Critic Sociology at 729–30 (cited in note 44); Naomi Oreskes,
Science and Public Policy: What’s Proof Got to Do with It?, 7 Envir Sci & Pol 369, 370–74
(2004); William R. Freudenburg, Risky Thinking: Facts, Values and Blind Spots in Societal
Decisions about Risks, 72 Reliab Eng & Sys Safety 125, 126–27 (2001); Sheila Jasanoff and
Brian Wynne, Science and Decisionmaking, in Steve Rayner and Elizabeth L. Malone, eds,
1 Human Choice and Climate Change 1, 30 (Battelle 1998).
46 See Yaron Ezrahi, Utopian and Pragmatic Rationalism: The Political Context of
Scientific Advice, 18 Minerva 111, 114–15 (1980). See also Pielke, Honest Broker at 39–53
(cited in note 43).
47 See, for example, Thomas O. McGarity and Wendy E. Wagner, Bending Science: How
Special Interests Corrupt Public Health Research 3 (Harvard 2008); Wendy Wagner and Rena
Steinzor, Introduction, in Wendy Wagner and Rena Steinzor, eds, Rescuing Science from
Politics: Regulation and the Distortion of Scientific Research 1, 9 (Cambridge 2006); E. Donald
Elliott, Strengthening Science’s Voice at EPA, 66 L & Contemp Probs 45, 51 (2003).
There is a third alternative, in which scientists and policy makers attempt to lay out the full
range of uncertainty based on the different choices that might be made for each individual
inference or assumption. The problem is that if there are a nontrivial number of inferences or
assumptions, and each has a reasonably large range of possible outcomes, the spread of
uncertainty for the final assessment very quickly becomes very large. We are then left in the
same situation as before, where scientists that stick only to the particular data that they have
are unable to provide useful information for policy makers. Of course, one could ask for only
“reasonable” choices to be made for the inferences and assumptions to narrow the uncertainty,
but then we return to the problem of the exercise of judgment by scientists.
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 483

policy.48 Various scholars have argued that “[c]ourts could increase
transparency by demanding clearer explanations of the policy
judgments that necessarily underlie regulatory decisions, and deferring
to those judgments when they are explained.”49 To reward agencies
that (1) identify uncertainty, (2) explain how they resolved that
uncertainty by making value choices, and (3) clarify what those value
choices were, courts could uphold decisions that articulate those
points, and remand decisions without such analysis.50 Relatedly, “the
unvarnished views of agency scientists or advisory panels” might be
made publicly accessible to increase transparency, constraining the
ability of political appointees to reject scientific advice without explicit
policy rationales.51
The challenge to separating out the science from the policy is
that it is unclear whether it is truly feasible to adequately separate
the two components for the large number of environmental law and
policy decisions that depend on environmental science.52 Value
choices are deeply embedded in much of the work of environmental
science, so embedded that many policy makers and even scientists
may not be aware of all of the implicit policy-based choices that they
have made in developing and using scientific information.53
48 J.B. Ruhl and James Salzman, In Defense of Regulatory Peer Review, 84 Wash U L
Rev 1, 43–47 (2006).
49 Doremus and Tarlock, 26 Pub Land & Res L Rev at 28 (cited in note 16). For other,
similar proposals, see Kathryn A. Watts, Proposing a Place for Politics in Arbitrary and
Capricious Review, 119 Yale L J 2, 32–33 (2009); Wagner, 95 Colum L Rev at 1712 (cited in
note 27).
50 Doremus and Tarlock, 26 Pub Land & Res L Rev at 29 (cited in note 16).
51 Id at 30. Other proposals include creating an ombudsman for environmental agencies
to address and investigate complaints about the blurring of science and policy. See Doremus,
86 Tex L Rev at 1645–46 (cited in note 22); Angus Macbeth and Gary Marchant, Improving the
Government’s Environmental Science, 17 NYU Envir L J 134, 160, 162 (2008), and developing
a separate science agency.
52 See, for example, Robert G. Hetes, Science, Risk, and Risk Assessment and Their
Role(s) Supporting Environmental Risk Management, 37 Envir L 1007, 1015 (2007) (noting that
the “underlying assumptions and analytical approaches” necessary for the use of science in
environmental law “are all informed by policy, judgment, and statute”); David E. Adelman,
The Art of the Unsolvable: Locating the Vital Center of Science for Environmental Law &
Policy, 37 Envir L 935, 938 (2007) (arguing that “thorny moral questions . . . are interwoven
with methodological considerations” in environmental science and have exacerbated
controversies); Deborah M. Brosnan, Science, Law, and the Environment: The Making of a
Modern Discipline, 37 Envir L 987, 999 (2007) (arguing that individuals should recognize how
science and law each are embedded in environmental decisions, as the two “cannot be easily
separated out”).
53 See Vern R. Walker, Transforming Science into Law: Transparency and Default
Reasoning in International Trade Disputes, in Wagner and Steinzor, eds, Rescuing Science from
Politics 165, 174 (cited in note 47). See also Stephen Bocking, Nature’s Experts: Science,
Politics, and the Environment 189 (Rutgers 2004); Doremus and Tarlock, 26 Pub Land & Res L
Rev at 15 (cited in note 16); Doremus, 86 Tex L Rev at 1627–28 (cited in note 22).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
484 The University of Chicago Law Review [79:471
For instance, by identifying a species, scientists create a category
of knowledge that is used in the legal system to protect the
environment through the ESA.54 Yet many of the assumptions,
inferences, and interpretations that must be made in the process of
identifying a species do not on their face necessarily require policy
choices: choices in methodology (for example, to rely heavily on the
physical characteristics of individual animals or plants, as opposed to
genetic information) will determine whether or not a particular
population will be identified as a separate species.55 But usually,
methodological choices will not regularly correlate with consistently
finding species more or less often; thus, one cannot consistently say
that, for instance, relying on physical characteristics will result in
identifying more species and therefore expanded regulatory
protection under the ESA. There have been claims in the scientific
community that some scientists have manipulated those
methodological choices in order to identify new species that help
advance the policy goals of protecting ecosystems or rare
populations.56 But this depends on the application of taxonomic
analysis in a particular context—in other areas, identifying significant
physical or genetic differences among populations might result in less
regulatory protection for individual populations or members of a
species (such as when there are claims that some members of a
population are hybrids that do not warrant protection under the
ESA).57 Because of the interaction of the legal context with the
diverse methodological choices, separating policy from science will
have to be case by case.
Another example from the ESA context is the determination of
whether a management or regulatory decision would lead to the
extinction of a species, a central regulatory question that has been
identified as an example of a “science” question.58 But the question
54 David Takacs, The Idea of Biodiversity: Philosophies of Paradise 186 (Johns Hopkins 1996).
55 See Holly Doremus, Listing Decisions under the Endangered Species Act: Why Better
Science Isn’t Always Better Policy, 75 Wash U L Q 1029, 1089–1111, 1133–34 (1997).
56 See notes 288–95 and accompanying text.
57 For example, hybridization between coyotes and endangered red wolves might result
in a determination that individual red wolves or red wolf populations no longer qualify for
protection under the ESA because they have diverged from the pure red wolf populations. For
a discussion of the various difficulties in mapping red wolves’ genetic ancestry, see generally
Craig R. Miller, Jennifer R. Adams, and Listette P. Waits, Pedigree-Based Assignment Tests for
Reversing Coyote (Canis latrans) Introgression into the Wild Red Wolf (Canis rufus)
Population, 12 Molecular Ecol 3287 (2003) (positing that “[t]he principal threat to the
persistence of the endangered red wolf . . . in the wild is hybridization with the coyote”).
58 See Ruhl and Salzman, 84 Wash U L Rev at 51 (cited in note 48) (identifying the
assessment of the economic and other costs of designation of a group as an endangered species
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 485

whether a particular decision will lead to the extinction of a species is
plagued with tremendous uncertainty. It requires significant
assumptions and inferences to answer, and therefore again it will
require judgments that will frequently implicate policy.59 For
instance, a scientist answering this question would have to decide
(among other questions) which of the commercial or field data is
“reliable” enough to warrant inclusion in any analysis, what
additional factors (besides the management or regulatory decision at
issue) might affect the species’s viability in the future, how those
factors will interact with the management or regulatory decision, and
how those factors will vary in the future over time and space.
Accordingly, the scientist will have to make judgment calls about
whether to include or exclude data or factors, and how to
characterize uncertainty—judgment calls that will surely affect the
ultimate decision about whether the species’s viability is threatened
by the proposed decision.
Or take the process of risk assessment for chemicals. This is
probably the area in which the interaction of science and policy has
been most closely studied and best understood, and yet even now
many observers question whether we have truly separated the
science and policy questions and whether it is feasible to do so on an
ongoing basis.60
Certainly the more sophisticated proposals to separate science
and policy are aware of these obstacles, and attempt to address them.
For instance, requiring the public disclosure of information produced
by civil-service employees relies in part on the distinction between
those employees (who presumably are more likely to have scientific
training and be more involved in collecting scientific information)
and political appointees (who presumably will be more focused on
policy implications).61 But it remains a difficult challenge, both
against the benefits to the designation as a “science” question). This is a question that the ESA
requires to be answered based solely on the “best available science.” 16 USC § 1533(b)(1)(A).
59 See Doremus, 75 Wash U L Q at 1112–28 (cited in note 55).
60 See, for example, Adelman, 37 Envir L at 937–39 (cited in note 52); Adam Babich, Too
Much Science in Environmental Law, 28 Colum J Envir L 119, 165 (2003); Valerie Watnick, Risk
Assessment: Obfuscation of Policy Decisions in Pesticide Regulation and the EPA’s Dismantling of
the Food Quality Protection Act’s Safeguards for Children, 31 Ariz St L J 1315, 1333–36 (1999);
Rushefsky, Making Cancer Policy at 41–44 & table 2.5, 180 (cited in note 25).
61 See Doremus, 86 Tex L Rev at 1644–45 (cited in note 22). See also Doremus and
Tarlock, 26 Pub Land & Res L Rev at 30 (cited in note 16). But there may still be problems
with political appointees “burying” themselves into the civil service. See Doremus, 86 Tex L
Rev at 1643–44 (cited in note 22). And, as will be developed more fully below, the scientists
themselves may have consciously or unconsciously already buried important policy
implications into scientific work. See Part II. Another proposed structural solution is the
establishment of independent advocates on behalf of science either within or outside the
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
486 The University of Chicago Law Review [79:471
because of the sheer number of stages at which the distinction must
be drawn for any individual policy choice62 and because of the
potential difficulty of separating out science from policy at any given
stage.63
Instead of trying to separate science and policy, we could try to
understand how policy preferences manifest themselves through the
diverse range of scientific disciplines. One reason this possibility hasn’t
been explored much to this date is that, in general, the environmental
law and science literature has focused on relatively large-scale
divisions in environmental science, primarily between “agency
science,” or “regulatory science,” and “pure science,” or “research
science.”64 My goal here is to build on these basic distinctions and to
explore how a more nuanced view of the range of scientific disciplines
in environmental science might affect environmental law. As it turns
out, that exploration will provide us with a number of options for
agency. See note 51. The problem remains, however, how to decide which aspects of a
management or regulatory problem are “science” such that the independent voice should have
primacy.
62 For example, there are many individual stages required to assess whether a species
might be jeopardized by a regulatory or management decision, many of which will have policy
implications. See notes 58–59 and accompanying text.
63 This is not to say that there are not a lot of (relatively) easy cases in which a particular
component of an overall policy decision can be identified as science based or policy based. At
one end of this spectrum, you might have the simple question of how many fish are actually
present in the net that is used to sample the fish population. At the other end of this spectrum,
you might have the question of whether your policy goal is to have a large or small population
size for that fish population.
64 Emily Hammond Meazell, Super Deference, the Science Obsession, and Judicial
Review as Translation of Agency Science, 109 Mich L Rev 733, 747–48 (2010); Jasanoff, The
Fifth Branch at 75–77 (cited in note 22); Doremus and Tarlock, 26 Pub Land & Res L Rev
at 27 (cited in note 16); A. Dan Tarlock, Environmental Law: Ethics or Science?, 7 Duke Envir
L & Pol F 193, 215 (1996); Peter H. Schuck, Multi-culturalism Redux: Science, Law, and
Politics, 11 Yale L & Pol Rev 1, 20 (1993). The former category is directed toward specific
policy ends and is more focused on the synthesis and interpretation of knowledge, while the
latter is directed by the interests of the scientists themselves. See Jasanoff, The Fifth Branch
at 77–79 (cited in note 22).
There are a few scholars who have briefly touched on a diversity within science greater
than the distinction between pure and applied science: David Adelman has noted that
“[s]cience is . . . inherently pluralistic, as the different scientific disciplines attest, and a unitary
conception of environmental science is neither a desirable end nor a viable goal.” Adelman,
37 Envir L at 939 (cited in note 52). Holly Doremus has, in the context of discussing the role
that science plays in identifying species for protection under the ESA, noted the existence of
disciplinary diversity. Doremus, 75 Wash U L Q at 1066–71 (cited in note 55). See also Schuck,
11 Yale L & Pol Rev at 20 (cited in note 64) (“Although it is common to speak of the scientific
‘community’ (and I do so here), such usage is a rather quaint way to describe what has in fact
become a fragmented profession.”); Douglas A. Kysar and James Salzman, Environmental
Tribalism, 87 Minn L Rev 1099, 1108, 1126–27 (2003) (noting disciplinary diversity in reaction
to Bjّrn Lomborg’s book, The Skeptical Environmentalist: Measuring the Real State of the
World).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 487

dealing with the dysfunctions of science and environmental law. But
first, I begin with what scientific disciplines are, their characteristics,
and how they might affect the production of knowledge in
environmental science.

[أحمد أبو زنط]

II. SCIENTIFIC DISCIPLINES
The complexity of environmental science makes it difficult, if
not impossible, for any one group of scientists to develop an
effective, unified perspective of all of the various natural resources of
interest to humans.65 Instead, the range of perspectives is shaped and
framed by different scientific disciplinary groups within
environmental science, from epidemiology to toxicology, wildlife
management to ecology, climate science to range science.66
Here I develop how external and internal forces shape the
existence and structure of disciplines, and the perspectives or
paradigms that frame the information that disciplines produce. I
elucidate these points using examples from environmental science.
The themes that I develop here—in particular, the peer-driven
nature of much of science and the inevitable tension between
external and internal pressures in shaping science—will be the basis
for my analysis of the role of scientific disciplines in policy making in
Part III.
In making these points, I do not mean to imply that
environmental science faces no constraints from the natural world in
reaching conclusions. In fact, the scientific method often does
produce fairly clear results: chemicals have been identified as
carcinogenic or toxic through lab experiments and epidemiology;67
threats to species from hunting, chemical pollution, or habitat
alteration have been identified and in some cases ameliorated;68 the
sources of many kinds of water pollution have been pinpointed and
65 Sarewitz, 7 Envir Sci & Pol at 390 (cited in note 44); Brian Wynne, Knowledges in
Context, 16 Sci, Tech, & Hum Values 111, 114 (1991).
66 David Sarewitz has noted that “nature itself—the reality out there—is sufficiently rich
and complex to support a science enterprise of enormous methodological, disciplinary, and
institutional diversity.” Sarewitz, 7 Envir Sci & Pol at 386 (cited in note 44). See also Bocking,
Nature’s Experts at 14–15 (cited in note 53); Carolan, 34 Critic Sociology at 726 (cited in
note 44) (“The disciplines serve an important analytic purpose: namely, they break down into
epistemologically manageable parts what is in reality an immensely complex world.”).
67 See, for example, R.A. Lemen, J.M. Dement, and J.K. Wagoner, Epidemiology of
Asbestos-Related Diseases, 34 Envir Health Persp 1, 2–7 (1980) (providing an overview of
historical asbestos-related epidemiological studies and results).
68 See, for example, Nicole Strong, Comment, The American Bald Eagle: Still a Protected
National Symbol, 12 Great Plains Nat Res J 232, 234–36 (2008) (describing impact of the
pesticide DDT on raptors).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
488 The University of Chicago Law Review [79:471
addressed.69 My argument is instead that (1) there are nontrivial
areas in environmental science for which the evidence from the
natural world cannot—at least at this point in time—resolve all
reasonable uncertainty; (2) in those areas of underdetermination, the
perspectives of scientists and scientific disciplines can significantly
shape the information that is produced; and (3) those perspectives
are in turn shaped by a range of factors external and internal to the
discipline. These three points are much more important in the
context of “cutting edge” science that is more fallible—precisely the
kind of science that is much more likely to be relevant for policy
making. But as time proceeds and additional information is
collected, the issues of underdetermination will often decrease. And
while these issues are more prevalent in the fields of environmental
science because of dynamism and complexity, those fields still
frequently provide important conclusions informed by the natural
world, just like the more paradigmatic hard science disciplines such
as physics or chemistry.70
A. The State of Global Fisheries: An Example of How Disciplines
Matter in Environmental Science
One example of how the different perspectives from different
scientific disciplines can produce conflicts with significant policy
implications is the conflict between marine ecologists and fisheries
scientists over the state of global fisheries.71 This is an issue of crucial
importance for the people that depend on the oceans for food and
livelihoods as well as for those concerned about the protection of
marine species and ecosystems.
In 2003 and 2006, a relatively junior scientist, Boris Worm,
published papers in Nature and Science, the two most prestigious
scientific journals, contending that global fisheries were collapsing
because of inadequate management, overfishing, and other human-
69 See, for example, John Harte, Toward a Synthesis of the Newtonian and Darwinian
Worldviews, Physics Today 29, 32 (Oct 2002).
70 I am adopting the middle ground in the so-called science war between positivists and
relativists. See D. Michael Risinger, The Irrelevance, and Central Relevance, of the Boundary
between Science and Non-science in the Evaluation of Expert Witness Reliability, 52 Vill L
Rev 679, 679–86 (2007) (noting that the debate is over “the proper proportion to be accorded
to social construction factors in comparison to the external phenomena under examination”);
John Ziman, Real Science: What It Is, and What It Means 234–36 (Cambridge 2000) (coming to
the conclusion that “scientific knowledge is both found and made” since academic science
gives great weight to the empirical findings of research, while still requiring the formulation of
theories by scientists).
71 I am indebted to Holly Doremus for helping to identify this example.
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 489

caused impacts.72 Given the policy importance of the conclusions for
marine and fisheries management, the papers received significant
press attention.73 The papers also received significant criticism from
numerous fisheries scientists who assailed Worm’s methodologies
and asserted that his doom-and-gloom statements were overstated.74
While the scientific debates appear highly technical to an
outsider, at heart they are the result of different judgments being
applied by scientists from different disciplinary backgrounds as to
the appropriate data and methods to be used to understand the state
of global fisheries given extremely limited information.75 Worm “is a
rising star among marine ecologists; soft-spoken and mediasavvy,
. . . a passionate conservationist.”76 His leading antagonist in
the dispute, Ray Hilborn, is an award-winning fisheries scientist.77
Marine ecologists generally study a wide range of natural marine
systems, not just those that might be relevant for human exploitation
(such as fish stocks); fisheries scientists have traditionally focused on
72 Ransom A. Myers and Boris Worm, Rapid Worldwide Depletion of Predatory Fish
Communities, 423 Nature 280, 282 (2003) (“Our analysis suggests that the global ocean has lost
more than 90% of large predatory fishes.”); Boris Worm, et al, Impacts of Biodiversity Loss on
Ocean Ecosystem Services, 314 Sci 787, 790 (2006) (asserting that if present trends continue, all
global fisheries will collapse by 2048).
73 Erik Stokstad, Détente in the Fisheries War, 324 Sci 170, 170 (2009).
74 See, for example, Ray W. Hilborn, Biodiversity Loss in the Ocean: How Bad Is It?,
316 Sci 1281, 1281 (2007); Steven Murawski, Richard Methot, and Galen Tromble, Biodiversity
Loss in the Ocean: How Bad Is It?, 316 Sci 1281, 1281 (2007) (calling Worm’s proxy for
abundance “inadequa[te]”); John C. Briggs, Biodiversity Loss in the Ocean: How Bad Is It?,
316 Sci 1282, 1282 (2007); Ray Hilborn, Faith-Based Fisheries, 31 Fisheries 554, 554–55 (2006)
(criticizing Myers and Worm’s paper as an example of “faith-based” fisheries research); Tom
Polacheck, Tuna Longline Catch Rates in the Indian Ocean: Did Industrial Fishing Result in a 90%
Decline in the Abundance of Large Predatory Species?, 30 Marine Pol 470, 470 (2006); Franz
Hِlker, et al, Comments on Impacts of Biodiversity Loss on Ocean Ecosystem Services,
316 Sci 1285, 1285c (2007); John Hampton, et al, Decline of Pacific Tuna Populations
Exaggerated?, 434 Nature E1, E1 (2005); Carl Walters, Folly and Fantasy in the Analysis of Spatial
Catch Rate Data, 60 Can J Fish Aquat Sci 1433, 1433–34 (2003). See also Stokstad, 324 Sci at 170
(cited in note 73) (“Many fisheries scientists were appalled. Trained in quantitative techniques for
determining the abundance of fish stocks, they questioned the methods used in Worm’s global
assessment, such as a reliance on the mass of fish reported caught.”).
75 See Stokstad, 324 Sci at 170–71 (cited in note 73).
76 Id at 171.
77 Id (noting that Hilborn was awarded the Volvo Environment Prize in 2006). Other
critics were also leading fisheries managers and scientists. See, for example, Hِlker, et al,
316 Sci at 1285c (cited in note 74) (presenting a challenge to Worm’s findings and conclusions
brought by the authors, who are all European fisheries scientists); Walters, 60 Can J Fish
Aquat Sci at 1433 (cited in note 74) (admonishing scientists who use “nonspatial” catch-pereffort
data, an argument made by the author who has collaborated with Hilborn); Murawski,
Methot, and Tromble, 316 Sci at 1281 (cited in note 74) (presenting commentary from three
scientists at the National Marine Fisheries Service, the agency in charge of fisheries regulation
in the United States); Briggs, 316 Sci at 1282 (cited in note 74) (presenting a critique of the
Worm study brought by an author who is a fisheries scientist).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
490 The University of Chicago Law Review [79:471
gathering data on fish species that are used for human benefit in
order to manage them for maximum human use over the long run.78
A key part of the dispute was the use of “catch per effort” data.
Fish are elusive and hard to count, given that their natural habitat is
so inhospitable to humans. One way to estimate their population size
is to measure how much effort commercial fishermen must exert to
catch fish. The advantage of these data is that they are plentiful:
commercial fishermen have been collecting these data for decades
and for many of the fisheries around the world.79 It allows for a
global, historical overview of fisheries trends that is otherwise
difficult to obtain.80 This was the data that Worm relied upon in his
two controversial papers.81
These data were the focus of the critics of Worm’s papers.82 They
argued that “catch-per-effort” data necessarily require an
understanding of the “effort” that commercial fishers put into
fishing, and that therefore “catch-per-effort” data may give a
misleading impression of the status of fish species.83 For instance,
“catch-per-effort” may decline substantially (in other words, it might
require much more effort for fishermen to catch fish) but that may
not necessarily reflect changes in the population of a fish species,
depending on how the effort is distributed in space and time, and
how the species reacts to greater fishing pressures.84 Thus, one fishery
scientist called the use of these data “folly” because of the need for
significant assumptions to extrapolate the data across time and
space.85
78 Stokstad, 324 Sci at 170 (cited in note 73).
79 See Myers and Worm, 423 Nature at 280 (cited in note 72); Worm, et al, 314 Sci at 788
(cited in note 72).
80 See Myers and Worm, 423 Nature at 280 (cited in note 72).
81 See id at 280; Worm, et al, 314 Sci at 788 (cited in note 72). Worm and his coauthor
provided detailed justifications for why they relied upon this data in the articles themselves and
in technical appendices, and also provided some estimates of the uncertainty associated with
their conclusions.
82 See, for example, Hilborn, 31 Fisheries at 554 (cited in note 74).
83 See, for example, Hampton, et al, 434 Nature at E1 (cited in note 74); Murawski,
Methot, and Tromble, 316 Sci at 1281 (cited in note 74).
84 See Polacheck, 30 Marine Pol at 471–74 (cited in note 74).
85 See Walters, 60 Can J Fish Aquat Sci at 1433–34 (cited in note 74). Walters instead
argued for “reasonable assumptions” that resulted in conclusions that fisheries were far less
overexploited than Myers and Worm concluded. Id at 1434–35 & figure 1. Importantly,
Walters conceded that both his approach and Worm’s approach involved an element of
“fantasy” and that Walters’s approach (which rejected making simple assumptions about how
known data might be representative of missing data) required “strong judgmental decisions
about what assumptions to use.” Id at 1434. For Walters, his “reasonable assumptions”
contrasted with “unwise ones [that are] based on the mistaken notion that statistical rigor or
scientific objectivity . . . can somehow substitute for judgment.” Id. In short, the fisheries
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 491

Fisheries scientists instead emphasized an alternate source of
information: fisheries assessments. These are complicated analyses of
the status of particular fish stocks based, in part, on “catch-pereffort”
data, but also on surveys conducted by research vessels and
other data sources.86 Fisheries scientists in their critiques emphasized
that these sources are more accurate and reliable than simple “catchper-
effort” data alone.87 The problem is, however, that only a few fish
stocks around the world have had fisheries assessments conducted
for them, and most assessments only go back for a very limited
period of time.88 Fisheries assessments therefore cannot provide a
large-scale and long-term assessment of the status and trends of
global fisheries.89
That is the heart of the conflict. Worm, as a marine ecologist,
was interested in getting a global, long-term assessment of the
impacts of human activities on the oceans. His concern was that
looking only at the best-studied fish stocks for a relatively short
period of time would greatly underestimate the negative impacts
humans have caused on marine systems, since many of those impacts
have occurred either on fish stocks that are not well studied (often in
developing countries) or before fisheries assessments occurred.
Underestimating negative impacts would, in turn, undermine the
scientific case for the dramatic changes in marine policy that Worm
believed were necessary to protect all marine biodiversity.90
scientist critics of Worm did not, and could not, argue that there was a definitive reason why
their conclusions were scientifically superior to those of Worm.
86 See Polacheck, 30 Marine Pol at 474 (cited in note 74); Murawski, Methot, and
Tromble, 316 Sci at 1281 (cited in note 74).
87 See Polacheck, 30 Marine Pol at 474 (cited in note 74); Walters, 60 Can J Fish Aquat
Sci at 1433–35 (cited in note 74). See also Hampton, et al, 434 Nature at E1 (cited in note 74);
Murawski, Methot, and Tromble, 316 Sci at 1281 (cited in note 74).
88 See Polacheck, 30 Marine Pol at 470–71 (cited in note 74); Boris Worm, et al,
Biodiversity Loss in the Ocean: How Bad Is It?, 316 Sci 1282, 1283 (2007).
89 See Stokstad, 324 Sci at 170–71 (cited in note 73).
90 These points are made clear in a book review that Worm wrote for Science shortly after
his controversial 2006 article. See Boris Worm, Book Review, Armageddon in the Oceans,
314 Sci 1546, 1546 (2006). The reviewed book was a thriller that depicted a global ecological crisis
based in the oceans that imminently threatened human societies. Worm approvingly cites the
book for emphasizing the need for scientists to play a role now in saving oceans. Worm approves
of criticism in the book that scientists are “shown to be slow to communicate their results, usually
waiting for absolute certainty.” Id. Worm emphasizes the importance of putting data together to
get global, long-term perspective so we can understand how threatened oceans are:
This is what emerges as the most interesting message from the book: that the oceans are
changing on a global scale, that our understanding of these changes always lags behind
them, and that too often we are too slow, too conservative, or simply too unimaginative to
put all the pieces together.
Id (describing approvingly the book’s mix of truth and fiction).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
492 The University of Chicago Law Review [79:471
Fisheries scientists, on the other hand, emphasized caution in
their estimates and the need to avoid both over- and underestimating
fish populations.91 This perspective can be understood as coming out
of the discipline’s long history providing information to be used in
assessing the population size of fish stocks so that appropriate
harvest levels can be set by regulatory agencies.92 In this context, risks
exist in both directions: overestimating fish populations carries the
risk of setting harvest levels too high and causing the collapse of a
fish stock (the risk that Worm highlighted); but underestimating fish
populations carries the risk of leaving a fish in the sea, unharvested,
not benefiting humans, with resulting social and economic impacts.
Thus, a dispute over whose science was “better” or “worse” is
actually a fundamental disagreement between two scientific
disciplines over what factors are most important in making difficult
judgments about data, methodologies, and interpretation of data.
One discipline (marine ecology) values the importance of reaching
conclusions that are broad in scale for both time and space, even if
the result is higher uncertainty for those estimates; another discipline
(fisheries science) values the importance of precision in estimates
even at the expense of being very limited in making statements at
larger levels of time and scale. These are judgment calls about when
an answer is adequate to justify communicating it to a broader
scientific or policy audience. It has epistemic implications, but it is
affected by other factors, such as how the discipline values risks with
respect to policy outcomes.93 There is no right or wrong answer from
91 See, for example, Walters, 60 Can J Fish Aquat Sci at 1434 (cited in note 74)
(emphasizing the need for “reasonable assumptions” in fisheries science).
92 See, for example, Josh Eagle and Barton H. Thompson Jr, Answering Lord Perry’s
Question: Dissecting Regulatory Overfishing, 46 Ocean & Coastal Mgmt 649, 651–53 (2003); Ehsan
Masood, Fisheries Science: All at Sea When It Comes to Politics?, 386 Nature 105, 106 (1997); Ray
Hilborn, Ellen K. Pikitch, and Robert C. Francis, Current Trends in Including Risk and Uncertainty
in Stock Assessment and Harvest Decisions, 50 Can J Fish Aquat Sci 874, 875 (1993).
93 Marine ecology or fisheries science are not unique in having a range of values shape their
conclusions about what is an adequate answer worthy of communication in areas where there are
high levels of uncertainty. At their heart, all decisions about statistical significance (that is,
conclusions about whether results are statistically unusual enough that they warrant publication)
are decisions of this sort, and they necessarily implicate policy. See, for example, Berry J. Brosi
and Eric G. Biber, Statistical Inference, Type II Error, and Decision Making under the US
Endangered Species Act, 7 Frontiers Ecol & Envir 487, 488 (2009). For discussions of these issues
in other fields, see, for example, Scott D. Halpern, Jason H.T. Karlawish, and Jesse A. Berlin, The
Continuing Unethical Conduct of Underpowered Clinical Trials, 288 J Am Med Assn 358, 359
(2002) (discussing the role that statistical analysis plays in decisions about ethical medical
research); S.J.L. Edwards, et al, Why “Underpowered” Trials Are Not Necessarily Unethical,
350 Lancet 804, 807 (1997).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 493

a scientific perspective in this dispute—instead, the dispute is about
values and perspectives underlying the science.
B. Defining Scientific Disciplines
On an intuitive level, anyone who has spent a fair amount of
time at a major research university understands the concept of
scientific disciplines: the different departments (from economics to
biology, from sociology to physics) are oriented around one level and
conception of disciplines. While there is great diversity in the
definitions of a scientific discipline,94 for purposes of this Article I use
a simple, general one: disciplines are epistemic subdivisions of
science. My definition includes a wide range of scales for a discipline,
from relatively large units (such as physics or chemistry), to smaller,
more informal subunits (such as experimental nuclear physics or
evolutionary biology).95 My focus here is on the relatively larger
units, though not necessarily those formally identified as university
departments.
In my analysis, I will draw on two examples from environmental
science: conservation biology and wildlife management. These fields
are similar enough that a comparison helps illuminate many of my
points about how scientific disciplines and their perspectives
develop. Moreover, conservation biology is a new and specialized
field that has not received formal recognition as a department at
many institutions,96 and using it as an example shows the possibilities
and challenges of identifying relatively informal disciplinary
structures.
Both conservation biology and wildlife management (or wildlife
biology) focus on the management of populations of nondomes-
94 See, for example, Russel McCormmach, Editor’s Foreword, 3 Hist Stud Phys Sci ix, ix
(1971); Robert E. Kohler, From Medical Chemistry to Biochemistry: The Making of a
Biomedical Discipline 1–2 (Cambridge 1982).
95 See Ziman, Real Science at 193 (cited in note 70) (adopting this defintion of
“discipline”). Disciplines at varying scales are defined by sociologists, historians of science, and
other scholars using a range of terms. Terms used to define disciplines include “epistemic
communities,” Peter M. Haas, Introduction: Epistemic Communities and International Policy
Coordination, 46 Intl Org 1, 2–3 (1992), “invisible colleges,” Diana Crane, Invisible Colleges:
Diffusion of Knowledge in Scientific Communities 35 (Chicago 1972), and “specialties,”
Warren O. Hagstrom, The Scientific Community 159–62 (Basic Books 1965). For further
examples of terms used to describe scientific disciplines, see Ludwik Fleck, Genesis and
Development of a Scientific Fact 20–51, 94–95 (Chicago 1979) (developing the related concept
of a “thought collective”); Thomas S. Kuhn, The Structure of Scientific Revolutions 177
(Chicago 3d ed 1996) (developing the concept of a “scientific community”).
96 See Susan K. Jacobson, Graduate Education in Conservation Biology, 4 Conservation
Bio 431, 431–32 (1990).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
494 The University of Chicago Law Review [79:471
ticated species. The most obvious difference between the two is that
wildlife management has historically focused on individual game
species (what might be called “deer and ducks”),97 seeking to identify
the maximum number of animals an ecosystem could support
(“carrying capacity”) and to understand how to artificially
manipulate “habitats and populations” to allow for larger
“harvestable surpluses” of those species for hunting.98 Conservation
biology in contrast seeks to answer questions about how to protect
biodiversity in all its forms (ecosystems, species, and populations).99
It values protecting biodiversity for its own sake (not necessarily
based on direct human use).100 A useful simplification is that a
wildlife biologist traditionally would publish a paper on the quality
of white-tailed deer habitat in a particular county,101 while a
97 See Thomas C. Edwards Jr, The Wildlife Society and the Society for Conservation
Biology: Strange but Unwilling Bedfellows, 17 Wildl Socy Bull 340, 341–42 (1989); Paul R.
Krausman, Wildlife Management in the Twenty-First Century: Educated Predictions, 28 Wildl
Socy Bull 490, 490 (2000); Frederic H. Wagner, American Wildlife Management at the
Crossroads, 17 Wildl Socy Bull 354, 356 (1989); Stanley A. Temple, et al, What’s So New about
Conservation Biology?, Trans 53rd N Am Wildl & Nat Res Conf 609, 610–11 & table 1 (1988).
Studies of publications in wildlife journals have found a heavy historical bias toward game
species. Robert A. Powell, et al, Dynamics of Content and Authorship Patterns in the Wildlife
Society Journals (1937–2007), 74 J Wildl Mgmt 816, 816 (2010); Mari N. Jensen and Paul R.
Krausman, Conservation Biology’s Literature: New Wine or Just a New Bottle?, 21 Wildl Socy
Bull 199, 201 table 1 (1993); Fred L. Bunnell and Linda A. Dupuis, Conservation Biology’s
Literature Revisited: Wine or Vinaigrette?, 23 Wildl Socy Bull 56, 56 (1995); R. Douglas Slack
and Nova J. Silvy, Have the Wildlife Society’s Publications Kept Pace with the Profession?,
Trans 55th N Am Wildl & Nat Res Conf 164, 167–70 (1990).
98 See Nathan F. Sayre, The Genesis, History, and Limits of Carrying Capacity, 98 Annals
Assn Am Geographers 120, 125–26 (2008).
99 Temple, et al, What’s So New about Conservation Biology? at 610–11 & table 1 (cited
in note 97) (“Conservation biology has no taxonomic bias and, at least in principle, treats all
taxa equitably.”).
100 See Takacs, The Idea of Biodiversity at 6, 35, 114–15 (cited in note 54); Reed F. Noss,
Who Will Speak for Biodiversity?, 3 Conservation Bio 202, 202 (1989) (“The fact that we are
conservation biologists suggests that we value biodiversity.”); Peter F. Brussard and John C. Tull,
Conservation Biology and Four Types of Advocacy, 21 Conservation Bio 21, 22 (2007); Reed F.
Noss, Values Are a Good Thing in Conservation Biology, 21 Conservation Bio 18, 20 (2007). For a
discussion of how conservation biology is less focused on the utilitarian benefits to humans of
wildlife conservation compared to wildlife biology, see Reed Noss, Is There a Special
Conservation Biology?, 22 Ecography 113, 118 & table 1 (1999); Reed Noss, Aldo Leopold Was a
Conservation Biologist, 26 Wildl Socy Bull 713, 717 (1998); Bruce A. Wilcox, Editorial,
1 Conservation Bio 188, 188 (1987); Jack Ward Thomas and Hal Salwasser, Bringing
Conservation Biology into a Position of Influence in Natural Resource Management,
3 Conservation Bio 123, 125 (1989); Michael E. Soulé, What is Conservation Biology?,
35 BioSci 727, 728 (1985).
101 See Malcolm L. Hunter Jr, Aardvarks and Arcadia: Two Principles of Wildlife
Research, 17 Wildl Socy Bull 350, 350 (1989).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 495

conservation biologist would publish a paper on the conservation
status of a cave spider.102
Wildlife management is the older discipline, having developed
beginning in the 1930s,103 while conservation biology was created in
the 1980s.104 Wildlife managers saw the rise of conservation biology in
the 1980s as threatening; they asserted that conservation biology
sought to answer the same questions wildlife management had been
exploring for decades.105 In the intervening decades, there have been
claims that the two fields have converged, as wildlife managers have
become more interested in questions about the conservation of a
wider range of species and ecosystems, and less interested in
maximizing game species populations to benefit humans.106
C. How Disciplinary Structures Are Created and Shaped
Disciplines are the products of a range of intellectual, political,
economic, social, and institutional factors, and as those factors
change, the structure of disciplines will also change over time,
potentially creating new disciplines through the splitting or merging
of existing fields.107 For instance, the new field of tropical medicine
developed in early-twentieth-century Britain as a result of internal
102 See, for example, Francis G. Howarth, Ecology of Cave Arthropods, 28 Ann Rev
Entomology 365, 382 (1983).
103 See notes 182–85 and accompanying text.
104 See notes 163–65 and accompanying text.
105 Noss, 22 Ecography at 113 (cited in note 100) (noting that “wildlife and fisheries
biologists, foresters, range managers, and other applied scientists . . . feel threatened by this
ostensibly new ‘metadiscipline’ that has suddenly grabbed the spotlight of scientific and policy
interest”); id at 114 (“[A] professional rivalry of sorts has developed between conservation
biologists . . . and wildlife biologists.”); James G. Teer, Book Review, Conservation Biology:
The Science of Scarcity and Diversity, 52 J Wildl Mgmt 570, 571 (1988) (leading wildlife
biologist questioning why conservation biologists felt the need to create a new society instead
of joining “an already established professional society whose interests and emphasis have been
and are on conservation of the natural world”); Reed F. Noss, The Failure of Universities to
Produce Conservation Biologists, 11 Conservation Bio 1267, 1267 (1997) (asserting that wildlife
resource “departments have successfully fought implementation of interdisciplinary programs
in conservation biology, ostensibly because they already have the subject covered within their
curricula. (I suspect the real reason may be the envy and fear among those in resource
management fields that conservation biology has stolen the professional spotlight from
them.)”).
106 Robert D. Brown and Larry A. Nielsen, Leading Wildlife Academic Programs into the
New Millennium, 28 Wildl Socy Bull 495, 497 (2000); Jack Ward Thomas and Daniel H.
Pletscher, The Convergence of Ecology, Conservation Biology, and Wildlife Biology: Necessary
or Redundant?, 28 Wildl Socy Bull 546, 547 (2000). See also notes 279–85 and accompanying
text.
107 Gerard Lemaine, et al, Introduction: Problems in the Emergence of New Disciplines, in
Gerard Lemaine, et al, eds, Perspectives on the Emergence of Scientific Disciplines 1, 16
(Mouton 1976).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
496 The University of Chicago Law Review [79:471
intellectual changes (the development of the germ theory of disease,
which made the study of the causes of disease much more tractable)
and external social, political, and institutional pressures (the need for
the development of a medical discipline that could allow Europeans
to live in tropical regions for extended periods of time to run the
colonial system and the need of the British Colonial Office to train
doctors who could serve in the colonies).108 I will discuss many of the
external factors in the next Section, since they overlap with the
factors that shape the perspectives of disciplines. What I will focus
on here is how the disciplinary nature of science is a necessary
outcome of the communication and specialization among scientists,
and how that both shapes the structure of disciplines and allows an
understanding of the borders of disciplines.
Patterns of communication among scientists are shaped by, and
in turn shape, disciplines. At the larger, more formal levels,
institutions such as university departments, PhD training programs,
academic journals, and professional societies will provide structures
for the sharing and certification of knowledge that also demarcate
and develop borders between disciplines.109 The intellectual history of
a discipline (including the prior disciplines from which it developed)
will shape patterns of communication among scientists, as
communication is more frequent among closely related disciplines.110
Subject matter will shape patterns of communication, as material
that is seen as too distant or irrelevant will not be read, accepted for
publication, or even understood.111 Relatedly, the intellectual history
of a discipline (that is, which prior disciplines it developed out of)
will shape patterns of communication among scientists, as
communication is more frequent among closely related disciplines.112
Journals, peer review, and review articles function as gatekeeping
devices that both create and maintain disciplinary borders by
controlling what kinds of communications “count” within a
108 Michael Worboys, The Emergence of Tropical Medicine: A Study in the Establishment
of a Scientific Specialty, in Lemaine, et al, eds, Perspectives 75, 93 (cited in note 107).
109 See, for example, M.J. Mulkay, G.N. Gilbert, and S. Woolgar, Problem Areas and
Research Networks in Science, 9 Sociology 187, 189 (1975); S.W. Woolgar, The Identification
and Definition of Scientific Collectivities, in Lemaine, et al, eds, Perspectives 233, 234 (cited in
note 107); Crane, Invisible Colleges at 35 (cited in note 95).
110 See, for example, Kohler, From Medical Chemistry at 1–2 (cited in note 94); Crane,
Invisible Colleges at 115–21 (cited in note 95); Hagstrom, The Scientific Community at 217–20
(cited in note 95).
111 See, for example, Crane, Invisible Colleges at 108, 115–21 (cited in note 95).
112 See, for example, Kohler, From Medical Chemistry at 214, 253–55 (cited in note 94)
(noting deference in biochemistry to chemistry, the parent discipline).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 497

particular field.113 At smaller, more informal scales, patterns of
citation and publication rates provide evidence of subdisciplinary
borders and the emergence of new fields: citation networks are
tighter within rather than across disciplines or subdisciplines,
reflecting the underlying social relationships that help establish
disciplines, such as collaboration, teaching, and citation.114
The importance of communication stems from the inevitable,
even necessary, role that disciplinary specialization plays in the
development of scientific information. Specialization—and
accordingly, the development of disciplines—happens because of the
limits on the amount of information that researchers can absorb, and
the choices that researchers must make to “communicate with those
who are concerned with problems similar in some way to their
own.”115 In this way, disciplines are central to the functioning of
science.116
Communication patterns and related institutional structures can
therefore be used to help identify and distinguish disciplines.
Researchers in information science have used these tools to study
how disciplines and interdisciplinarity shape the production of
knowledge. Techniques to determine whether a particular article is
“interdisciplinary” range from simple ones based on the
categorization of research by subject matter (such as keyword and
classification headings for published articles), to the use of
departmental affiliations for listed authors, to the publication and
citations of articles by journals in different disciplines.117
113 Crane, Invisible Colleges at 115–22, 128 (cited in note 95); Hagstrom, The Scientific
Community at 23, 210, 224 (cited in note 95); Saad Z. Nagi and Ronald G. Corwin, The
Research Enterprise: An Overview, in Saad Z. Nagi and Ronald G. Corwin, eds, The Social
Contexts of Research 1, 22 (Wiley 1972).
114 Crane, Invisible Colleges at 41 (cited in note 95) (discussing the influence of informal
relationships, such as between students and teachers, in publication and evaluation).
115 Mulkay, Gilbert, and Woolgar, 9 Sociology at 189 (cited in note 109); Woolgar,
Definition of Scientific Collectivities at 234 (cited in note 109); Ziman, Real Science at 46–49,
189–90 (cited in note 70); Derek J. de Solla Price, Little Science, Big Science 62–91
(Columbia 1963).
116 Specialization, by ensuring a diversity of perspectives and information in science, may
also help produce scientific knowledge. See Philip Kitcher, The Advancement of Science:
Science without Legend, Objectivity without Illusions 344 (Oxford 1993).
117 See Joachim Schummer, Multidisciplinarity, Interdisciplinarity, and Patterns of
Research Collaboration in Nanoscience and Nanotechnology, 59 Scientometrics 425, 435–36
(2004) (describing some of the “scientometric approaches to measuring interdisciplinarity”);
Marيa Bordons, Fernanda Morillo, and Isabel Gَmez, Analysis of Cross-Disciplinary Research
through Bibliometric Tools, in Henk F. Moed, Wolfgang Glنnzel, and Ulrich Schmoch, eds,
Handbook of Quantitative Science and Technology Research: The Use of Publication and
Patent Statistics in Studies of S&T Systems 437, 437 (Kluwer 2004); Mu-Hsuan Huang and Yu-
Wei Chang, A Study of Interdisciplinarity in Information Science: Using Direct Citation and Co01
BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
498 The University of Chicago Law Review [79:471
With these tools, even fields as similar as conservation biology
and wildlife management can be distinguished to some extent. For
instance, in terms of subject matter, as noted above, wildlife
management historically has focused more on single-species game
management that has direct connections to human use (such as
hunting). In addition, conservation biology is generally more
theoretical than wildlife management,118 and answers questions at
larger temporal and spatial scales.119
These differences reflect different intellectual pedigrees.
Conservation biology developed out of the more theoretical field of
ecology and related areas (such as evolutionary and population
biology), while wildlife management developed out of the more
applied fields of resource management, such as forestry.120 The
authorship Analysis, 37 J Info Sci 369, 369 (2011); Thomas W. Steele and Jeffrey C. Stier, The
Impact of Interdisciplinary Research in the Environmental Sciences: A Forestry Case Study, 51 J
Am Socy Info Sci 476, 477 (2000); Jian Qin, F.W. Lancaster, and Bryce Allen, Types and
Levels of Collaboration in Interdisciplinary Research in the Sciences, 48 J Am Socy Info Sci 893,
894–97 (1997); Powell, et al, 74 J Wildl Mgmt at 818 (cited in note 97); Sydney J. Pierce,
Boundary Crossing in Research Literatures as a Means of Interdisciplinary Information
Transfer, 50 J Am Socy Info Sci 271, 273–74 (1999).
118 See Krausman, 28 Wildl Socy Bull at 493 (cited in note 97); Jacobson, 4 Conservation
Bio at 432 (cited in note 96); Edwards, 17 Wildl Socy Bull at 340–41 (cited in note 97); Thomas
A. Gavin, What’s Wrong with the Questions We Ask in Wildlife Research?, 17 Wildl Socy
Bull 345, 345–46 (1989); Wagner, 17 Wildl Socy Bull at 358 (cited in note 97); Gregory H.
Aplet, Richard D. Laven, and Peggy L. Fiedler, The Relevance of Conservation Biology to
Natural Resource Management, 6 Conservation Bio 298, 299 & table 1 (1992); Tim W. Clark,
Developing Policy-Oriented Curricula for Conservation Biology: Professional and Leadership
Education in the Public Interest, 15 Conservation Bio 31, 33 (2001); Wilcox, 1 Conservation Bio
at 188 (cited in note 100); Noss, 11 Conservation Bio at 1267 (cited in note 105). Studies of
publications in wildlife and conservation biology journals have reached similar conclusions.
See, for example, Jensen and Krausman, 21 Wildl Socy Bull at 201 table 1 (cited in note 97);
Bunnell and Dupuis, 23 Wildl Socy Bull at 59–60 & figure 4 (cited in note 97); I. Fazey,
J. Fischer, and D.B. Lindenmayer, What Do Conservation Biologists Publish?, 124 Bio
Conservation 63, 71 (2005).
119 See, for example, Bunnell and Dupuis, 23 Wildl Socy Bull at 61 (cited in note 97);
Noss, 26 Wildl Socy Bull at 718 (cited in note 100) (comparing conservation biology with
similar fields and noting that “conservation biology emphasizes a long-term perspective, [and]
focuses on several levels of biological organization (e.g., genes through ecosystems)”); Soulé,
35 BioSci at 728 (cited in note 100).
120 See, for example, David Ehrenfeld, Editorial, 1 Conservation Bio 6, 6–7 (1987) (listing
relevant disciplines for conservation biology by the first editor in chief of Conservation Biology
that includes ecologists, taxomonists, and population biologists, but not wildlife biologists);
Joshua J. Lawler, et al, Conservation Science: A 20-Year Report Card, 4 Front Ecol & Envir
473, 474 table 1 (2006) (providing a list of journals surveyed for assessment of conservation
science that focuses on applied ecology journals); Jacobson, 4 Conservation Bio at 431–32
(cited in note 96); Noss, 22 Ecography at 115–16 (cited in note 100); Curt Meine, Michael
Soulé, and Reed F. Noss, “A Mission-Driven Discipline”: The Growth of Conservation Biology,
20 Conservation Bio 631, 639 (2006) (“Conservation biology’s scientific foundations lie at the
interface of systematics, genetics, ecology, and evolutionary biology.”); Gavin, 17 Wildl Socy
Bull at 345 (cited in note 118) (arguing that conservation biology developed out of “basic”
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 499

different disciplinary origins are reflected to some extent in the
institutional homes for conservation biology and wildlife
management programs,121 and to a lesser extent, the differences in
educational backgrounds of faculty members in each program.122
Journals provide the clearest evidence of distinctions between
wildlife management and conservation biology. Studies of the
journals published by the leading wildlife management professional
society—the Wildlife Society (TWS)—and the leading conservation
biological sciences); Noss, 26 Wildl Socy Bull at 718 (cited in note 100) (claiming that
conservation biology “is clearly dominated by biologists, and especially by ecologists and
geneticists”); Steven R. Beissinger, On the Limits and Directions of Conservation Biology,
40 BioSci 456, 456 (1990) (“In the past decade, the principles of population and ecosystem
ecology, biogeography, and population genetics have been applied to the problems of
preserving biological diversity. This synthesis has been termed conservation biology.”); José
Sarukhلn, Conservation Biology: Views from the Ecological Sciences, 20 Conservation Bio 674,
674 (2006) (“If there is an area of science with which conservation biology has been intimately
related, it is ecology. . . . Conservation biology is ecology applied to preserving species,
populations, and communities.”); Temple, et al, What’s So New about Conservation Biology?
at 610–11 & table 1 (cited in note 97) (contrasting conservation biology with wildlife
management due to the fact that it “is a synthetic discipline whose practitioners hail from a
diverse range of backgrounds, including population genetics, demography, community ecology,
ecosystem ecology and evolutionary biology”).
121 Wildlife management programs are somewhat more likely, for instance, to be located
in “natural resources management” departments than in ecology departments compared to
conservation biology programs. Of 414 wildlife undergraduate and graduate programs listed by
the main wildlife management professional society, the Wildlife Society, 111 (26 percent) were
located in departments that had “wildlife” or “natural resources” in their title, while 161 (39
percent) were located in biology departments, 88 (25 percent) were in environmental studies or
science departments, and only 9 (2 percent) were in ecology or zoology departments. The
Wildlife Society, Table of Wildlife Degrees, online at http://joomla.wildlife.org/documents
/wildlife_degrees.xls (visited Dec 15, 2011). In contrast, of 99 conservation biology programs
listed by the main conservation biology professional society, the Society for Conservation
Biology that had accessible websites, 14 (14 percent) were in wildlife or natural resource
departments, 40 (40 percent) were in biology departments, 19 (19 percent) were in
environmental studies or sciences departments, and 21 (21 percent) were in ecology or zoology
departments. Society for Conservation Biology, University Program Database, online at http://
www.conbio.org/Resources/Programs/Search/ps.cfm (http://www.conbio.org/Resources/Programs/Search/ps.cfm) (visited Dec 15, 2011) (search keyword
“conservation biology”). See also William J. Matter and Robert J. Steidl, University
Undergraduate Curricula in Wildlife: Beyond 2000, 28 Wildl Socy Bull 503, 505 (2000) (“Most
wildlife programs are housed in larger academic units that offer other natural resource
curricula such as fishery, forestry, rangeland, and watershed science.”).
122 Noss, 22 Ecography at 116 (cited in note 100); Temple, et al, What’s So New about
Conservation Biology? at 610–11 & table 1 (cited in note 97). A review of the faculty
backgrounds from 25 conservation biology and 25 wildlife management programs (randomly
selected from the lists provided by TWS and SCB) found that of 398 wildlife management
professors, 38 (10 percent) had PhD degrees in wildlife science or natural resources, 73
(18 percent) in ecology or zoology, 182 (46 percent) in biology, 8 (2 percent) in environmental
science and 97 (24 percent) did not fall in any of those categories, while for 529 conservation
biology professors, 90 (15 percent) had PhD degrees in wildlife science or natural resources,
153 (32 percent) in ecology or zoology, 118 (22 percent) in biology, 8 (2 percent) in
environmental studies, and 160 (30 percent) in other categories.
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
500 The University of Chicago Law Review [79:471
biology society—Society for Conservation Biology (SCB)—show
striking differences in subject-matter and author affiliations.123 Those
differences are also replicated in the membership of the professional
societies that publish the journals.124
The comparison of conservation biology and wildlife
management shows how two fields that study very similar subject
matter areas can nonetheless have important distinctions based on
intellectual history and patterns of communication. But the
comparison also demonstrates the challenges of distinguishing
among disciplines, as shown by the data on the educational
backgrounds of faculty members in the respective programs: the two
fields have converged to some extent in terms of perspective,
(primarily as wildlife managers have become more like conservation
biologists and as communication between the disciplines has
increased) and the lack of formal departmental identification for
conservation biology (almost always) and wildlife management
(frequently) makes line drawing more difficult. In Part III, I will
draw on both the possibilities and challenges of distinguishing among
disciplines when I discuss the use of scientific disciplines as legal- or
institutional-design tools.
123 See, for example, Jacobson, 4 Conservation Bio at 432 (cited in note 96) (concluding
that the Journal of Wildlife Management (the leading TWS publication) is dominated by
articles about game species and that relatively few Conservation Biology (the leading SCB
publication) articles are authored by researchers with wildlife manager affiliations); Temple, et
al, What’s So New about Conservation Biology? at 610 (cited in note 97) (reporting that “a
review of the 889 papers published in The Journal of Wildlife Management (1982–87) revealed
that only 75 (8 percent) were on topics that would be appropriate for publication in
Conservation Biology”); Slack and Silvy, Have the Wildlife Society’s Publications Kept Pace
with the Profession? at 167–70 (cited in note 97) (concluding after a study of wildlife
management journals from the 1930s through 1989 that a vast majority of articles are on game
animals, and up to 40 percent of articles at times are on deer and waterfowl); Jensen and
Krausman, 21 Wildl Socy Bull at 200 (cited in note 97) (finding Conservation Biology authors
are more academic, much more likely to come from “basic” (biology) departments, and much
more unlikely to come from wildlife departments); Powell, et al, 74 J Wildl Mgmt at 816, 819–
23 (cited in note 97) (noting that despite a rise in conservation, nongame, and multiple species
studies in wildlife journals, those types of studies are still a minority, and few articles study
endangered species or invertebrates). But see Thomas and Pletscher, 28 Wildl Socy Bull at 547
(cited in note 106) (“Further, to scientists who have published in various journals, it seems
obvious that often articles published in one journal could have just as readily appeared in
another.”); Noss, 22 Ecography at 120 (cited in note 100).
124 There is significant overlap among the members of TWS, SCB, and the ESA, the main
professional society for ecologists. See Thomas and Pletscher, 28 Wildl Socy Bull at 547 (cited
in note 106). However, a study of membership patterns found that many more ESA members
are also members of SCB than of TWS. David M. Lawrence, Marjorie M. Holland, and
Deborah J. Morrin, Profiles of Ecologists: Results of a Survey of the Membership of the
Ecological Society of America: Part I: A Snapshot of Respondents, 74 Bull Ecol Socy Am 21, 31
table 14 (1993).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 501

D. How Disciplinary Structures Shape the Production of
Scientific Information
1. The importance of disciplinary perspectives.
The structure of disciplines within science is important because
each discipline brings very different perspectives to scientific work.125
(By perspectives, I mean attitudes about both how to describe and
analyze the natural world, and how to value the natural world.) Those
perspectives—sometimes called “paradigms,” or “worldviews,”126—will
shape (and in turn can be shaped by) the methodological tools used by
a discipline, the interpretations of ambiguous data, the kinds of
problems the discipline investigates, and the goals for research.127 The
contrasting viewpoints of marine ecologists and fisheries scientists
toward the use and reliability of data demonstrate how important
perspectives are in producing information.
Perspectives exist in part because they allow for easy
communication within a discipline, “since all the fundamentals are
agreed upon: what is a problem, what is a solution, what standards of
125 For a call for legal scholars to be more attentive to the ways in which values and
perspectives in environmental science shape environmental law and policy, see William Boyd,
Ways of Seeing in Environmental Law: How Deforestation Became an Object of Climate
Governance, 37 Ecol L Q 843, 849–51, 916 (2010).
126 These terms were popularized by the work of Thomas Kuhn. See Kuhn, The Structure
of Scientific Revolutions at x, 5–6, 111 (cited in note 95). In referencing the terminology, I do
not endorse all of the statements by Kuhn in that work (or the interpretations of those
statements) that paradigms might be “incommensurable” or that there is no basis for judging
the truth or falsity of paradigms based on facts from the natural world. Laudan, Science and
Relativism at 121–45 (cited in note 41) (noting adoption of incommensurability as justification
for a radical relativist critique of science, and criticisms of the concept of incommensurability).
My goal is much more limited. In the context of the complexity and dynamism of
environmental science, the facts available to us from the natural world are more incomplete
and less determinative of the truth or falsity of different hypotheses or perspectives. See
notes 15–19 and accompanying text. Accordingly, in the context of environmental science,
paradigms and perspectives will have greater role to play in determining the ways in which
science develops information than in many other scientific areas. That does not mean that the
natural world places no constraints on the development of environmental science, providing no
basis for comparing different disciplines.
127 Crane, Invisible Colleges at 32, 136–37 (cited in note 95) (noting contingency as to
whether a particular test can be seen as proof, depending on the intellectual environment of
the relevant discipline); David Robbins and Ron Johnston, The Role of Cognitive and
Occupational Differentiation in Scientific Controversies, 6 Soc Stud Sci 349, 353 (1976) (noting
importance of a theoretical framework for shaping facts and their interpretation within a
discipline); John Law, The Development of Specialties in Science: The Case of X-ray Protein
Crystallography, in Lemaine, et al, eds, Perspectives 123, 124 (cited in note 107) (arguing that
disciplines are based on paradigms, which constrain which problems can be worked on and
which cannot); Doremus, 75 Wash U L Q at 1066 (cited in note 55); Carolan, 34 Critic
Sociology at 728 (cited in note 44).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
502 The University of Chicago Law Review [79:471
accuracy are appropriate, what techniques may and may not be used,
what journals and text books are worth reading and what can safely
be left on the shelf, and so on.”128 For instance, disciplines will
develop “black boxes” of theories or claims that are presumed to
represent truth.129 These black boxes become the foundations for
future intellectual progress within that field, by allowing scientists to
focus on particular research questions, instead of questioning all of
the necessary assumptions for any research to proceed.130 But they
also necessarily cabin to some extent the range of inquiry.
Until the 1980s, a dominant perspective in the field of ecology
was an understanding of the natural world as governed by relatively
simple models of the interactions among species, interactions that
produced relatively stable outcomes in the natural environment.131
Predator and prey species would interact in ways that would
maintain each other’s population levels within certain levels; a forest
that had been cleared for agriculture or by a fire would eventually
return to a “climax” state dominated by specific species. The
underpinnings of this perspective were a series of assumptions about
how the natural world operated and how it could be modeled. Those
assumptions had not, in fact, been tested with real-world data more
than a few times (and those few tests had found the assumptions
wanting), but ecologists did not question the underlying assumptions
because the models and general perspective were essential for
pursuing productive field work.132 Only when accumulating field data
contradicted the predictions of the “balance of nature” perspective
did ecologists begin to question those assumptions.133
The role that perspectives play in shaping knowledge is not
comprehensive or even necessarily dominant; perspectives change in
the face of the accumulation of evidence from experiments and
128 David Collingridge and Colin Reeve, Science Speaks to Power: The Role of Experts in
Policy Making 19 (St Martin’s 1986).
129 See Sheila Jasanoff, Transparency in Public Science: Purposes, Reasons, Limits, 69 L &
Contemp Probs 21, 35, 37 (2006); Carolan, 34 Critic Sociology at 729 (cited in note 44);
Oreskes, 7 Envir Sci & Pol at 370–72, 380 (cited in note 45).
130 See Kitcher, Advancement of Science at 80–82, 84–89 (cited in note 116). Variation
within a discipline on important issues will nonetheless be significant and may be important for
the development of new scientific knowledge. Id at 82 n 34, 306–08.
131 See Daniel B. Botkin, Discordant Harmonies: A New Ecology for the Twenty-First
Century 27–49 (Oxford 1990).
132 Id at 33, 41–42.
133 Id at 37–41 (describing two studies that called into question the assumption of
stability).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 503

debates from within or outside the discipline.134 The natural world
will constrain the types of perspectives that can achieve consensus
within a discipline.135 But when it comes to interpreting and
understanding ambiguous data and managing uncertainty, “[w]here
reasonably possible, scientists tend to interpret their observations as
consistent with whatever theory currently commands the most
adherents, even if other interpretations are equally or even more
plausible.”136
These perspectives will often be influenced by the values of the
members of the discipline.137 The controversy over the global rate of
extinction of species is an example of how values shape perspectives
and how perspectives shape the production of information in
conservation biology. There is no question that humans have
dramatically shaped the world in the past century and a half.
However, for conservation biologists, a key question is the
implications for global biodiversity: How many species have gone
extinct or will go extinct if current trends continue?
The problem is (again) that there are extremely limited data
about even the total number of species on the planet, let alone how
many of those species have gone extinct as a result of human
activities (and will go extinct if present trends continue).138 The
question becomes even more complicated as conservation biologists
attempt to understand how climate change might accelerate the loss
of species over the next century.
To answer this question, conservation biologists have relied
heavily on an equation that relates the area of intact habitat to the
number of species that will persist over time: more intact habitat
means more species persist.139 This “species-area curve” analysis was
134 See generally Naomi Oreskes, The Rejection of Continental Drift (Oxford 1999)
(describing the decades-long process by which argument and evidence accumulated in geology
leading to a consensus around the theory of continental drift).
135 See Ziman, Real Science at 197–98 (cited in note 70).
136 Doremus, 75 Wash U L Q at 1066, 1068 (cited in note 55).
137 Sarewitz, 7 Envir Sci & Pol at 392 (cited in note 44) (“Even the most apparently
apolitical, disinterested scientist may, by virtue of disciplinary orientation, view the world in a
way that is more amenable to some value systems than others.”); Michael S. Carolan, Scientific
Knowledge and Environmental Policy: Why Science Needs Values, 3 Envir Sci 229, 230 (2006)
(“In the end, environmental science rests, at least in part, on normative assumptions.”).
138 See Paul Voosen, Scientists Clash on Claims over Extinction ‘Overestimates,’ NY Times
(May 18, 2011), online at http://www.nytimes.com/gwire/2011/05/18/18greenwire-scientistsclash-
on-claims-over-extinction-ove-96307.html?pagewanted=all (visited Dec 15, 2011)
(describing the historic and renewed controversy over the uncertainty inherent in the
calculations scientists use to estimate extinction).
139 See, for example, Stuart L. Pimm and Robert A. Askins, Forest Losses Predict Bird
Extinctions in Eastern North America, 92 Proc Natl Acad Sci 9343, 9343 (1995).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
504 The University of Chicago Law Review [79:471
used to estimate how many species might go extinct if deforestation
rates in locations such as tropical rainforests continued.140 It has since
been extended to assess the implications of climate change for
biodiversity. “Climate envelope” models use global-scale models
estimating how climate will change across the planet in the next
decades or century. They then compare the changed climatic regime
to existing climate patterns to determine how much suitable habitat
is gained or lost; reductions in habitat are then used (via the
“species-area curve”) to estimate how many species might go extinct
as a result of climate change.141
“Climate envelope” models have proved quite controversial. A
“climate envelope” study in 2004 tentatively predicted that 15 to
37 percent of global species would be “committed to extinction” as a
result of climate change by 2050.142 Critics challenged the significant
assumptions entailed in the use of such models, such as the inability
of species to adapt to new climates, or the relationship between
climate and current distributions of species.143 Critics also noted that
the use of “climate envelope” models often assumes the worst-case
scenario in terms of impacts of climate change on species.144
But even the fundamental relationship between species numbers
and area has been challenged. Despite the relationship’s widespread
use by conservation biology, critics have argued that scientists still do
not fully understand why the relationship exists.145 Recent research
140 See, for example, Thomas M. Brooks, Stuart L. Pimm, and Joseph O. Oyugi, Time Lag
between Deforestation and Bird Extinction in Tropical Forest Fragments, 13 Conservation
Bio 1140, 1141 (1999).
141 See, for example, John W. Williams, Stephen T. Jackson, and John E. Kutzbach,
Projected Distributions of Novel and Disappearing Climates by 2100 AD, 104 Proc Natl Acad
Sci 5738, 5739 (2007); Wilfried Thuiller, et al, Climate Change Threats to Plant Diversity in
Europe, 102 Proc Natl Acad Sci 8245, 8247 (2005); Chris D. Thomas, et al, Extinction Risk
from Climate Change, 427 Nature 145, 145 (2004).
142 See Thomas, et al, 427 Nature at 145 (cited in note 141).
143 See, for example, Owen T. Lewis, Climate Change, Species-Area Curves, and the
Extinction Crisis, 361 Phil Trans Royal Socy Bull 163, 167–68 (2006); Daniel Botkin, et al,
Forecasting the Effects of Global Warming on Biodiversity, 57 BioSci 227, 230–31 (2007).
144 See Lewis, 361 Phil Trans Royal Socy Bull at 169 (cited in note 143); Botkin, et al,
57 BioSci at 228 (cited in note 143). But see Arndt Hampe, Bioclimate Envelope Models: What
They Detect and What They Hide, 13 Global Ecol & Biogeography 469, 470 (2004) (arguing
“climate envelope” models might underestimate the risks of climate change for biodiversity).
145 See Lewis, 361 Phil Trans Royal Socy Bull at 165–66 (cited in note 143) (describing the
species-area relationship approach as deceptively simple in that it conceals a number of
unfounded assumptions); Robert J. Whittaker, et al, Conservation Biogeography: Assessment
and Prospect, 11 Diversity & Distributions 3, 12 (2005); Stephen Budiansky, Extinction or
Miscalculation?, 370 Nature 105, 105 (1994); Voosen, Scientists Clash, NY Times (cited in
note 138) (reporting that “[f]or decades, it has been an open secret among conservationists. An
elegant equation widely used to calculate how many species will go extinct from deforestation
and habitat destruction—one of the ‘laws’ of ecological theory—was a little shaky”).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 505

has questioned whether the relationship systematically overstates the
impact of habitat destruction on species survival rates,146 though that
research has itself been criticized.147
My point here is not to resolve these disputes. Instead, I want to
note that the significant reliance of conservation biologists on these
tools, despite their acknowledged weaknesses, is the outcome of a
discipline that places a strong value on biodiversity protection. There
are other tools to assess the impacts of climate change and habitat
destruction on biodiversity,148 but those tools require tremendous
amounts of data—data that exist only for a fraction of species.149 If
conservation biologists limited themselves to those other tools, they
might make more precise estimates of the risks of species extinction
for particular species, but they would be unable to make large-scale
(in space and time) statements about the impacts of human activities
on global biodiversity.150 Such statements are crucial for a discipline
that is focused on addressing the impacts of human activities on
biodiversity at a global scale.151 This is particularly true for a
discipline that sees itself as responding to a “crisis” and therefore
must produce information that is at a large enough scale that it can
be useful for policy makers in decision making in the very near
146 See Fangliang He and Stephen P. Hubbell, Species-Area Relationships Always
Overestimate Extinction Rates from Habitat Loss, 473 Nature 368, 368 (2011).
147 See Voosen, Scientists Clash, NY Times (cited in note 138); Stuart Pimm, This Week’s
Claim that the Species Extinction Crisis Is Overblown Is a Sham, Natl Geo News Watch
(May 21, 2011), online at http://newswatch.nationalgeographic.com/2011/05/21/this-week%E2
%80%99s-claim-that-the-species-extinction-crisis-is-overblown-is-a-sham/ (visited Dec 15,
2011); Bob O’Hara, Species-Area Relationships Don’t Overestimate Extinction Rates from
Habitat Loss, Deep Thoughts and Silliness Blog (Nature Publishing Group 2011), online at
http://blogs.nature.com/boboh/2011/05/31/species-area-relationships-dont-overestimateextinction-
rates-from-habitat-loss (visited Nov 21, 2011) (providing a step-by-step analysis of
He and Hubbell’s conclusions to argue that they are incorrect); T.M. Brooks, Extinctions:
Consider All Species, 474 Nature 284, 284 (2011).
148 See, for example, Botkin, et al, 57 BioSci at 230–32 (cited in note 143) (describing
alternative models).
149 See Richard G. Pearson and Terence P. Dawson, Predicting the Impacts of Climate
Change on the Distribution of Species: Are Bioclimate Envelope Models Useful?, 12 Global
Ecol & Biogeography 361, 366–67 (2003) (noting that “climate envelope” models are much
more useful for large numbers of species with limited data).
150 See Botkin, et al, 57 BioSci at 230 (cited in note 143); Lewis, 361 Phil Trans Royal
Socy Bull at 165–66, 169–70 (cited in note 143); Antoine Guisan and Wilfried Thuiller,
Predicting Species Distribution: Offering More Than Simple Habitat Models, 8 Ecol Let 993,
1004 (2005).
151 Thus, even those scientists who have questioned climate envelope and species-area
curve models have emphasized that they do so out of a desire to ensure that more accurate
information is available, both to improve conservation management and to reduce the risk that
conservation biologists are seen as “crying wolf.” See He and Hubbell, 473 Nature at 370 (cited
in note 146); Lewis, 361 Phil Trans Royal Socy Bull at 164, 170 (cited in note 143).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
506 The University of Chicago Law Review [79:471
future (and, ideally, inspire policy makers to action to protect
biodiversity in the very near future).152 Conservation biologists are
comfortable accepting the uncertainty in these kinds of estimates for
those reasons.153 The discipline therefore relied on the species-area
curve as a tool (and to some extent put it in a black box) despite its
inevitable weaknesses, based in part on values (the importance of the
protection of biodiversity) that contribute to the overall perspective
of the discipline (understanding the status and trends of biodiversity
as the primary goal of research).

[أحمد أبو زنط]

2. How disciplines can shape the nature and content of their
own perspectives.
Disciplinary perspectives are shaped both by the discipline itself
and by forces outside the discipline. Perhaps the most important
internal constraint is the training of scientists while graduate
students, which acculturates them into the expectations of the
discipline, including its fundamental precepts.154 Even after training is
completed, social or peer pressure from colleagues within a
discipline (particularly leading figures) can deter scientists from
publishing results that are inconsistent with the dominant
perspective—particularly through peer-review gatekeeping
mechanisms for publication, funding, hiring, and tenure decisions.155
Positive peer reputation may be the most important factor in
152 For more on conservation biology as a crisis discipline, see, for example, Meine, Soulé,
and Noss, 20 Conservation Bio at 635 (cited in note 120) (providing a brief historical overview
of conservation biology’s characterization a crisis discipline); Soulé, 35 BioSci at 727 (cited in
note 100) (explaining why conservation biology is a crisis discipline). For more on the
importance of generalizability, see, for example, Noss, 22 Ecography at 116 (cited in note 100)
(arguing that “because case-specific information is always poorer than desired for conservation
planning, empirical generalizations derived from prior experience often provide the only
reasonable foundation for decision-making”).
153 Soulé, 35 BioSci at 727 (cited in note 100) (“In crisis disciplines, one must act before
knowing all the facts . . . . A conservation biologist may have to make decisions or
recommendations about design and management before he or she is completely comfortable
with the theoretical and empirical bases of the analysis. Tolerating uncertainty is often
necessary.”).
154 Cyrus C.M. Mody and David Kaiser, Scientific Training and the Creation of Scientific
Knowledge, in Edward J. Hackett, et al, eds, The Handbook of Science and Technology Studies
377, 377–85 (MIT 3d ed 2008); H.M. Collins, Changing Order: Replication and Induction in
Scientific Practice 142 (SAGE 1985); Hagstrom, The Scientific Community at 9–10 (cited in
note 95).
155 See Steven P. Feldman, The Culture of Objectivity: Quantification, Uncertainty, and the
Evaluation of Risk at NASA, 57 Hum Rel 691, 697 (2004); Mulkay, Gilbert, and Woolgar,
9 Sociology at 195 (cited in note 109); Sheila Jasanoff, Science, Politics, and the Renegotiation
of Expertise at EPA, 7 Osiris 194, 196 (1992); Doremus, 75 Wash U L Q at 1062 & n 181 (cited
in note 55).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 507

professional advancement.156 For instance, biologists who took the
unpopular position of criticizing the species designation of a sea
turtle population “felt considerable pressure to downplay or
‘reinterpret’” the relevant data from “members of the international
sea turtle research community” including “journal editors that were
deciding the fate of [the biologists’] papers and program officers
deciding the fate of [their] grants.”157
Training and peer pressure can help perpetuate a discipline’s
perspective. Other internal factors might shape the content of that
perspective. For instance, the methodological choices of a discipline
may significantly affect the perspectives of the discipline.
Meteorologists regularly work with complicated computer models to
attempt to forecast weather at the local and regional level—
producing predictions that can be quite unreliable.158 As a discipline,
meteorologists are all too aware of the flaws and foibles of models as
forecasting tools. It is probably no surprise that meteorologists have
been skeptical of the ability of global climate models to provide
reliable predictions of future climate impacts from increases in
atmospheric greenhouse gases.159 On the other hand, the global
nature of the information and the methods that most climate
scientists use frame the policy questions that climate scientists
believe are relevant to their work and create a focus on
international, rather than local or domestic, solutions.160
There may be significant self-selection mechanisms by which
perspectives are developed and maintained—scientists may enter
into the study of a particular field because of personal preferences
(for example, people who love birds may become ornithologists)—
that in turn will inform their use of judgments about ambiguous
data.161 As one commentator has put it:
156 Hagstrom, The Scientific Community at 29–33 (cited in note 95); Mulkay, Gilbert, and
Woolgar, 9 Sociology at 195 (cited in note 109).
157 Brian W. Bowen and Stephen A. Karl, In War, Truth Is the First Casualty, 13 Conservation
Bio 1013, 1013–14 (1999).
158 This unreliability has led to an abundance of jokes about the predictive abilities of
weather forecasters.
159 See text accompanying notes 1–3.
160 See, for example, Boyd, 37 Ecol L Q at 866–68 (cited in note 125); Paul N. Edwards,
Representing the Global Atmosphere: Computer Models, Data, and Knowledge about Climate
Change, in Clark A. Miller and Paul N. Edwards, eds, Changing the Atmosphere: Expert
Knowledge and Environmental Governance 31, 33 (MIT 2001); Simon Shackley, Epistemic
Lifestyles in Climate Change Modeling, in Miller and Edwards, eds, Changing the Atmosphere 107,
108–13 (MIT 2001).
161 Doremus, 75 Wash U L Q at 1073 (cited in note 55) (noting that intuitions and
judgments of scientists may “be strongly influenced by subjective preferences”).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
508 The University of Chicago Law Review [79:471
When it comes to evaluating conflicting evidence, people tend
to trust evidence of the kind which they and their close
colleagues have dedicated their lives to obtaining, in part for
social reasons, and in part because they have an intellectual,
aesthetic, or ethical affinity for that kind of scientific work,
which helps to explain why they chose to pursue that kind of
research in the first place. Often these commitments are both
affective and epistemic. Field scientists like field work—they
like being out in the fresh air and sunshine—and they also
believe it to be more likely to capture the basic truths about the
natural world, messy though it may be. In contrast, laboratory
scientists enjoy working in the lab—they enjoy building and
tinkering—and they also believe it to produce knowledge of
greater specificity and rigor than field science. . . . Scientists may
also choose a particular line of inquiry because it aligns with
their normative commitments: (field biologists caring about
nature, economists caring about the efficient management of
monetary resources), in which case they are apt to defend their
work strongly on (implicit) normative grounds.162
An example of these factors at work in shaping a discipline’s
perspective is the history of conservation biology.163 The field’s
creation was part of a concerted effort by scientists concerned about
the disappearance of species and ecosystems to increase their
traction in the policy arena. For instance, the term “biodiversity” was
created by scientists interested in protecting species, ecosystems, and
other natural living resources. Their goal was to increase their
political leverage to shape policy by emphasizing the importance of
their expertise in making decisions about what should be protected
and how to protect it.164 It was first broadly used as part of the
“National Forum on Biodiversity” in 1986, a political event intended
to raise the profile of species conservation, and the organizers of the
forum specifically coined the term as a “glitzy” one that could
advance the political mission of conservation biology.165
The importance of internal factors means that science is often a
very peer-driven enterprise.166 Journals and government grant
162 Oreskes, 7 Envir Sci & Pol at 375 n 7 (cited in note 45).
163 Takacs, The Idea of Biodiversity at 35 (cited in note 54).
164 Id at 1, 5–8.
165 Id at 37–38.
166 See Nagi and Corwin, The Research Enterprise at 22 (cited in note 113); Law, The
Development of Specialties in Science at 125 (cited in note 127).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 509

funding are peer reviewed and disciplinary based,167 and disciplinary
associations are peer run. The collective imprimatur of scientists in a
field is what ultimately determines what is “knowledge” and “truth”
within the field, what work is valid, which scientists should be
praised.168
The peer-driven nature of disciplines is an important basis for
the public legitimacy and political power of science.169 As a
decentralized institution or social structure—a characteristic
emphasized by disciplinary fragmentation—science is hard for
outsiders to directly control.170 The legitimacy of science when it is
deployed in the political arena comes, in part, from the perception by
the public that scientific information is often difficult for political
actors to shape to achieve political goals,171 and therefore as more
likely to be authentic and trustworthy.172 Thus, skeptics of climate
science regularly attempt to discredit the field by pointing to
government funding of climate scientists. The contention is that the
government agencies use the threat of climate change and the need
for additional research to justify larger budgets from Congress, and
167 The National Science Foundation grant-making process gives scientists a dominant
role. See Steven Goldberg, Culture Clash: Law and Science in America 44–49 (NYU 1994).
This peer-review process is usually organized by discipline. See Jasanoff, The Fifth Branch
at 65–68 (cited in note 22).
168 Fleck, Genesis and Development at 118–25 (cited in note 95); Joseph Rouse,
Knowledge and Power: Toward a Political Philosophy of Science 120 (Cornell 1987) (stating
that “[i]n science, the standards of rational acceptability are not individual but social”); Kuhn,
The Structure of Scientific Revolutions at 168 (cited in note 95); Hagstrom, The Scientific
Community at 12–24 (cited in note 95).
While the peer-driven nature of scientific disciplines to some extent imposes conformity on
scientists, there is a countervailing pressure—the dramatic rewards provided to those who
develop new, original understandings of the world that become widely accepted within the
discipline. See, for example, Ziman, Real Science at 40–41, 182–85 (cited in note 70). This
pressure—and the desire of scientists for fame and recognition—might be an important factor
driving the development of scientific information, including the development of challenges to
scientific orthodoxy. See Kitcher, The Advancement of Science at 72–74 (cited in note 116).
169 See Pielke, The Honest Broker at 93 (cited in note 43).
170 See Daryl E. Chubin and Edward J. Hackett, Peerless Science: Peer Review and U.S.
Science Policy 6 (SUNY 1990). But see Ziman, Real Science at 74–76, 177–81 (cited in note 70).
171 Sheldon Krimsky, Publication Bias, Data Ownership, and the Funding Effect in Science:
Threats to the Integrity of Biomedical Research, in Wagner and Steinzor, eds, Rescuing Science
from Politics 61, 67 (cited in note 47); Pielke, The Honest Broker at 88–89 (cited in note 43).
172 See Jasanoff, The Fifth Branch at 208 (cited in note 22) (“Knowledge generated by
parties without a clear stake in its application might indeed be more resistant to the
deconstructive pressures of U.S. regulation.”); Fein, Comment, 99 Cal L Rev at 555 (cited in
note 4).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
510 The University of Chicago Law Review [79:471
the scientists exaggerate the risks of climate change to justify
additional research that requires funding for themselves.173
3. How outside institutions can shape disciplinary perspectives.
Outside pressures shape both the creation of disciplines and
their perspectives. Disciplines may be created or supported by
institutions in order to increase their profile and improve their ability
to obtain resources.174 For instance, American “land grant”
universities focused on the “services” they could provide to the local
economy and society and so focused on “practical” disciplines such
as agricultural sciences.175 Those external forces will also shape the
information produced by those disciplines they have created or
nurtured by supporting work in some directions and deterring work
in others.176 In the development of biochemistry in the United States,
the dominance of land grant universities meant that there was a
focus on “applied” science.177 Later, as biochemistry moved to
medical schools, the focus changed to supporting clinical work and
medical school service teaching and on research that could support
medical practice.178
In the context of environmental science, industry is the main
client for much of resource science such as forestry or fisheries
science,179 and that dominance might shape the values of those
disciplines, which tend to assume that resources are to be used.180
Foresters “use terms such as ‘tree crops,’ draw an analogy between
farming and forestry, and often claim that undisturbed forests will
173 See, for example, Patrick J. Michaels, Meltdown: The Predictable Distortion of Global
Warming by Scientists, Politicians, and the Media 221–35 (Cato 2004). I do not endorse this
rather cynical view of climate science.
174 See Kohler, From Medical Chemistry at 7, 15–16, 211 (cited in note 94); Joseph Ben-
David and Randall Collins, Social Factors in the Origins of a New Science: The Case of
Psychology, 31 Am Sociological Rev 451, 458–61 (1966) (describing how institutional
structures encouraged the development of psychology in Germany); Charles E. Rosenberg,
Factors in the Development of Genetics in the United States: Some Suggestions, 22 J Hist
Med 27, 28–29 (1967) (arguing that social and institutional context helps explain how genetics
developed as a formal discipline in the United States); Joseph Ben-David, Scientific Growth: A
Sociological View, 2 Minerva 455, 469–76 (1964).
175 Kohler, From Medical Chemistry at 4–5 (cited in note 94).
176 Schuck, 11 Yale L & Pol Rev at 16 (cited in note 64); Kohler, From Medical Chemistry
at 324 (cited in note 94) (arguing that “[p]articular scientific styles flourish only where
intellectual priorities are congruent with institutional structures and goals”); Robbins and
Johnston, 6 Soc Stud Sci at 354 (cited in note 127).
177 Kohler, From Medical Chemistry at 96–97 (cited in note 94).
178 Id at 214, 216–17.
179 Bocking, Nature’s Experts at 85–87 (cited in note 53).
180 Id at 86.
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 511

‘deteriorate’—that is, become less productive.”181 Likewise, the
mission of public agencies that are major clients of a discipline can
also have a significant impact on a discipline’s development.
Wildlife management provides an excellent example of how
outside factors can shape a discipline’s perspective. The field of
wildlife management developed in the 1930s, with its own journals,
societies, and departments in universities.182 A primary impetus for its
creation was the rise of state and federal agencies, such as the new
federal Bureau of Biological Survey—the predecessor to the United
States Fish and Wildlife Service (FWS)—that sought to restrict
hunting and increase game populations.183 The new federal agency in
particular needed tools to monitor, count, and predict wildlife
populations so that it could set quotas for duck and waterfowl
hunting under a new regulatory system.184 The new discipline could
provide technical support for these goals, but perhaps even more
important, “[t]he mantle of science gave game managers
legitimacy.”185 Accordingly, the agency supported the development of
the discipline: the Biological Survey encouraged its employees to join
the new Wildlife Society, and its employees were 10 percent of the
new organization’s members.186
Given this institutional context, it is no surprise that the
discipline was heavily oriented in a utilitarian, resource-maximizing
direction. One of the founders of the field, Aldo Leopold, defined
game management as “the art of making land produce sustained
annual crops of wild game for recreational use” and the field was
modeled on other “land-cropping arts” such as forestry and
agricultural sciences.187 Wildlife management heavily relied on the
181 Id.
182 Thomas R. Dunlap, Saving America’s Wildlife 76, 78 (Princeton 1988) (describing the
development of game management as a separate academic discipline from ecology, noting that
TWS was founded in 1936, with the Journal of Wildlife Management first published two years
later in 1938); Thomas R. Dunlap, Organization and Wildlife Preservation: The Case of the
Whooping Crane in North America, 21 Soc Stud Sci 197, 200–01 (1991); Samuel P. Hays,
Beauty, Health, and Permanence: Environmental Politics in the United States, 1955–1985 19
(Cambridge 1987).
183 Dunlap, Saving America’s Wildlife at 35–40 (cited in note 182); Hays, Beauty, Health,
and Permanence at 19–20 (cited in note 182).
184 See Dunlap, Saving America’s Wildlife at 37–38 (cited in note 182); Dunlap, 21 Soc
Stud Sci at 200 (cited in note 182).
185 Dunlap, Saving America’s Wildlife at 76 (cited in note 182).
186 Id at 78.
187 Id at 70–71, 76–77, citing Aldo Leopold, Game Management 3 (Scribner’s 1933). See
also Dunlap, Organization and Wildlife Preservation, 21 Soc Stud Sci at 201 (cited in note 182);
Hays, Beauty, Health, and Permanence at 21 (cited in note 182); R. Ben Peyton, Wildlife
Management: Cropping to Manage or Managing to Crop?, 28 Wildl Socy Bull 774, 776 (2000).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
512 The University of Chicago Law Review [79:471
paradigms of “maximum production, efficiency, and expert
management” that had dominated forestry.188 The applied,
agricultural focus of many American land grant universities also
provided a welcoming home for such a utilitarian discipline.189
From the 1930s onward, academic wildlife science programs
“worked hand-in-hand with agencies” to advance the dominant
paradigm of managing wildlife populations for hunting and
harvesting.190 State wildlife agencies still hire the largest share of
wildlife management program graduates, with federal agencies a
close second, and together the two comprise the majority of
employment opportunities.191 Wildlife scientists regularly note (and
sometimes bemoan) the heavy influence of these agencies on
curricula and research programs.192

[أحمد أبو زنط]

III. EXPLORING THE ROLE OF DISCIPLINES IN SHAPING
ENVIRONMENTAL LAW
If the perspectives of disciplines are so integral to the
development of information in environmental science, and if those
perspectives interact with the governance institutions and politics of
environmental law, then separating the “science” and the “policy” in
environmental law is a challenging task. But what if we utilized the
intertwinement of science and policy in the world of environmental
law?
188 Dunlap, Saving America’s Wildlife at 76–77 (cited in note 182).
189 Id at 77.
190 John F. Organ and Eric K. Fritzell, Trends in Consumptive Recreation and the Wildlife
Profession, 28 Wildl Socy Bull 780, 781 (2000). See also Peyton, 28 Wildl Socy Bull at 774
(cited in note 187).
191 Robert M. Muth, et al, Passing the Torch of Wildlife and Fisheries Management:
Comparing the Attitudes and Values of Younger and Older Conservation Professionals, Trans
77th N Am Wildl and Nat Res Conf 178, 183–84 (Wildlife Management Institute 2002).
192 See, for example, Brown and Nielsen, 28 Wildl Socy Bull at 499 (cited in note 106);
Gavin, 17 Wildl Socy Bull at 348–49 (cited in note 118); J. Michael Scott, et al, Conservation of
Biological Diversity: Perspectives and the Future for the Wildlife Profession, 23 Wildl Socy
Bull 646, 650 (1995); James M. Peek, A Look at Wildlife Education in the United States, 17
Wildl Socy Bull 361, 361 (1989); Daniel M. Keppie, To Improve Graduate Student Research in
Wildlife Education, 18 Wildl Socy Bull 453, 455 (1990); Brown and Nielsen, 28 Wildl Socy Bull
at 496 (cited in note 106).
My discussion of the underlying values of the fields of conservation biology and wildlife
management, and how those values shape the scientific output of those fields, is not a critique
of those fields as scientific disciplines or of the scientists in those disciplines. Society (or at least
important public and private organizations) decided that protection of endangered species and
increasing the availability of game for sport hunters were both important social goals. These
fields developed to serve those goals, and they have often performed well in achieving those
goals.
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 513

Here I develop this possibility and how it might improve the
relationship between science and law in the environmental context.
By explicitly recognizing the ways in which value choices infuse
various disciplines in environmental law, we can design our legal and
institutional structures to take into account the disciplinary
perspectives: either to embrace them (when those perspectives are
thought to advance our overall policy goals in the area) or to offset
them (when they instead interfere with our policy goals).193
Alternatively, we can choose to rely on a more diverse range of
perspectives in order to reduce the bias that might come from one
particular discipline, if we believe that such bias is improper. The
choice between embracing, offsetting, or balancing between different
disciplines will often depend on whether the legal or institutional
designer seeks to prioritize achieving a particular policy goal at the
expense of a greater risk of lower quality information. Regardless of
the specific design choice, being cognizant of the importance of
disciplinary perspectives allows a potentially simpler way of
understanding how policy and science frequently interact than trying,
for every regulatory decision, to identify every assumption and
inference and the implicit or explicit value choices that drive each of
them.
A. Using Scientific Disciplines as Legal- or Institutional-Design
Tools
Scientific disciplines can help us accomplish particular policy
goals in a range of ways. We might privilege one discipline over
another where both are seen as plausibly relevant for the policy
process. For instance, a fisheries management program might
privilege either the information produced by marine biologists or,
alternatively, the information produced by fisheries scientists,
depending on whether the policy goal prioritizes the protection of all
forms of marine biodiversity, or the assurance of some level of
human exploitation of particular fish stocks. But even when there
193 The concept of legal and institutional design is based on administrative law and
political-science scholarship exploring how agency structures and procedures might be used by
legislatures to shape implementation of a management or regulatory program by an
administrative agency. See Gersen, Designing Agencies at 339–42 (cited in note 6) (discussing
“structure and process” theory); Lisa Schultz Bressman, Procedures as Politics in
Administrative Law, 107 Colum L Rev 1749, 1785 (2007) (providing overview of literature);
Mathew D. McCubbins, Roger G. Noll, and Barry R. Weingast, Administrative Procedures as
Instruments of Political Control, 3 J L, Econ, & Org 243, 248–53 (1987) (developing the
concept of using administrative structure and process to solve principal–agent problems in
government agencies, and considered a seminal political science article).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
514 The University of Chicago Law Review [79:471
may be only one scientific discipline that is plausibly relevant,
privileging that discipline in the policy making process can shape the
information that is produced, and thus in turn the decisions that are
based on that information. I develop an example of this shortly when
I discuss the use of range science by the Forest Service.
In this Part, I explore four different choices for using scientific
disciplines as legal- and institutional-design tools: (1) using a
discipline to constrain administrative agency decision making,
(2) using a discipline to legitimate administrative agency decision
making, (3) insulating a discipline from administrative agency
decision making, or (4) balancing the role of multiple disciplines in
the decision-making process. These four examples help illuminate
the possible ways in which disciplines might be used, but they are not
exhaustive. My intent is also not to specifically endorse the use of
any of these particular tools in general; the desirability of any of
these options will depend on the particulars of any given regulatory
or management program. I conclude with a discussion of some of the
challenges to using specific scientific disciplines as legal- or
institutional-design tools, such as the fluidity of the concept of a
scientific discipline and the possibility that a discipline might change
its perspectives over time—though it is important to keep in mind
that these challenges may exist even if disciplines are not consciously
or explicitly considered in the design of regulatory or management
structures.
Throughout the following discussion, I draw on examples from
statutes, regulations, and case law. However, I want to emphasize
that my reliance on these examples is not necessarily intended to
demonstrate that Congress, agency policy makers, or courts
consciously chose to rely on disciplines as legal- or institutionaldesign
tools. Instead, the examples are intended to show how some
of the processes and possibilities that I explore might play out in the
real world.
1. Option one: constraining.
We might use a discipline as an ex ante constraint on the
implementation of an environmental statute by an administrative
agency, drawing on the peer-driven independence of the discipline.194
194 By ex ante, I mean that the legal or institutional designer (for example, the legislature)
sets up a system in advance (at the time the management or regulatory program is established)
to constrain how the administrative agency implements the program, rather than attempting to
control the agency through ongoing monitoring of the agency’s performance. See Gersen,
Designing Agencies at 334–42 (cited in note 6).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 515

As long as a nontrivial number of members of the discipline are
nongovernment employees, a direct mandate from the executive or
bureaucratic leaders to those scientists to adopt a particular position
on an issue is unlikely to succeed. Likewise, as long as at least some
of the funding for a discipline comes from outside the government, it
will be more difficult for the legislature to use its budgetary power to
impose a consensus on a particular question, unlike how an agency
might respond to similar sanctions.
Accordingly, if scientists from a particular discipline are given
some sort of voice in the decision-making process—such as
regulatory peer review, in which agency regulatory decisions are
reviewed by scientists before being implemented—then they can
provide effective ex ante control of bureaucracy.195 Another
possibility is that judicial review of agency decision making may pay
special attention to a particular scientific discipline in examining an
agency’s decision.
This role for scientific disciplines has appeal for two reasons:
(1) the tremendous need for flexibility in developing and responding
to new information in the dynamic world of environmental science,
and (2) the importance of using ex ante constraints to control the
risk of future slippage in the implementation of environmental
statutes by agencies.
Starting with the second point, slippage is particularly important
in environmental decision making because there are good reasons to
believe that psychological, economic, and political pressures will lead
us to systematically implement environmental policy in ways which
are socially suboptimal.196 Many commentators have noted that the
diffuse, subtle, and long-term nature of many environmental harms
makes environmental law particularly vulnerable to public choice
failures in the political process: the costs of organizing to achieve
environmental benefits may be high, but any one individual will
receive a relatively small share of the (socially quite large) overall
benefit that is distributed among most or all members of the public,
and free-riders cannot be excluded easily, if at all.197 On the other
side, those who pay for the environmental benefits will often (though
not always) be relatively small in number, and their costs will be
195 For a discussion of the ability of peer-reviewed science to act as a neutral ex ante control
on agency policies, see Stuart Shapiro and David Guston, Procedural Control of the Bureaucracy,
Peer Review, and Epistemic Drift, 17 J Pub Admin Rsrch & Theory 535, 543 (2006).
196 See Richard J. Lazarus, Super Wicked Problems and Climate Change: Restraining the
Present to Liberate the Future, 94 Cornell L Rev 1153, 1173–75 (2009).
197 Id at 1183; Richard J. Lazarus, The Making of Environmental Law 41 (Chicago 2004).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
516 The University of Chicago Law Review [79:471
much higher per capita then the per capita benefits from the
environmental amenities. As a result, there is a classic organizational
advantage for the opponents of most environmental regulation.198
The obstacles are exacerbated by the subtle and long-term
nature of environmental harms, which can take decades to manifest
themselves. Myopia often leads us to overvalue short-term benefits
and costs and undervalue long-term benefits and costs.199 Moreover,
the substantial uncertainty in environmental law helps encourage the
avoidance of costly short-term choices to avoid long-term
problems.200 Myopia and procrastination might endlessly prevent
individuals or communities from making the short-term sacrifices
necessary to achieve long-term goals, particularly where short-term
decisions not to sacrifice will result in “individually negligible”
harms, but the accumulation of procrastination over time will result
in “cumulatively devastating” outcomes.201
Some environmental protection statutes do get passed. The
wave of environmental legislation passed in the early-to-mid-1970s
has been explained either as a result of a “republican moment” in
environmental law in which deliberation by the public led to a focus
on the common good and political activism,202 or as a result of an
unusual convergence of public choice dynamics203—both explanations
rely on the catalyzing influence of high-profile environmental
catastrophes.204 Whatever the story, the problem of long-term
implementation still remains, even after the temporary conditions
198 See, for example, Steven P. Croley, Public Interested Regulation, 28 Fla St U L Rev 7,
35–38 (2000); Matthew D. Zinn, Policing Environmental Regulatory Enforcement:
Cooperation, Capture, and Citizen Suits, 21 Stan Envir L J 81, 126–31 (2002); Eric Biber, The
Importance of Resource Allocation in Administrative Law, 60 Admin L Rev 1, 40–49 (2008);
Daniel A. Farber, Taking Slippage Seriously: Noncompliance and Creative Compliance in
Environmental Law, 23 Harv Envir L Rev 297, 307–08 (1999).
199 See Lazarus, 94 Cornell L Rev at 1174–75 (cited in note 196); Lazarus, The Making of
Environmental Law at 223 (cited in note 197); Cass R. Sunstein, Endogenous Preferences,
Environmental Law, 22 J Legal Stud 217, 239 (1993).
200 See Lazarus, 94 Cornell L Rev at 1175 (cited in note 196).
201 See Chrisoula Andreou, Environmental Preservation and Second-Order Procrastination,
35 Phil & Pub Aff 233, 240 (2007).
202 See Daniel A. Farber, Politics and Procedure in Environmental Law, 8 J L, Econ, &
Org 59, 66–67 (1992).
203 See Christopher H. Schroeder, Rational Choice versus Republican Moment—
Explanations for Environmental Laws, 1969–73, 9 Duke Envir L & Pol F 29, 43–56 (1998).
204 See Farber, 8 J L, Econ, & Org at 67 (cited in note 202); Schroeder, 9 Duke Envir L &
Pol F at 45–46 (cited in note 203). Another, more cynical, explanation is that politicians might
respond to the public outcry with symbolic legislation that they have no intent of
implementing, in which case they would not be interested in adopting the kinds of
precommitment mechanisms discussed here. See Murray Edelman, The Symbolic Uses of
Politics 22–29 (Illinois 1964).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 517

that allowed for the enactment of legislation have disappeared:205
“Subsequent legislative amendments, limited budgets,
appropriations riders, interpretive agency rulings, massive delays in
rulemaking, and simple nonenforcement are more than capable of
converting a seemingly uncompromising legal mandate into nothing
more than a symbolic aspirational statement.”206 For a sincere
legislator seeking to ensure effective environmental policy,
developing strong precommitments in environmental legislation that
require effective future implementation is extremely important.207
An appealing precommitment tool is to specify the outcomes for
future regulatory decisions. The problem is that environmental law is
a highly dynamic field.208 Thus, for precommitment to succeed in
environmental law requires “institutional design features that allow
for [ ] flexibility but insulate programmatic implementation to a
significant extent.”209
205 See Farber, 8 J L, Econ, & Org at 63, 72–73 (cited in note 202).
206 Lazarus, 94 Cornell L Rev at 1156 (cited in note 196). See also Farber, 23 Harv Envir
L Rev at 298–99 (cited in note 198).
207 See Lazarus, 94 Cornell L Rev at 1197 (cited in note 196); Murray J. Horn, The
Political Economy of Public Administration 53–54, 183 (Cambridge 1995) (arguing that
commitment problems are central to institutional design, particularly when interest groups
have disparate organizational strengths).
208 See Lazarus, 94 Cornell L Rev at 1180 (cited in note 196); Lazarus, The Making of
Environmental Law at 192 (cited in note 197) (“Broad delegations of lawmaking authority are
necessary . . . because of the sheer complexity of environmental standard setting, [which]
requires deliberations based upon a vast array of informational inputs. . . . The relevant
information . . . is constantly changing in light of new information and technology.”). See also
Daryl J. Levinson, Parchment and Politics: The Positive Puzzle of Constitutional Commitment,
124 Harv L Rev 657, 696 (2011). For examples of prespecified lists created by Congress, see
42 USC § 7412(b)(1) (listing 189 toxic substances to be regulated by the EPA under the Clean
Air Act); 33 USC § 1317(a)(1) (identifying 65 toxic pollutants to be regulated by the EPA
under the Clean Water Act). Such lists were often enacted by Congress as a frustrated
response to perceived foot dragging by agencies. See, for example, Kenneth M. Murchison,
Learning from More Than Five-and-a-Half Decades of Federal Water Pollution Control
Legislation: Twenty Lessons for the Future, 32 BC Envir Aff L Rev 527, 552–55 (2005). They
have been criticized, however, as inflexible, overambitious, and underinformed. See, for
example, William Wombacher, Note, There’s Cologne in the Water: The Inadequacy of U.S.
Environmental Statutes to Address Emerging Environmental Contaminants, 21 Colo J Intl
Envir L & Pol 521, 542–43, 554 (2010); John C. Dernbach, The Unfocused Regulation of Toxic
and Hazardous Pollutants, 21 Harv Envir L Rev 1, 34, 51–53 (1997). See also Stephen Breyer,
Breaking the Vicious Circle: Toward Effective Risk Regulation 39–42 (Harvard 1993); Richard
H. Pildes and Cass R. Sunstein, Reinventing the Regulatory State, 62 U Chi L Rev 1, 96–99
(1995); R. Shep Melnick, The Political Roots of the Judicial Dilemma, 49 Admin L Rev 585,
586, 589–91 (1997); Robert L. Fischman, The Divides of Environmental Law and the Problem
of Harm in the Endangered Species Act, 83 Ind L J 661, 681 (2008). In any case, such lists
eventually become out of date. See Doremus and Tarlock, 26 Pub Land & Res L Rev at 24
(cited in note 16).
209 Lazarus, 94 Cornell L Rev at 1158 (cited in note 196).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
518 The University of Chicago Law Review [79:471
Scientific disciplines might be a very useful solution to this
problem:210 they are flexible in that they develop and respond to new
information through a decentralized, peer-based process; but if the
values of the relevant disciplines are consonant with the underlying
statutory goals, they can provide an important constraint on agency
implementation in the future that is (relatively) immune to shortterm
political pressures. 211
Consider the problem of global fisheries discussed in Part II.A.
There is significant uncertainty about the status of those fisheries
given the limited existing data. A legislature concerned about (for
instance) the protection of marine biodiversity could rely upon the
perspectives of marine ecologists (as opposed to fisheries scientists)
in developing the regulatory and management structure. Marine
ecologists will generally give the benefit of the doubt to protecting
marine biodiversity in relying on data, developing methodologies,
making assumptions, and drawing inferences. As shown in the
example of the use of “catch-per-effort” data, they will be willing to
tolerate uncertainty when making predictions that marine resources
need additional protection.212 That information will in turn be used by
the regulatory and management process to reach outcomes that are
more protective of marine resources than otherwise might occur.
The legislature could even provide that an agency that ignored or
overrode this information could be held accountable in court.
The ESA illustrates how courts might hold implementing
agencies accountable for disregarding the perspectives of particular
scientific disciplines. Science is obviously integral to the implementtation
of the ESA in order to inform agency decisions about what
species are endangered and what steps are required to protect and
restore those species. The implementation of the ESA is also
extremely controversial—there are hundreds of court cases in which
environmentalists, industry, and other parties have challenged the
implementation of the statute by the relevant agencies.213
210 See Shapiro and Guston, 17 J Pub Admin Rsrch & Theory at 541 (cited in note 195)
(arguing that regulatory peer review can constrain a shirking or corrupt agency that violates
the legislature’s preferences). Richard Lazarus has argued that one way to achieve this
precommitment with flexibility is through “more neutral, objective scientific expertise” that
gains access to the policy process through advisory bodies or reports. Lazarus, 94 Cornell L
Rev at 1220–22 (cited in note 196).
211 See Levinson, 124 Harv L Rev at 680 (cited in note 208).
212 See notes 78–90 and accompanying text.
213 See, for example, Tennessee Valley Authority v Hill, 437 US 153, 161–64 (1978);
Miccosukee Tribe of Indians of Florida v United States, 566 F3d 1257, 1262–64 (11th Cir 2008).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 519

In those cases, courts often rely upon the perspectives of
ecology, wildlife management, conservation biology, and related
disciplines to determine whether an agency has failed to meet its
ESA obligations. These are all disciplines that to one extent or
another have perspectives that value the protection of wildlife and
biodiversity.214 Normally it is extremely difficult for a plaintiff to
successfully challenge an agency’s decision.215 But plaintiffs do
sometimes succeed in ESA litigation, and when they do, it is often
because they are able to point to a conclusion within the fields of
ecology, wildlife management, conservation biology, or other
relevant biological disciplines that the agency is wrong.216 In these
cases, the reading of uncertain information that is given by particular
disciplines is privileged by the courts over the interpretation by an
administrative agency of the same information.
While there is no indication that Congress in enacting the ESA
sought this kind of outcome,217 a hypothetical legislator who wished
214 While there are differences among these disciplines, all place high value on the
conservation of wildlife. See Parts II.B and II.C.
215 See, for example, Marsh v Oregon Natural Resource Council, 490 US 360, 377–85 (1989).
216 See, for example, Western Watersheds Project v Foss, 2005 WL 2002473, *15–16 (D
Idaho 2005) (overturning the FWS’s decision not to list plant species under the ESA in part
because the agency’s assessment ignored peer reviews by biologists and ecologists finding that
listing was warranted and because the agency’s conclusion that a 64–82 percent chance of
extinction within 100 years did not warrant listing contradicted a guideline provided by the
International Union for Conservation of Nature, a leading conservation biology professional
and advocacy society, that a 10 percent risk of extinction in 100 years warrants protection);
Center for Biological Diversity v Kempthorne, 607 F Supp 2d 1078, 1089–91 (D Ariz 2009)
(scrutinizing the statements of a leading “jaguar expert . . . at the Wildlife Conservation
Society” and rejecting an agency decision not to prepare recovery plan or designate critical
habitat for jaguar based on the expert’s statements since the statements did not support such a
decision); Defenders of Wildlife v Babbitt, 958 F Supp 670, 676, 681–82 (DDC 1997)
(overturning the FWS’s refusal to list lynx for protection under the ESA where “not a single
biologist or Lynx expert employed by the FWS disagreed with the recommendation” to list the
species, state wildlife agency biologist studies supported listing, and where The Wildlife Society
concluded that species population was in decline); Northern Spotted Owl v Hodel, 716 F
Supp 479, 481–83 (WD Wash 1988) (overturning the agency’s decision not to list owl species
where experts on “population viability” and “the acknowledged founder of the discipline of
‘conservation biology’” concluded that the species should be listed); Western Watersheds
Project v Fish and Wildlife Service, 535 F Supp 2d 1173, 1178, 1180, 1186 (D Idaho 2007)
(remanding the agency’s decision not to list sage-grouse for ESA protection where the decision
was contradicted by a report issued by a “group of State agency wildlife biologists who were
experts on the sage-grouse” that was also “peer-reviewed by an independent group of scientists
selected by the Ecological Society of America” and supported by the conclusions of a “panel of
seven outside scientists with expertise in sage-grouse biology and ecology, sagebrush
community ecology, and range ecology and management”).
217 The ESA requires the FWS and the National Oceanic and Atmospheric
Administration (NOAA) to use the “best scientific and commercial data available” in making
many decisions. See 16 USC §§ 1533(b)(1)(A), 1536(a)(2). This language derives from a series
of predecessor statutes: the 1966 Endangered Species Preservation Act, Pub L No 89-669
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
520 The University of Chicago Law Review [79:471
to reverse and halt the trend towards species extinction might indeed
seek to use disciplines in this way. Such a legislator might understand
that while Congress as a general matter might endorse the protection
of all endangered species, when and if debates ever came down to
the decision about whether to protect a particular species at the
expense of particular interest groups or communities, the political
support might be much weaker.218 The general benefits of protecting
species might be overridden regularly (at a socially undesirable rate)
in the context of specific, concrete decisions. Such a legislator might
also believe that Congress in 1973 knew so little about which species
should be protected ten or twenty or thirty years in the future—or
even how many species might require protection—that specific rules
§ 1(c), 80 Stat 926, repealed by the ESA § 14, 87 Stat at 903, required the implementing
agencies to “seek the advice and recommendations of interested persons and organizations
including, but not limited to, ornithologists, ichthyologists, ecologists, herpetologists, and
mammologists” in making listing decisions. 1966 Endangered Species Preservation Act § 1(c),
Pub L No 89-669, 80 Stat 926, repealed by the ESA § 14, 87 Stat at 903. In 1969, the best
available science requirement was made explicit. Endangered Species Conservation Act of
1969 § 3(a), Pub L No 91-135, 83 Stat 275, 275, repealed by the ESA § 14, 87 Stat at 903. In the
popular consciousness of the 1960s, ecology (mentioned in the 1966 Act) was strongly
associated with the protection of the environment in general and the conservation of
endangered species before the appearance of a distinct field of conservation biology in the
1980s. See Takacs, The Idea of Biodiversity at 11–30 (cited in note 54); Bocking, Nature’s
Experts at 56–58, 61–62 (cited in note 53); Roderick Frazier Nash, The Rights of Nature: A
History of Environmental Ethics 55–86 (Wisconsin 1989); John Opie, Nature’s Nation: An
Environmental History of the United States 413–15 (Harcourt Brace 1998); Sylvia Noble Tesh,
Uncertain Hazards: Environmental Activists and Scientific Proof 40–61 (Cornell 2000); Hays,
Beauty, Health, and Permanence at 26–32 (cited in note 182). For instance, Washington State’s
environmental protection agency was created in 1970 and is still called the Department of
Ecology. See Maria McLeod, Historically Speaking: An Oral History in Celebration of the First
35 Years, 1970–2005 5 (Washington State Department of Ecology 2005), online at
http://www.ecy.wa.gov/pubs/0501006.pdf (visited Dec 15, 2011). Public surveys in this period
identified a correlation between a “scientific approach” to resource management and a general
support for environmental protection. Hays, Beauty, Health, and Permanence at 33–34, 256–58
(cited in note 182) (noting that “[f]rom the Mid-1960s on, ecology as a scientific discipline
became central in environmental inquiry, and much of its perspective was adopted and adapted
by environmentalists in their views as to what kinds of scientific knowledge and technologies
should be advanced,” including the need to protect biodiversity). However, we have almost no
legislative history that could provide further insights into Congress’s intent. Holly Doremus’s
extremely thorough review of that legislative history shows very little discussion of the topic in
Congress. Doremus concludes that the legislative history shows that the best available science
mandate was “generally intended to ensure objective, value-neutral decision making by
specially trained experts.” Doremus, 34 Envir L at 419 (cited in note 15). See also Doremus,
75 Wash U L Q at 1130 (cited in note 55) (expressing skepticism that Congress intended to
privilege particular scientific disciplines through its use of the best available science concept).
218 See Tennessee Valley Authority, 437 US at 152–53, 184 (noting that the original
legislative intent of Congress in enacting the ESA was to “halt and reverse the trend toward
species extinction, whatever the cost,” but the Supreme Court also noted that Congress
subsequently appropriated funds for the completion of a dam that might drive a species into
extinction).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 521

(for example, protect these species now, or protect at least fifty
species a year) are inadequate. But by legally privileging particular
scientific disciplines, a legislator can draw on a decentralized, peerbased
social organization that is outside the direct control of future
legislators or the executive branch, that has expertise in developing
the relevant information needed to resolve uncertainty in the future
and respond to dynamic, complex change, and that has underlying
values that will lead it to push for implementation of the statute
consistent with the legislator’s purposes.
2. Option two: legitimation.
A different choice would be to closely tie the relevant scientific
discipline to the administrative agency structure and use the
discipline to legitimate and advance that agency’s implementation of
the relevant statutes.
From the agency’s perspective, it can use science to provide the
benefits of legitimacy, as discussed earlier.219 But in addition, limiting
the political or legal scope of discussion to a particular discipline may
make it easier for the agency to frame the information relevant for
its decisions, and therefore make the agency’s decision-making
process simpler. The larger the scope of the relevant science, the
more diverse viewpoints are available, the easier it is to find
dissenters; the harder it is to meet whatever legal standard might
exist to change the legal status quo, the easier it is to argue to wait
until the “science is clearer.” On the other hand, by privileging a
particular disciplinary perspective, agencies can cabin the range of
scientific information and arguments that are available for a policy
debate and therefore help shape and advance particular outcomes.
Disciplinary restrictions can be an important way to tame what is
otherwise the very decentralized and peer-driven social structure of
science—but in a subtle way that appears to endorse, rather than
conflict with, science.
From an institutional or legal design perspective, legitimation of
agency decision making may be a desirable goal because the agency
may be seen as necessary for the long-term implementation of the
relevant policy. The institutional power and structure of a major
regulatory or management agency might be essential to offset other
important interest groups (such as regulated industry) in future
implementation. A sympathetic scientific discipline can provide that
219 See notes 29–38, 169–73 and accompanying text. See also Jasanoff, 7 Osiris at 194, 203,
215 (cited in note 155).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
522 The University of Chicago Law Review [79:471
agency with additional political and legal heft that it needs for future
political battles.
An example of how disciplines can provide legitimacy for an
administrative agency, and how an agency might use a discipline to
shape the information relevant for its decision making, is the
interaction between the discipline of range science (the study of the
impacts of grazing on grasslands) and the Forest Service. While
range science has roots that go back to the nineteenth century, the
discipline was nurtured in its formative years by the Forest Service.220
From the beginning, the Forest Service did not limit itself to
managing the timber on its lands. It also looked to manage the
grazing of livestock on those lands—something that hitherto had
been regulated by the states, if at all.221 Grazing regulation was
immediately controversial, requiring two decisions by the Supreme
Court to confirm the agency’s regulatory powers,222 and that
controversy never disappeared, as the Forest Service found itself
regularly locked in battles with grazers and their congressional allies
whenever it sought to reduce permitted grazing levels.223
In fighting those battles, the Forest Service concluded that data
produced by scientists would provide a formidable political and legal
ally.224 In the early twentieth century, government scientists
developed estimates of the maximum grazing capacity for
rangelands, estimates that were then used by the Forest Service to
create systematic estimates of whether rangelands were being grazed
in excess of their “carrying capacity.”225 As historian William Rowley
put it:
[I]n order for its officers in the field to administer with
confidence and professionalism, the Forest Service needed to
220 See N.F. Sayre and M. Fernandez-Gimenez, The Genesis of Range Science, with
Implications for Current Development Policies, in N. Allsopp, et al, eds, Proceedings of the
VIIth International Rangelands Congress 1976, 1978 (2003); Nathan F. Sayre, Eric Biber, and
Greta Marchesi, Social and Legal Effects on Monitoring and Adaptive Management: A Case
Study of National Forest Grazing Allotments, 1927–2007, Socy & Nat Resources (forthcoming
2012). Thanks to Nathan Sayre for assistance in the research of this history.
221 See Omaechevarria v Idaho, 246 US 343, 352 (1918) (upholding state regulation of
grazing on federal land); Light v United States, 220 US 523, 529, 535 (1911) (noting historic
regulation of grazing by states and absence of federal regulation).
222 United States v Grimaud, 220 US 506, 511 (1911); Light, 220 US at 535.
223 See Nancy Langston, Forest Dreams, Forest Nightmares: The Paradox of Old Growth
in the Inland West 206–16 (Washington 1995); William D. Rowley, U.S. Forest Service Grazing
and Rangelands: A History 180–87 (Texas A&M 1985).
224 Sayre and Fernandez-Gimenez, The Genesis of Range Science at 1978–79 (cited in
note 220).
225 See Rowley, U.S. Forest Service Grazing and Rangelands at 99, 101 (cited in note 223);
Sayre, et al, Monitoring as a Social Process (cited in note 220).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 523

make decisions rooted in information derived from objective,
scientific study. . . . [T]he professional manager could speak
much more authoritatively if he could show that ‘studies have
been conducted.’ On the basis of this research, decisions could
be made in the interests of professional management of the
forest and its resources. If the administrators of the range were
to professionalize themselves and their tasks, they needed a
body of knowledge from which they could derive authority. . . .
Range science was the unspoken but necessary source of
authority for aggressive range-management policies.226
Accordingly, the Forest Service adopted the field of range
science as the basis for its grazing-management program, in part
because the field promised a quantifiable estimate of how much
grazing a rangeland could sustainably support and of whether a
particular pasture was exceeding that carrying capacity. Through its
research stations and management actions, the Forest Service helped
guide the field of range science to adopt the very “carrying capacity”
perspective that would be most useful to the agency.227 Over the
decades, the Forest Service increasingly adopted quantitative
estimates of range condition based on range science and hired more
and more range scientists (“range conservation” personnel in the
bureaucracy) to conduct estimates of grazing conditions in order to
continue its struggles with recalcitrant permittees.228
One epoch in the 1940s helps demonstrate the pattern: Congress
had held hearings in which grazing permittees expressed their
unhappiness with federal management and threatened to divest the
Forest Service of its grazing lands.229 The resulting political
environment meant that the agency had to tread carefully.230 It was
no coincidence that in the late 1940s, the Forest Service turned to a
leading range scientist “to develop a system for measuring range
conditions and trends, usable throughout national forests as a
226 Rowley, U.S. Forest Service Grazing and Rangelands at 111 (cited in note 223). See
also Thomas G. Alexander, From Rule-of-Thumb to Scientific Range Management: The Case of
the Intermountain Region of the Forest Service, in Char Miller, ed, American Forests: Nature,
Culture, and Politics 179, 179–80 (Kansas 1997).
227 Sayre and Fernandez-Gimenez, The Genesis of Range Science at 1980 (cited in
note 220) (noting that the “practical exigencies of management and administration determined
the research agenda—and thus indirectly the findings—of range science”).
228 See Sayre, et al, Monitoring as a Social Process (cited in note 220); Sayre and
Fernandez-Gimenez, The Genesis of Range Science at 1978 (cited in note 220).
229 See William Voigt Jr, Public Grazing Lands: Use and Misuse by Industry and
Government 110–21 (Rutgers 1976); Rowley, U.S. Forest Service Grazing and Rangelands
at 173–230 (cited in note 223).
230 See Rowley, U.S. Forest Service Grazing and Rangelands at 188 (cited in note 223).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
524 The University of Chicago Law Review [79:471
management tool that would be uncomplicated, scientifically sound,
and acceptable to permittees who owned livestock using national
forest rangeland.”231 That system has become the basic methodology
for range monitoring for national forests and many other rangelands
in the United States.232 These developments occurred just as the
discipline of range science was developing its institutional structures,
such as the formation of the Society for Range Management in 1948,
and the creation of masters and doctoral programs in range science
in many universities between 1930 and 1950.233
The Forest Service also shaped how range science developed
information. Range science emphasized the notion of a fixed
carrying capacity for rangelands, based on ecological theories that
held that any ecosystem that was protected from disturbance would
naturally proceed towards a climax state (“successional” theory).234 In
the context of rangelands, this meant that any difference between the
“natural” state of a grassland and its existing state was more likely to
be the result of overgrazing, and reduction of grazing was the
appropriate solution.235
However, early range scientists working in the American
Southwest observed that rangeland conditions varied tremendously
from year to year, often regardless of grazing pressures. Carrying
capacity in these rangelands did not appear to be fixed, but instead
highly variable depending on climatic conditions.236 Yet despite these
observations, these range scientists eventually embraced the concept
of fixed carrying capacity, in part because of the legal and
231 Lloyd W. Swift, Kenneth William Parker, 1904–1973, 1 Wildl Socy Bull 153, 153–54 (1973).
232 Sayre, et al, Monitoring as a Social Process (cited in note 220).
233 See Sayre, The Genesis of Range Science at 1980 (cited in note 220). See also Clinton
H. Wasser, Elbert H. Reid, and Arthur D. Smith, A History of the Society for Range
Management, 1948–1985 1–3 (Society for Range Management 1987), online at http://
www.rangelands.org/pdf/SRM History 1948-1985.pdf (visited Dec 15, 2011).
234 National Research Council, Rangeland Health: New Methods to Classify, Inventory,
and Monitor Rangelands 52–62 (National Academy 1994); Society for Range Management,
New Concepts for Assessment of Rangeland Condition, 48 J Range Mgmt 271, 272–73 (1995).
235 See Nathan F. Sayre, Ranching, Endangered Species, and Urbanization in the
Southwest: Species of Capital 76–77 (Arizona 2002).
236 Nathan F. Sayre, Recognizing History in Range Ecology: 100 Years of Science and
Management on the Santa Rita Experimental Range, USDA Forest Service Proceedings
RMRS-P-30 1, 5 (2003) (describing how an early researcher’s “reports contain numerous
remarks, however, that suggest he had doubts about the concept of carrying capacity when
applied ‘in a region where the seasons, the altitude, the slope, and the rainfall are so
variable’”); id at 6 (“[B]oth of [the leading range science researchers in the Southwest]
expressed reservations, tacitly or explicitly, about the central premise of the system of
rangeland administration institutionalized over the following decades.”).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 525

bureaucratic needs for such a concept.237 It would take decades for
range science (and other disciplines, such as ecology) to question this
concept.238 The choice of a particular discipline, and the perspective
that the discipline took (shaped in significant part by the
administrative agency it was allied with), resulted in a significant
change in how the natural world was viewed.
The Forest Service’s interactions with range science provide an
excellent example of the way in which an administrative agency’s
reliance on a particular scientific discipline not only provides
legitimacy for the agency directly, but can also shape the production
of information in ways that supports the agency’s political and legal
goals and narrows the range of information available for political
and legal debate. Range science, under the influence of the Forest
Service, developed a concept of carrying capacity that downplayed
other influences on rangeland conditions (such as climatic
variability) and emphasized the role played by active management of
grazing levels by the Forest Service. This in turn provided political
and legal support for the agency to take steps to manage grazing
levels on Forest Service allotments.
3. Option three: insulation.
A legal or institutional designer might be concerned that a
discipline could interfere with future implementation, perhaps
because the perspective of the discipline could undercut achievement
of desired policy goals. Legislators might accordingly try to reduce
the influence of that discipline. They might impose high publicparticipation
requirements, heavy involvement of political
appointees in the decision-making process, explicit statutory
language mandating that courts not defer to the agency on technical
questions, or explicit efforts to force the development and use of
237 See id at 12–13 (describing how “[r]ange scientists generated carrying capacity
estimates that aspired to be independent of fluctuating rainfall, and economic and political
constraints compelled ranchers and agencies to interpret proper stocking in terms of static
carrying capacities,” and noting that “highly variable carrying capacities might have [been]
economically and administratively impractical”); Sayre, Ranching, Endangered Species, and
Urbanization at 62–65 (cited in note 235) (noting early work that recognized the extreme
variability of Southwest rangelands, but that carrying capacity was “the conceptual foundation
[of] leasing, the cornerstone of range reform,” and therefore fixed carrying capacities had to be
determined); Nathan F. Sayre, Climax and “Original Capacity”: The Science and Aesthetics of
Ecological Restoration in the Southwestern USA, 28 Ecological Restoration 23, 27–28 (2010)
(noting that fixed carrying capacity was needed if range science was to be useful for
management).
238 See National Research Council, Rangeland Health at 62 (cited in note 234).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
526 The University of Chicago Law Review [79:471
competing scientific disciplines in the decisionmaking process,
among other tools.
One possible example of how disciplines might have been used
in this way is the effort to reform the Forest Service in the 1970s.239
Historically, there was a tight relationship between the Forest
Service and the discipline of forestry.240 Graduates from forestry
programs provided the vast majority of professional employees for
the Forest Service for many years.241 The discipline of forestry
emphasized the measurement and classification of forests to
maximize timber production,242 just as the Forest Service emphasized
those goals in its management.243 The Forest Service had significant
influence on forestry research on questions such as whether fire
helped or hindered timber management.244
In the 1960s and 1970s there was a significant public movement
to try and reorient the Forest Service around a wider range of goals
besides timber production. During that time frame, Congress
instructed the Forest Service to manage for a wider range of goals
and to ensure that minimum environmental standards were met on
Forest Service lands.245 One tool that Congress relied upon was
instructing the agency to draw on a wide range of disciplinary
239 See Eric Biber, Too Many Things to Do: How to Deal with the Dysfunctions of
Multiple-Goal Agencies, 33 Harv Envir L Rev 1, 18 (2009).
240 The founder of the Forest Service, Gifford Pinchot, also helped found the Society of
American Foresters and the Yale School of Forestry—the first major forestry school in
America—in 1900. See Samuel Trask Dana and Sally K. Fairfax, Forest and Range Policy: Its
Development in the United States 84 (McGraw-Hill 1980). See also Char Miller, Gifford
Pinchot and the Making of Modern Environmentalism 117, 196, 279 (Island 2001).
241 See Paul W. Hirt, A Conspiracy of Optimism: Management of the National Forests
since World War Two xxxvii, 59–60 (Nebraska 1994); Henry Clepper, Professional Forestry in
the United States 2, 49 (Johns Hopkins 1971).
242 As a select committee of university professors of forestry concluded in a congressional
report, “The core of forestry professionalism, the central tenent [sic] of professional dogma, is
sustained yield timber management.” In other words, managing forests in order to maximize
the amount of timber that can be produced in perpetuity from the forest. Select Committee of
the University of Montana, A University View of the Forest Service, S 91–115, 91st Cong, 2d
Sess 22 (1970). See also Hirt, A Conspiracy of Optimism at 6 (cited in note 241); Nancy
Langston, Environmental and Human Change in Old-Growth Forests, 7 Rsrch Soc Probs &
Pub Pol 253, 260 (1999).
243 For instance, the agency made aggressive efforts after World War II to convert
“decadant old growth” forests into rapidly growing, young forests that would provide much
greater levels of timber production under active Forest Service management. See Langston,
7 Rsrch Soc Probs & Pub Pol at 253, 258–60 (cited in note 242).
244 See Ashley L. Schiff, Fire and Water: Scientific Heresy in the Forest Service 15–50
(Harvard 1962).
245 See Biber, 33 Harv Envir L Rev at 18–20 (cited in note 239).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 527

perspectives in conducting its planning and decision making.246 Over
time, this has resulted in a greater diversity of scientific disciplines
among the staff of the Forest Service, which in turn appears to have
had a significant effect on the agency’s decision making, causing it to
consider environmental impacts of timber production to a much
greater degree.247
246 See, for example, 16 USC § 1604(b) (requiring agency to use a “systematic
interdisciplinary approach” in planning); 16 USC § 1604(f)(3) (requiring agency to use an
“interdisciplinary team” to prepare planning documents). The provision was originally enacted
in 1974 as part of the Forest and Rangeland Renewable Resources Planning Act of 1974
(RPA) § 5(b), Pub L No 93-378, 88 Stat 476, 477. It was part of a forest reform bill introduced
by Senator Hubert Humphrey to provide more balance and environmental protection in Forest
Service decision making. See National Forest Environmental Management Act of 1973, S 2296,
93d Cong, 1st Sess, in 119 Cong Rec S 26797 (daily ed July 31, 1973); 119 Cong Rec S 27174
(daily ed Aug 1, 1973) (statement of Sen Lee Metcalf). Much of the rest of the original bill was
substantially changed in response to industry pressure, eliminating substantive restrictions on
timber cutting, for example, and making it more of a procedural, planning statute that was in
part intended to alleviate a perceived timber shortage. See Amendment No 641, 93d Cong, 1st
Sess, in 119 Cong Rec S 36106 (daily ed Nov 7, 1973) (statements by Sen Humphrey); Forest
and Rangeland Environmental Management Act of 1974, S Rep No 93-686, 93d Cong, 2d Sess 5,
8–9; Forest and Related Resources Planning Act of 1974, HR Rep No 93-1163, 93d Cong, 2d
Sess 1–2. Nonetheless, the interdisciplinary provision remained and became part of the final
law. There was relatively little discussion of the provision itself during the legislative process,
and what references there were emphasized that the provision could improve balance in
agency decision making. See S Rep No 93-686 at 12 (cited in note 246) (“The further
requirement that such plans shall use a systematic interdisciplinary approach to achieve
integrated consideration of physical, biological, economic and other sciences, is designed to
assure that a balanced, comprehensive methodology will be employed.”). Timber industry
representatives who testified at hearings for the bill indicated no opposition and claimed that
interdisciplinary work was already standard practice in the Forest Service. See National Forest
Environmental Management Act, Hearing on S 2296 before the Subcommittee on Environment,
Soil Conservation and Forestry of the Committee on Agriculture and Forestry, 93d Cong, 1st
Sess 78 (1973) (statement of William E. Towell, Executive Vice President, American Forestry
Association); id at 99 (statement of Arnold D. Ewing, Executive Vice President, North West
Timber Association). Interestingly, the Society of American Foresters explicitly endorsed the
provision, perhaps because the society asserted that it already represented a wide range of
practitioners in addition to foresters, and therefore the provision would provide additional
employment opportunities. Forest and Rangelands Environmental Management Act, Hearings
on HR 11320 before the Subcommittee on Forests of the Committee of Agriculture, 93d Cong,
1st Sess 67–68 (1973) (statement of H.R. Glascock Jr, Executive Vice President, Society of
American Foresters).
247 See Biber, 33 Harv Envir L Rev at 22 (cited in note 239). The RPA provision requiring
a reliance on a range of disciplines makes it difficult for courts applying the statute to use the
conclusions of particular disciplines to constrain agency decision making. For instance, in
Sierra Club v Marita, 46 F3d 606 (7th Cir 1995), the court rejected claims by environmental
groups and conservation biologists that the Forest Service was legally required to use
conservation biology principles because conservation biology provided the most relevant
science for shaping agency management plans to achieve the goal of biodiversity protection. Id
at 617, 619–24. The court concluded that the agency had great discretion to draw on a range of
scientific tools and was not required to rely specifically on conservation biology. Id at 619–24.
Marita shows that the use of disciplines in one way under a statute may exclude the use of
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
528 The University of Chicago Law Review [79:471
4. Option four: balancing.
So far the analysis has assumed that there is a particular policy
goal—conservation of endangered species, or active management of
forests to maximize output—that the legal or institutional designer
seeks to achieve and the designer selects particular disciplines to
help achieve that goal, perhaps to offset other political pressures.
But there may be many situations in which the policy goal is not so
specific or where the concern about political pressures is not so
severe. Here, the legal and institutional designer might be more
concerned about trying to get the “best” information (however
defined) to identify the relevant problems and possible solutions.
In this context, drawing on a diverse range of scientific
disciplines may be a good option. A wide range of disciplinary
perspectives will ensure that a wide range of values will be brought
to bear, that the biases or blind spots of any one discipline are more
likely to be identified by individuals in other areas, that important
aspects of a particular problem will be identified, or that
interdisciplinary debates might provoke a broader range of creative
solutions.
Blind spots might be technical or methodological. For instance,
the EPA has been criticized for relying too heavily on experimental
toxicology data (for example, tests on lab animals) for determining
health risks from chemicals or pollution and too little on
observational epidemiological data.248 The concern is that this narrow
focus means that the EPA has ignored useful information that could
help with regulation in the context of high levels of uncertainty.249 It
appears that one of the reasons why environmental review
documents did not accurately identify the risks of a massive oil spill
from deepwater drilling in the Gulf of Mexico (as occurred with the
Deepwater Horizon disaster) is the lack of appropriate scientific
expertise in reviewing agencies, such as engineers who might have
properly evaluated the risks from a blowout.250 In the example of the
disciplines in other ways under that statute. This is a specific example of a more general point:
statutes may limit the ability of agencies to use scientific disciplines in particular ways.
248 See Mark R. Powell, Science at EPA: Information in the Regulatory Process 64, 123,
194–95 (Resources for the Future 1999) (giving an example from an EPA air regulatory
program).
249 Id at 64, 123. In some contexts, reliance on toxicology is seen as being more “riskaverse”
in terms of avoiding harms from potentially toxic chemicals, and therefore more likely
to result in regulation, than reliance on epidemiology. See Rushefsky, Making Cancer Policy
at 28, 32, 41, 45–46 (cited in note 25).
250 See Holly Doremus, Through Another’s Eyes: Getting the Benefit of Outside
Perspectives in Environmental Review, 38 BC Envir Aff L Rev 247, 261–69, 274 (2011).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 529

conflict between fisheries scientists and marine ecologists discussed
in Part II.A, a détente was reached between the two sides with an
effort to work together to develop mutually acceptable
methodologies for assessing global fisheries; the result was a series of
papers that both sides could agree upon as accurately reflecting the
status of the world’s marine biological resources.251
Blind spots may also involve important social goals or values. If
all of our information is gathered and applied within the framework
of a single discipline, and we are not conscious about the value
choices that the discipline is based on, then we may be blind to the
inevitable tradeoffs that those implicit value choices force us to
make. For instance, the definition of a species for conservation
purposes necessarily “requires the making of value judgements.”252
Statistical genetics tests intended to determine whether two
populations are separate subspecies or species worthy of protection
under the ESA necessarily require prior assumptions as to (a) what
types of errors we are more willing to accept and (b) what level of
risk of each error we are willing to accept. Are we more willing to
accept a 5 percent chance that we conclude a population is not a
separate subspecies when in fact it is one (a Type II error), or a
5 percent chance that we conclude a population is a separate
subspecies when in fact it is not one (a Type I error), and which of
the two is the more important decision criterion?253
In a recent scientific fight over whether a population of jumping
mice in the Front Range of Colorado and Wyoming was a subspecies
that warranted legal protection, different scientists took different
positions based on the outcomes of their studies, but few recognized
that at the heart of their disagreement was a difference of opinion
about which risks were more problematic. Those that believed that
the Preble’s jumping mouse was not a separate subspecies
emphasized the risk of Type I errors in their analyses, while their
opponents emphasized the risk of Type II errors.254 But the problem
is even worse because the statistical tests used by the scientists
(particularly those that attempt to estimate the risk of Type II error)
also implicitly require an understanding of what should be the
minimum genetic difference between populations that justifies
calling them species or subspecies—without such an understanding,
251 See Stokstad, 324 Sci at 170–71 (cited in note 73).
252 Carolan, 17 Envir Polit at 450–51 (cited in note 33).
253 Id at 455–56. See Brosi and Biber, 7 Front Ecol & Envir at 493 (cited in note 93).
254 Carolan, 17 Envir Polit at 456–57 (cited in note 33).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
530 The University of Chicago Law Review [79:471
one simply cannot conduct the relevant tests.255 The scientists
involved in the controversy had hardly considered this question.256 In
other words, bias and blind spots as to crucial (implicit) value choices
inevitable in this kind of work meant that important policy questions
were predetermined (unconsciously) by the researchers.257
Because of the disciplinary blind spots, taxonomists and
conservation biologists have not yet developed a concept of a
“minimum level of change or difference” that needs to be met in
order for a species or subspecies designation to be made.258 The
problem, however, is that identification of species or subspecies for
regulatory protection necessarily implies the expenditure of scarce
societal resources at the expense of other social goals—in other
words, a tradeoff. Biologists who take the value of protecting
biodiversity as a given are not well-placed to identify such
tradeoffs,259 and accordingly have not developed analytic tools that
would require identifying tradeoffs, making decisions about those
tradeoffs, and translating them into (for instance) an assessment of
how much genetic difference between populations society should
value and therefore identify as worthy of protection as a protected
species or subspecies. It will require outsiders—perhaps economists,
perhaps judges, perhaps scientists from other disciplines—with
different perspectives to force such a process to occur.260 These are
the reasons Sheila Jasanoff has called for diversity of disciplines to
be used in expert advisory bodies that agencies or policy makers rely
upon.261
255 See Brosi and Biber, 7 Front Ecol & Envir at 493 (cited in note 93). See also Carolan,
17 Envir Polit at 449–50 (cited in note 33).
256 See Brosi and Biber, 7 Front Ecol & Envir at 488–89 (cited in note 93).
257 Funtowicz and Ravetz, Three Types of Risk Assessment at 265 (cited in note 17)
(noting the inevitable value choices implicated by choices as to statistical significance and
statistical analysis tools).
258 See Brosi and Biber, 7 Front Ecol & Envir at 488–89, 492–93 (cited in note 93).
259 Doremus, 86 Tex L Rev at 1617–19, 1627 (cited in note 22) (arguing that “if scientists
do not recognize their values as such, and fail to concede the possibility that others might hold
different values, they cannot consciously monitor the extent to which their values influence
their scientific work,” and noting that “the assumption that there can be only one view about
the relative value of conservation is widespread among [conservation biology graduate
students]”).
260 See Brosi and Biber, 7 Front Ecol & Envir at 492 (cited in note 93) (suggesting the
role that courts could play in forcing this development in the ESA listing process); Holly
Doremus, Using Science in a Political World: The Importance of Transparency in Natural
Resource Regulation, in Wagner and Steinzor, eds, Rescuing Science from Politics 143, 160–64
(cited in note 47) (calling for disclosure of agency scientific reports as part of this process).
261 Jasanoff, 30 Sci & Pub Pol at 161 (cited in note 31). See also Doremus and Tarlock,
26 Pub Land & Res L Rev at 32 (cited in note 16); Doremus, 32 Ecol L Q at 302 (cited in
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 531

An example of the use of a diversity of disciplines in this way in
environmental policy making is the National Environmental Policy
Act of 1969262 (NEPA). NEPA requires the federal government to
conduct a review of the relevant environmental impacts before
conducting any major action that might significantly impact the
environment.263 NEPA further requires any such review to use an
“interdisciplinary approach.”264 One of the goals of NEPA was to
force consideration of a wider range of values in governmental
decision making, particularly environmental values;265 imposing a
requirement for interdisciplinary review of the impacts of proposed
projects can be seen as an effort to reduce disciplinary blind spots
that might ignore important environmental values.266
There is a significant tension in any choice between privileging
or distancing particular disciplines and balancing a diversity of
disciplines. In the first case, scientific disciplines are being used to
stabilize the policy-making process (through constraint, legitimation,
or insulation) against political or economic pressure in order to
ensure that the implementation process will be more likely to
achieve a particular goal. In the second case, a diversity of disciplines
is being used to increase the chances that important values are not
ignored, that important policy options are not overlooked, and that
important facts or information are not excluded.267 In general,
note 22); David E. Winickoff and Douglas M. Bushey, Science and Power in Global Food
Regulation: The Rise of the Codex Alimentarius, 35 Sci, Tech, & Hum Values 356, 372 (2010).
262 Pub L No 91-190, 83 Stat 852 (1970), codified at 42 USC §§ 4332–35.
263 See NEPA § 102, 83 Stat at 853.
264 See NEPA § 102(A), 83 Stat at 853.
265 See Biber, 33 Harv Envir L Rev at 35–41 (cited in note 239) (describing the history
and purposes of NEPA).
266 Id at 38–39 (suggesting that scientists hired by agencies to put its projects in
compliance with NEPA may be sequestered in institutional “cul de sacs” where their
perspectives are not given attention for decision-making purposes). The success of NEPA in
achieving these results is unclear.
NEPA and RPA both can be seen as using interdisciplinary decision-making requirements
to achieve different goals (balancing versus insulation), demonstrating how the same policy
tool might serve different goals in terms of using disciplines in institutional and legal design.
267 Developing the process of balancing across disciplines may require some complex
institutional design choices as well. A simple balancing process might just aggregate estimates
or preferences from different disciplinary representatives. But simple aggregation will only
result in a better approximation of the objective truth in the natural world if the distributions
of those perspectives are roughly symmetric (that is, their errors are equally distributed around
whatever the true value of the natural world is) and independent of each other (that is, their
errors are not correlated). These conditions might not be satisfied in many cases (for example,
disciplines might influence each other). If we know that these conditions are not satisfied, we
might want to give some disciplines a greater representation in the decision-making process (to
offset skewed distributions) or ensure that disciplines that are relatively independent of other
disciplines have representation in the process. Moreover, we may not know if either of these
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
532 The University of Chicago Law Review [79:471
balancing disciplines might be the choice that is preferable when we
are less concerned about public choice failures or political or
economic pressures on the decision-making process. The choice of
balancing a diversity of disciplines can be seen as prioritizing getting
“better” information and a more inclusive decision-making process,
but with less of a focus on a particular, specific policy outcome. The
first three options instead can be seen as prioritizing accomplishing
predetermined goals, with the attendant risk of using “worse”
information in the decision-making process, and at the risk of
excluding particular values.
5. Limits to using disciplines as institutional- or legaldesign
tools.
The discussion above lays out the ways in which scientific
disciplines might be useful institutional- or legal-design tools. But
there are also limits to using them in this way. Some of those limits
have to do with the external pressures that might shape disciplines,
reducing their utility; others have to do with the internal instability
of disciplines, the possibility that they may change, fracture, or
merge with other disciplines and therefore may not be predictable in
the future. Finally, there is the question of whether explicitly relying
on disciplines as policy tools undermines their effectiveness as policy
tools.
a) Outside pressures. Science is not immune to political pressures,
as the Forest Service’s shaping of forestry and range science
shows. Scholars have regularly bemoaned the ability of interest
groups or agencies to shape the production of scientific information
in the environmental arena. For instance, the differential ability of
various interest groups to mobilize effective challenges to new
scientific information (with regulated industry often portrayed as
having a substantial advantage268) might result in the skewing of
scientific outcomes in a systematic way.269 In response to unfavorable
studies, interest groups might marshal criticisms of the
conditions are satisfied, in which case we might choose to rely on other, less mechanistic
approaches, such as deliberation among a group of disciplinary representatives. Similar
problems can arise if our goal is to achieve some level of representation of a variety of social
values and goals through disciplinary balancing.
268 For instance, industry historically dominated the scientific advisory process for
pesticide regulation in EPA because of little opposition from other groups. See Jasanoff, The
Fifth Branch at 40, 143–45 (cited in note 22).
269 William R. Freudenburg and Robert Gramling, Scientific Expertise and Natural
Resource Decisions: Social Science Participation on Interdisciplinary Scientific Committees,
83 Soc Sci Q 120, 120, 128 (2002).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 533

methodologies or interpretations, reanalysis of the data, or other
tools to discredit the results and, perhaps, keep the unfavorable
research out of the peer-reviewed literature.270 The inevitable
existence of assumptions and inferences allows any such critic to
undermine results by deconstructing science.271 Interest groups can
actively sponsor research in efforts to shape outcomes as well,
reaching the desired results through hidden use of inferences,
assumptions, and study design choices.272 Research results can often
correlate with the origin of funding—a study of Bisphenol A (BPA)
research found that industry-supported research was much less likely
to find a correlation between BPA exposure and cancer than nonindustry-
supported research.273 If disciplines are so vulnerable to
outside pressures, then they may not be very useful in constraining
an administrative agency from bowing to outside forces itself. If the
discipline is very vulnerable to outside pressure, then we might be
better off insulating decision making from the discipline to avoid
future implementation problems.
These are real problems, but the questions are always relative
ones:274 For instance, in using disciplines to constrain agencies, are the
political pressures on scientific disciplines less than the political
pressures that might be brought to bear on administrative agencies?
The answer here will not be straightforward. For example, given its
history, range science might be less successful than conservation
biology at constraining the Forest Service but more successful at
legitimating it. Many disciplines have fundamental commitments that
will make them resistant to shaping by outside forces on particular
questions—it seems unlikely that outside pressure would (for
instance) result in conservation biology abandoning its strong
commitment to protecting biodiversity. The level of
underdetermination on particular questions or particular topics will
also affect malleability; the higher the underdetermination, the more
malleable the outcomes and the more susceptible to pressure a
270 McGarity and Wagner, Bending Science at 128–56 (cited in note 47).
271 See id at 133.
272 See id at 61–95. See also William R. Freudenburg, Seeding Science, Courting
Conclusions: Reexamining the Intersection of Science, Corporate Cash, and the Law,
20 Sociological F 3, 13–21 (2005) (describing efforts by a major corporation to support the
publication of a sociological article that might have been supportive of their position in highstakes
litigation).
273 McGarity and Wagner, Bending Science at 96 (cited in note 47), citing Frederick S. vom
Saal and Claude Hughes, An Extensive New Literature Concerning Low-Dose Effects of
Bisphenol A Shows the Need for a New Risk Assessment, 113 Envir Health Persp 926, 928 (2005).
274 See Neil K. Komesar, Imperfect Alternatives: Choosing Institutions in Law, Economics,
and Public Policy 5 (Chicago 1994).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
534 The University of Chicago Law Review [79:471
discipline might be. The particular answer for any particular case will
depend on the history and structure of the relevant discipline: what
kinds of intellectual and economic influences outsiders have on that
discipline, the nature of the research questions being asked, the
kinds of interest groups involved, and other factors.
b) Disciplinary dynamism, diversity, and distinctions. For some
choices in using disciplines as design tools,275 the management or
regulatory process is predicated on the perspectives of particular
disciplines that guide the production of information in particular
ways. That production of information, in turn, guides regulatory or
management outcomes.
But, as noted earlier, disciplines are not fixed monoliths; instead,
they are flexible, diverse, and changing structures. Disciplines might
well change their values and perspectives independently even if they
do resist outside political pressures. If legal or institutional designers
build a regulatory or management structure based on discipline A
taking a consensual position X on a key issue, they may be sorely
disappointed when discipline A’s position on X changes, is unclear,
or when discipline A no longer clearly exists as an institution. These
problems are exacerbated when diversity and dynamism occur in
multiple disciplines that might each be relevant for the particular
regulatory or management question.
(i) Disciplinary dynamism. Disciplinary dynamism includes the
possibilities that the discipline as a field may disappear or change so
significantly that it can no longer be usefully relied upon as an
institutional or legal design mechanism (structural changes), or that a
discipline might itself change its central values or perspectives such
that its role in the regulatory and management process
fundamentally changes (paradigm shift).
Structural changes are not unusual.276 Biochemistry emerged
from the prior discipline of physiology; conservation biology from
the field of ecology. In both cases, what was once one discipline
became two. Other fields disappear,277 and fields might merge as well.
If the legal or institutional structures depend, explicitly or implicitly,
on underlying scientific disciplines that might themselves change,
then this may upset those legal or institutional structures.
275 The option in which a discipline constrains policy making by an agency in the future,
and the option in which a discipline provides support for agency decision making in the future.
276 See Ziman, Real Science at 199 (cited in note 70).
277 See, for example, Charles S. Fisher, The Death of a Mathematical Theory: A Study in
the Sociology of Knowledge, 3 Arch Hist Exact Sci 137, 139 (1966) (describing the death of the
field of invariant theory in mathematics).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 535

Paradigm shifts (or “epistemic drift”) occur when a disciplinary
field’s perspective or knowledge strongly changes over time. From
the point of view of a legal or institutional designer, the problem is
that if one is counting on the discipline to provide a particular
perspective, the changed perspective or knowledge thus will
undermine (or at least alter) the discipline’s utility as an institutional
or legal design tool.278
Such paradigm shifts are also not uncommon. An example is the
gradual shift of wildlife management away from managing primarily
or exclusively for maximizing output of desired game species and
towards management for a wide range of game and “non-game”
species. The state regulatory systems established to control hunting
beginning in the early twentieth century were generally designed to
entrench a particular vision of wildlife management: maximizing
hunting opportunities for sport or recreational hunters. One way this
was accomplished was through state professional wildlife
management agencies staffed by professionals from the new field of
wildlife management, sympathetic to the goals of maximizing game
numbers for hunting.279 Most wildlife managers saw “their primary
responsibility as providing resources (fish and game) to their
clients—anglers, hunters, and trappers.”280
But today, younger wildlife scientists are much less focused on
management for the purposes of providing a crop for hunters and
much more focused on managing for the full range of native species,
whether valuable for game or not.281 Survey data of wildlife
management professionals shows that younger ones are more likely
to embrace ecosystem management and biodiversity protection at
the expense of managing for high levels of game species.282 Wildlife
management scientists in nonprofits and universities are also less
likely than those working in agencies to support hunting as the
278 See Shapiro and Guston, 17 J Pub Admin Rsrch & Theory at 544–45 (cited in note 195).
279 See text accompanying notes 182–92. See also Robert M. Muth and Wesley V.
Jamison, On the Destiny of Deer Camps and Duck Blinds: The Rise of the Animal Rights
Movement and the Future of Wildlife Conservation, 28 Wildl Socy Bull 841, 843 (2000)
(describing the “North American Wildlife Conservation Model” that emphasizes “intensive
management” of wildlife “based on professional training and scientific research”); Thomas D.I.
Beck, Citizen Ballot Initiatives: A Failure of the Wildlife Management Profession, 3 Hum
Dimensions Wildl 21, 23 (1998) (noting how wildlife management follows an “agricultural
model” in which deer are “harvest[ed]” and wetlands become managed as “production units”
for waterfowl).
280 Martin Nie, State Wildlife Policy and Management: The Scope and Bias of Political
Conflict, 64 Pub Admin Rev 221, 223 (2004).
281 Muth, et al, Passing the Torch at 178 (cited in note 191).
282 Id at 185 table 5, 189 table 7.
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
536 The University of Chicago Law Review [79:471
dominant goal of the profession.283 Curricula in wildlife management
departments at universities have changed to include “more rarespecies
conservation” and “more multidisciplinary management”
and less “species harvest management.”284
This paradigm shift caused substantial policy changes in
Colorado in the mid-1990s, when the state wildlife agency became
more open to other goals and imposed stricter regulations on the
trapping of furbearing animals, many of which were predators of
livestock. Agricultural and trapping interests responded by getting
the legislature to move regulatory power over “predator
management” to the commissioner of agriculture, seen as much
more friendly to trapping and livestock interests. Animal rights
groups responded with a successful initiative to prohibit trapping.285
What had been a decision-making process contained within a
regulatory agency and guided by disciplinary perspectives to advance
a particular policy outcome had been transferred to competing
agencies and decision-making structures where the voice of the
discipline was much less influential, and where outcomes were much
more fluid.
(ii) Disciplinary diversity. Relying on disciplines in the
regulatory or management process can be problematic if the
disciplines themselves have diverse perspectives on important
regulatory or management questions and therefore cannot provide
clear answers to those questions. The diversity of opinion on an
important question might be widely shared within a discipline
(dissensus),286 or there may be significant, vocal individuals within a
discipline that dispute the dominant position within the field
(heretics).287 Either way, a decision-making structure predicated on a
discipline providing a predictable answer to particular questions may
break down.
An example of dissensus is the dispute within biological
disciplines about how to define taxonomic groups such as species and
subspecies—a concept that is integral to the regulatory program
under the ESA, since it requires the identification of species that
require protection under the Act. The problem is that the “ESA’s
283 Organ and Fritzell, 28 Wildl Socy Bull at 784 (cited in note 190).
284 Id at 783.
285 Susan Cockrell, Crusader Activists and the 1996 Colorado Anti-trapping Campaign,
27 Wildl Socy Bull 65, 67 (1999).
286 See Robert Ackermann, Consensus and Dissensus in Science, 1986 Proceedings
Biennial Meeting Phil Sci Assn 99, 100.
287 See Steve Fuller, The Elusiveness of Consensus in Science, 1986 Proceedings Biennial
Meeting Phil Sci Assn 106, 113.
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 537

definition of ‘species’ is singularly uninformative.”288 Moreover, the
basic concept of species is not one about which biologists have a
consensus: there is great variation across taxonomic groups and
across geographic areas in terms of number of species that are
identified, and that variation appears to be based on different
standards among taxonomists, not on the biology of those different
taxonomic groups.289 This is in part because there is a great deal of
subjectivity in the definition of a species, with up to two dozen
different definitions in the scientific literature.290 This subjectivity
allows for a wide variety of ways to draw the line between what
counts and doesn’t count as a species.291 The differences can be
substantial—adoption of one definition (for instance, the
phylogenetic species concept) might result in a 48 percent increase in
the total number of species recognized worldwide.292
Given the definitional importance of the species concept for the
regulatory scope of the ESA, these disputes have tremendous policy
importance. Indeed, critics have argued that this ambiguity has been
used for policy purposes by various scientists, purportedly through
the creation of additional species in order to provide additional
political or legal support for conservation measures.293
288 Doremus, 75 Wash U L Q at 1089 (cited in note 55).
289 See Emma Marris, The Species and the Specious, 446 Nature 250, 251 (2007); Paul-
Michael Agapow, et al, The Impact of Species Concept on Biodiversity Studies, 79 Q Rev
Bio 161, 168–69 (2004); Nick J.B. Isaac, James Mallet, and Georgina M. Mace, Taxonomic
Inflation: Its Influence on Macroecology and Conservation, 19 Trends Ecol & Evol 464, 464–65,
467 (2004).
290 See Marris, 446 Nature at 251 (cited in note 289). See also Peter C.H. Pritchard, Status
of the Black Turtle, 13 Conservation Bio 1000, 1000 (1999); Agapow, et al, 79 Q Rev Bio at 163
(cited in note 289).
291 See Isaac, Mallet, and Mace, 19 Trends Ecol & Evol at 464 (cited in note 289) (noting
that ant taxonomists are much more amenable to identifying new species than butterfly
taxonomists); Doremus, 75 Wash U L Q at 1102 (cited in note 55) (noting different
professional traditions in taxonomy across different categories of animals and plants, and
distinction between “lumpers” who group taxa together and “splitters” who identify many
different species).
292 Agapow, et al, 79 Q Rev Bio at 62, 64 (cited in note 289).
293 See, for example, Bowen and Karl, 13 Conservation Bio at 1013 (cited in note 157);
James Mallet, Nick J.B. Isaac, and Georgina M. Mace, Response to Harris and Froufe, and
Knapp et al.: Taxonomic Inflation, 20 Trends Eco & Evol 8, 8 (2005); Robert M. Zink, et al,
Genetics, Taxonomy, and Conservation of the Threatened California Gnatcatcher,
14 Conservation Bio 1394, 1402–03 (2000); Doremus, 75 Wash U L Q at 1102–03 (cited in
note 55) (recounting quote from a FWS biologist that there was a bias towards identifying
species in order to increase protection).
The problem is even worse for the definition of subspecies, another category within the
ESA’s legal definition of “species.” Here there are no consistently applied definitions, leading
to even greater subjectivity. See, for example, Carolan, 17 Envir Polit at 449 (cited in note 33);
Susan M. Haig, et al, Taxonomic Considerations in Listing Subspecies under the U.S.
Endangered Species Act, 20 Conservation Bio 1584, 1586 (2006).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
538 The University of Chicago Law Review [79:471
To the extent that a range of biological disciplines constrains
agency decision making under the ESA, the dissensus eliminates that
constraint—where there is such a divergence of opinion, it is possible
for the agency to justify a decision either way it comes out, making
any legal challenges extremely difficult.294 That is the case even where
the agency’s application of the concept has been highly inconsistent
over time, as Holly Doremus documented in the 1990s.295
Heretics are individual members of a discipline that contest
fundamental values or principles of that discipline. For instance,
conservation biologist Rob Roy Ramey296 has challenged the focus of
the discipline on protecting the full range of biodiversity, including
subspecies and genetic diversity within species, and has argued
instead that biodiversity policy should practice triage and focus on
species that represent significant elements of evolutionary change.297
Such heretics can be drawn upon by interest groups who seek to
challenge the ability of the discipline to stabilize, legitimize, or
constrain decision making. For instance, Ramey conducted research
challenging whether a subspecies of jumping mouse in Colorado and
Wyoming was, indeed, a valid subspecies; developers and state and
local officials who opposed the impacts of ESA regulatory
protections for the jumping mouse drew upon this work in arguing
for development of previously-protected areas.298
Given the underdetermination in much of environmental
science and the stakes of many environmental regulatory and
management decisions, there is tremendous potential and motive for
294 See, for example, Alabama–Tombigbee Rivers Coalition v Kempthorne, 477 F3d 1250,
1255–62 (11th Cir 2007) (upholding agency decision to list an endangered species based on the
conclusion that it was separate from related fish species, despite conflicting evidence). For an
exception in which the agency’s conclusion was overturned, see Center for Biological Diversity
v Lohn, 296 F Supp 2d 1223, 1230, 1236–40 (WD Wash 2003). In this case, the court noted that
when there are extreme practical difficulties in affirmatively establishing evidence of different
taxa, such as the case for whales, consensus is an unreasonable standard. But as noted above, in
many circumstances there will be significant disagreement about species status among
scientists. In such situations, courts are likely to defer to the agency’s conclusion.
295 Doremus, 75 Wash U L Q at 1103–12 (cited in note 55).
296 David Holthouse, Building a Better Mousetrap, Denver Westword (Jan 20, 2005),
online at http://www.westword.com/2005-01-20/news/building-a-better-mousetrap/ (visited
Dec 15, 2011). Ramey has a doctorate in ecology and evolutionary biology, and self-identifies
as a conservationist. See also Rob Roy Ramey II and Laura MacAlister Brown, About WSI
(Wildlife Science International), online at http://www.wildlifescienceintl.com/WSI,_Inc
/About_WSI.html (visited Dec 15, 2011) (describing Ramey’s consulting firm and noting that
Ramey has “been working on endangered species conservation issues for 28 years”).
297 Holthouse, Building a Better Mousetrap at 6 (cited in note 296).
298 Id. Ramey explicitly justified his position in the jumping mouse controversy based on
his preferences for prioritizing protection of major elements of biodiversity.
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 539

interest groups to encourage this kind of heresy.299 As philosopher of
science Imre Lakatos put it, “A brilliant school of scholars (backed
by a rich society to finance a few well-planned tests) might succeed
in pushing any fantastic programme ahead, or, alternatively, if so
inclined, in overthrowing any arbitrarily chosen pillar of ‘established
knowledge.’”300 Dissenters always exist, but they become a lot more
important when the political or economic stakes of a decision that
might depend on science are very high. For instance, disputes over
climate science have risen because the political and economic stakes
became so high (particularly after the Kyoto Protocol) and because
the necessary uncertainty and ambiguities in climate science allow
for dispute to occur.301
(iii) Disciplinary distinctions. Finally, it will not always be easy
to distinguish among disciplines in a regulatory or management
program. First, if the particular discipline to be relied upon has not
been specifically identified, then more than one discipline might be
relevant for important policy decisions, creating the potential for
conflict.
Disciplinary conflict302 is exemplified by an ongoing dispute over
how to restore and manage the San Joaquin River and Sacramento
River delta in California, a major source of drinking and irrigation
water for the state, as well as a vital hub for fisheries and home to
other native species. After 2000, the population of a number of
299 See Brian Martin, Suppression of Dissent in Science, in William R. Freudenburg and
Ted I.K. Youn, eds, 7 Research in Social Problems and Public Policy 105, 106 (JAI 1999)
(stating that “[a] few dissenting experts are sometimes all it takes to turn unanimity into
controversy” and that “[t]he existence of controversy . . . usually serves to undercut the
legitimacy of the dominant position”).
300 Imre Lakatos, Falsification and the Methodology of Scientific Research Programmes, in
Imre Lakatos and Alan Musgrave, eds, 4 Criticism and the Growth of Knowledge: Proceedings
of the International Colloquium in the Philosophy of Science, London, 1965 91, 187–88
(Cambridge 1970). While I do not fully embrace a strong interpretation of Lakatos’s statement
in all fields of knowledge, in the context of environmental science, where underdetermination
is particularly important, it has a great deal of force. See also Carolan, 34 Critic Sociology
at 727 (cited in note 44); Jasanoff, 69 L & Contemp Probs at 38 (cited in note 129). See
generally Freudenburg, 20 Sociological F 3 (cited in note 272). See also Donald Ludwig, Ray
Hilborn, and Carl Walters, Uncertainty, Resource Exploitation, and Conservation: Lessons
from History, 260 Sci 17, 17 (1993) (discussing how the fishing industry relied upon dissenting
scientists to argue that overfishing was impossible for oceanic species in order to resist
regulation of California sardine harvests prior to the collapse of the fishery). Nonetheless, even
in environmental science there are limits to the extent to which outside funding can question
established principles. See Seth Borenstein, Skeptic Finds He Now Agrees Global Warming Is
Real, Boston Globe A2 (Oct 30, 2011).
301 See Carolan, 34 Critic Sociology at 730 (cited in note 44).
302 See James V. Spickard, Disciplinary Conflict in the Study of Religion: Anthropology,
Sociology, and “Lines in the Sand,” 14 Method & Theory Stud Relig 141, 142 (2002).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
540 The University of Chicago Law Review [79:471
native fish species, including the delta smelt, crashed. Pursuant to the
ESA, the FWS has proposed restrictions on the pumping of water
from the delta for irrigation. Irrigators challenged in court the FWS’s
Biological Opinion (BiOp) that imposed the pumping restrictions.303
A major source of controversy over the BiOp was whether the
FWS’s analysis of the status of the delta Smelt, and the impacts of
water pumping on it, was adequate. Water districts challenging the
BiOp obtained declarations from several fisheries biologists who
contended that the BiOp had improperly relied too much on
qualitative analysis, instead of developing a comprehensive,
quantitative “life-cycle model” for the smelt which would have
allowed for a more precise understanding of how water pumping was
affecting the species.304 During court hearings, fisheries biologists
testified on behalf of the water districts that development of a
quantitative “life-cycle model” is “standard operating procedure” for
fisheries management agencies,305 and asserted that such a modeling
exercise could have been completed within a matter of hours to at
most months.306
FWS experts and a National Research Council expert panel
responded that given the lack of monitoring data, the agency had
done the best that it could do, and that a quantitative model based
on inadequate data ran the risk of producing inaccurate and
misleading numbers that might lead to decisions that could
irreversibly harm the smelt.307 In the end, the court agreed that a lifecycle
model is a “standard tool used by fisheries scientists to evaluate
population-level impacts,”308 but nonetheless upheld the FWS’s
303 See Consolidated Delta Smelt Cases, 717 F Supp 2d 1021, 1024–25 (ED Cal 2010).
304 See, for example, Declaration of Dr. Richard B. Deriso, Consolidated Delta Smelt
Cases, Docket No 401, *2–4, 5, 11 (ED Cal filed Mar 23, 2010) (recording the statements of Dr.
Deriso, an expert in fisheries management, asserting that life-cycle models are necessary for
adequate status assessments); Declaration of Dr. Ray Hilborn in Support of Plaintiffs’ Motion
for Summary Judgment, Consolidated Delta Smelt Cases, Docket No 393, *2–3, 6 (ED Cal filed
Mar 23, 2010) (“Hilborn Declaration”) (recording the statements of Dr. Hilbornon on the
importance of life-cycle models for determining a species protection status, similar to Dr.
Deriso’s declaration, and noting Dr. Hilborn’s expertise in fisheries management).
305 Consolidated Delta Smelt Cases, 717 F Supp 2d at 1048.
306 Id at 1048–50.
307 See National Research Council, A Scientific Assessment of Alternatives for Reducing
Water Management Effects on Threatened and Endangered Fishes in California’s Bay–Delta 3,
25–26, 39 (National Academies 2010); Declaration of Ken B. Newman, Consolidated Delta
Smelt Cases, Docket No 484, *4–6 (ED Cal filed Jan 8, 2010).
308 Consolidated Delta Smelt Cases, 717 F Supp 2d at 1048.
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 541

conclusions on this issue because the plaintiffs had not submitted
their own models during the administrative process.309
The choice of disciplines in this case had the potential to change
any evaluation of the FWS’s BiOp. From the perspective of fisheries
scientists, heavily reliant on modeling in their work, the FWS’s
choice was flawed. From the perspective of conservation biologists,
however, there is much greater skepticism of model use; for instance,
conservation biologists have expressed caveats about the use of the
most frequent form of quantitative models in their discipline—
population viability analysis (PVA).310 Conservation biologists
emphasize that, given the often large gaps in the data record from
which any model might be constructed, researchers should be
skeptical of the use of models such as PVAs because of the risk that
limited data, processed unthinkingly through models, might produce
predictions that are insufficiently conservative and protective of
species, the primary goal in the discipline.311 With this type of sharp
disciplinary conflict, the ability of any one discipline to either
constrain or legitimate an agency is greatly reduced, as partisans on
either side can point to an alternate source of scientific information.

[أحمد أبو زنط]

Disciplinary conflict is part of a broader problem in relying on
scientific disciplines as legal- or institutional-design tools. Even
where a discipline has been specifically identified as relevant in the
statutes, regulations, or court decisions, it can be quite difficult to
draw lines between what are admittedly fuzzy concepts. As the
discussion of conservation biology and wildlife management in
Part II.B showed, it will not always be easy to distinguish between an
ecologist and a conservation biologist, or a fisheries scientist and a
conservation biologist, for example. An individual scientist might be
a member of multiple professional organizations (for example, both
the Ecological Society of America and the Society for Conservation
309 Id at 1050 (finding that no party who commented on the public review drafts of the
BiOp had submitted a life-cycle model).
310 See J. Michael Reed, et al, Emerging Issues in Population Viability Analysis,
16 Conservation Bio 7, 14–15 (2002); Barbara L. Taylor, The Reliability of Using Population
Viability Analysis for Risk Classification of Species, 9 Conservation Bio 551, 552, 557 (1995);
J. Michael Reed, Dennis D. Murphy, and Peter F. Brussard, Efficacy of Population Viability
Analysis, 26 Wildl Socy Bull 244, 245–46 (1998); Steven R. Beissinger and M. Ian Westphal, On
the Use of Demographic Models of Population Viability in Endangered Species Management,
62 J Wildl Mgmt 821, 832–33 (1998). But see Hilborn Declaration at *7, 13 (describing PVA as
a risk assessment tool that is commonly used by conservation biology), citing Wikipedia,
Population Viability Analysis, online at http://en.wikipedia.org/wiki/Population_viability
_analysis (visited Dec 15, 2011) (relying on Google searches to contend that such quantitative
analyses are the “accepted scientific standard” in ESA analysis).
311 See Taylor, 9 Conservation Bio at 557 (cited in note 310).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
542 The University of Chicago Law Review [79:471
Biology); the name of the department at which they are located
might not clearly identify their institutional home (for example, both
fisheries scientists and conservation biologists might be trained in an
ecology department); the borders between disciplines might be
disputed and movable (for example, conservation biologists might
claim for themselves the research of all endangered species including
fish, while fisheries scientists might claim for themselves the research
of all fish species, regardless of whether they are endangered). The
problem might be much worse if there are significant legal and
political stakes involved in deciding who is part of a particular
discipline.
(iv) Addressing the challenges raised by disciplinary diversity,
dynamism, and distinctions. These are all nontrivial problems or
costs to the explicit use of disciplines in this way, but they do not lead
to the conclusion that disciplines cannot be used this way. First, again
the comparison must be a relative one: Compared to the regulatory
and management system we would otherwise have, would reliance
on disciplines improve outcomes? The prior use of disciplines by
institutions such as the Forest Service to provide significant legal and
political support provides some strong evidence that disciplines are
not necessarily so unstable that they cannot serve as useful structures
to build management or regulatory systems upon. Indeed, range
science proved to be stable enough to provide decades of support for
the agency. There are a number of factors that might cut against the
legal or institutional instability caused by disciplinary dynamism and
diversity. While disciplines do change, they do have substantial
inertia as a whole; while one individual may change their perspective
quickly, a discipline as a whole may move much more slowly, with
change primarily driven by generational turnover.312 While there may
be dissensus or even heresy on important points, many of the most
fundamental assumptions and perspectives will not be questioned
and will be taken for granted. For instance, even Ramey concurs
with the importance of preserving some forms of biodiversity; the
debate is over whether to prioritize and how.313
Second, there are ways to ameliorate the problem, and the ways
in which courts have relied upon scientists in reaching decisions in
ESA cases shed some light on these tools. One might look to the
conclusions of established professional organizations and societies,
which might be more stable as institutional entities, less dynamic
312 Wildlife management exemplifies this gradual transformation process. See notes 279–85
and accompanying text.
313 See Holthouse, Building a Better Mousetrap at 6 (cited in note 296).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 543

compared to individuals, easier to define and identify, and harder to
manipulate (at least if their membership is substantial). It is
significant that in many cases where courts have overturned agency
decisions implementing the ESA, the basis has been conclusions
reached by leading professional organizations.314 One might resort to
“nose-counting,” looking at the weight of the published literature in
the leading journals of the field, which also will be relatively stable as
institutional entities and easier to define and identify.
Using in part these tools, courts in a significant number of cases
appear to have been able to reach judgments about which scientists
are relevant for their analysis of ESA decisions.315 The risk of error
might be small in many of these cases; after all, the perspectives of
related fields may be similar on many points (for instance, ecologists,
wildlife managers, and conservation biologists all have similar
attitudes towards many of the underlying issues in biodiversity
policy).316
Third, and perhaps most important, the challenges that
disciplinary diversity, dynamism, and distinctions pose to
environmental law are not limited to situations where we privilege or
distance any one particular discipline. The establishment of a
“balanced” decision-making process that represents a range of
disciplines will necessarily require addressing the question of which
disciplines to represent, which decision makers represent which
disciplines, and how the perspectives of each represented discipline
will interact. All of these assessments can be upset by the problems
of diversity, dynamism, and disciplinary line drawing.
And even if we try to avoid the problem by not explicitly or
consciously taking into account disciplines in our regulatory or
management process, the problem still will arise. Science still has to
be used in environmental law; there are often a range of relevant
disciplines, each with different perspectives that might result in the
production of different information. Even if only one discipline
appears relevant, its perspective will still often shape the information
produced (as shown by the example of range science and the Forest
Service). If we ignore these problems, then we will be blindly
drawing upon disciplinary perspectives, perhaps producing
314 See note 216.
315 See note 216.
316 These options are not perfect; for instance, there is evidence that industry has created
or shaped professional organizations and journals to provide legitimacy for its policy positions.
See generally McGarity and Wagner, Bending Science (cited in note 47). And to the extent that
we are relying on lower levels of scientific disciplinary structures, where there are far fewer
formal institutions, it will be much harder to undertake this kind of line drawing.
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
544 The University of Chicago Law Review [79:471
unintended consequences. And surely the interest groups, agencies,
and other actors in the decision-making process will not blind
themselves to the role that disciplines play, as shown by the skillful
use of disciplines by the Forest Service for decades. We are thus
forced to at least consider the role that disciplines play in policy, and,
in so doing, face the challenges posed by disciplinary diversity,
dynamism, and distinctions.
c) Undermining the legitimacy of disciplines. A third major problem
is that explicitly relying on scientific disciplines as legal- or institutionaldesign
tools in certain ways might be self-defeating. If policy makers
specifically rely on disciplines to shape policy, that might undermine the
legitimacy or public standing of disciplines by eroding their perceived
neutrality or independence. The problem here may be less serious when
using disciplines to constrain agencies: here the discipline still remains
autonomous from the administrative agency or political process. The
problem also may be less serious in the context of insulating disciplines
from the policy process, since the insulation will itself also emphasize
the discipline’s independence. Similarly, balancing disciplines might still
imply independence among the various groups. But where a discipline is
used to provide legitimacy for an agency, the very process of identifying
the disciplines as connected with and supportive of an administrative
agency may undermine its perceived independence and neutrality, and
therefore the legitimacy it can provide.
On the other hand, it is not as if disciplines have not been drawn
upon in this way in the past. As the US Forest Service’s history with
both range science and forestry makes clear, scientific disciplines
have been recruited, shaped, and utilized in policy making for
decades, and yet scientific disciplines still have significant amounts of
public legitimacy and still can provide significant legitimacy to
regulatory and management decisions. Again, the question is the
relative utility of disciplines for this goal as opposed to other ways of
increasing agency legitimacy and authority.
6. Assessing the use of disciplines as legal- and institutionaldesign
tools.
This Article suggests four ways that scientific disciplines may be
used as institutional- and legal-design tools. There are surely other
combinations and options: for instance, disciplines might be used in
combination with institutions other than agencies (for instance, a
discipline might be used to constrain or empower decision making by
courts). Across the diversity of ways in which legal and institutional
designers might want to account for disciplines, each option might
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 545

draw differently on the tension between internal and external forces
in shaping scientific disciplines.
The details will vary depending on the particular goals of a
regulatory or management program, the nature of the relevant
disciplines that will plausibly be drawn upon, the political or
economic context for decision making, and the underlying realities of
the natural world. One might even reject explicitly relying on
particular disciplines because of the problems that disciplinary
dynamism, diversity, and distinctions create. The important point is
that wise institutional and legal design in environmental law has to
take into account the paradigms and values of the scientific
disciplines that will produce the information used in the regulatory
or management process if it is ultimately to be fully effective.
Whether we consider disciplines consciously or not, they shape the
information used in environmental law

[أحمد أبو زنط]

B. Seeing Science and Law in a New Way
Even if in the end we do not rely on disciplines as an explicit
legal and institutional design tool, understanding scientific disciplines
as part of the policy process in environmental law helps us
understand the problematic relationship of environmental law and
science in a new, more effective way. Instead of trying to separate
the science from the policy for each stage in a complicated
environmental policy question, it may be much easier in many cases
to simply understand the likely perspective of the scientific
community that is involved in the decision-making process and
therefore the likely policy valence of the assumptions and judgments
made by that scientific community in its analyses and models.317
317 There are in fact two relevant dimensions of scale for the analysis of the interaction of
policy and science. One dimension is the scale of the relevant policy decision-making process:
Do we investigate each individual decision (for example, individual risk assessments for
particular chemicals) or the general process for decision making (for example, risk assessment
as a category)? The second dimension is the scale of our analysis of the interaction of science
and policy: Do we examine each individual application of science mixed with policy to
determine the scope of each category (for example, in developing this dose-response curve,
what elements are science-based and what are policy-based), or do we rely upon disciplines
(for example, was it toxicologists or epidemiologists who are involved in constructing the doseresponse
curve)? As described in Part I, prior scholarship has focused on examinations of
individual applications of science mixed with policy but applied that analysis to both levels of
scale for policy decision making (individual decisions or general processes). See notes 25, 27,
33, and 35. However, there may be limits to the ability to generalize about how individual
applications of science mixed with policy consistently correlate with policy conclusions for
general policy decision-making processes. See text accompanying notes 53–57. My disciplinary
approach can also be applied at both levels of scale (individual decisions, or general decisionmaking
processes), though it is more amenable to extrapolation to larger levels of scale and
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
546 The University of Chicago Law Review [79:471
Choosing a scientific discipline involves choosing a package of
interrelated policy perspectives and scientific facts and
methodologies.318
Recognizing the role of different scientific disciplines in bringing
values to environmental law and policy and the need for a choice
among disciplines and their values might be controversial among
those who believe that “neutral” science, or at least the perception of
neutral science, has given environmentalists a powerful public
rhetorical weapon against industry and other interest groups that
otherwise have heavy advantages in the political arena. Giving up
the neutrality of science—or at least recognizing its limits—might be
opposed as unilateral disarmament.319
But first of all, there are many situations in which
environmentalists have found themselves on the opposite side from
the dominant perspective of important scientific disciplines, and yet
have still successfully won important political battles. For example,
for many years, the nuclear industry and government agencies
enjoyed support from a public consensus among the relevant
physicists and engineers about the safety of nuclear power, but
therefore may be less costly to apply. More generally, there may be circumstances in which
fine-grained analyses are not particularly difficult or might be easily generalizable across
multiple decisions, in which case there may not be as great of an appeal for a discipline-based
analysis.
One weakness of relying on scientific disciplines is that, unlike fine-grained, particular
analyses of individual policy and science judgments for specific decisions, this method will be
less able to detect bad-faith efforts by agencies or interest groups to bury policy choices within
science. In such situations, the bad-faith actor might purport to rely on information from a
discipline with one perspective but adjust the relevant policy and science judgments to reflect
very different perspectives. Thus, my disciplinary approach is not a complete substitute for
efforts to separate science and policy.
318 See J.B. Ruhl, Reconstructing the Wall of Virtue: Maxims for the Co-evolution of
Environmental Law and Environmental Science, 37 Envir L 1063, 1080–82 (2007) (arguing that
environmental law and environmental science should be seen as co-evolved fields that are
interdependent and cannot be fully separated, and calling for greater attention to the
interconnection of law and science in training environmental lawyers, policy makers, and
scientists). See also Brosnan, 37 Envir L at 1005–06 (cited in note 52) (suggesting that greater
integration of science, law, and policy would lead to superior decisions, since “when anyone
enters into the messy realm of science, law, and policy they will, at some point, address topics
and make judgments beyond their expertise”). The understanding of disciplines as a “package”
of perspectives and knowledge has similarities to the “co-production” concept of science that
Sheila Jasanoff has developed. See, for example, Sheila Jasanoff, Ordering Knowledge,
Ordering Society, in Sheila Jasanoff, ed, States of Knowledge: The Co-production of Science
and Social Order 13, 19–36 (Routledge 2004) (describing how scientific knowledge is created
through a process of “co-production” by the physical reality of the world and the social
organization by scientists of that reality into an orderly and recognizable system of
knowledge).
319 The same concerns or arguments might be relevant for those who oppose greater regulation
on economic or other grounds. My responses here apply with equal force to them as well.
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 547

environmentalists successfully challenged a range of controversial
proposals for nuclear power plant locations by effectively developing
new sources of information from other scientific disciplines.320 In the
1970s and 1980s, it was industry that pushed for greater use of
“science” in regulatory decisions (such as in the context of the
assessment of the risks of potentially carcinogenic chemicals) while
environmental groups pushed to open the decision-making process
to more public involvement.321 Industry has called for the greater use
of scientific panels for pesticide regulation, in part because the view
of scientists in the relevant disciplines (such as pathology or
epidemiology) of what is “good science” often correlates well with a
skepticism of identifying negative health impacts from chemicals.322
There is therefore no necessary reason why plainly recognizing
the role that values play in environmental science will consistently
privilege one side of the debate or another. Even in the area of
biodiversity protection, where conservation biology has been tightly
allied with efforts to increase regulatory protections,323 this is the
case. As Holly Doremus has well documented, the George W. Bush
administration skillfully used the rhetoric of science to argue for
delaying or denying protection to endangered species: “[T]he
rhetoric of science is just as suited to blocking conservation measures
as it is to facilitating them.”324
And in the long run, efforts to disguise the role that
assumptions, inferences, and values play in environmental science
320 See Brian Balogh, Chain Reaction: Expert Debate and Public Participation in
American Commercial Nuclear Power, 1945–1975 257–62 (Cambridge 1991).
321 Jasanoff, The Fifth Branch at 15–16, 20–38 (cited in note 22); Ted Greenwood,
Knowledge and Discretion in Government Regulation 265 (Praeger 1984) (noting the call by
industry for more use of scientific advisory panels by OSHA).
322 Jasanoff, The Fifth Branch at 123, 133, 145 (cited in note 22) (describing how pressure
from the agricultural lobby encouraged adoption of scientific adivisory panels for pesticide
regulation under the EPA, and noting pressure from industry for the FDA to use more science
advisory panels and push from industry for the EPA to use science advisors in developing
carcinogen guidelines). Industry has also tried to invoke reviews by the National Research
Council to undermine regulatory decisions it disagrees with. See Fein, Comment, 99 Cal L Rev
at 468–69 (cited in note 4).
323 See Takacs, The Idea of Biodiversity at 112, 115–16, 121 (cited in note 54).
324 Doremus, 32 Ecol L Q at 258 (cited in note 22). See also Carolan, 17 Envir Polit
at 458–60 (cited in note 33) (noting how industry groups have framed the question as whether
“sound science” proves the “hypothesis” that regulation is required, which places a steep
statistical burden on the advocates of environmental regulation); Charles N. Herrick and Dale
Jamieson, Junk Science and Environmental Policy: Obscuring Public Debate with Misleading
Discourse, 21 Phil & Pub Pol Q 11, 15–16 (2001) (noting use of “junk science” rhetoric by
those who challenge environmental regulations); Doremus and Tarlock, 26 Pub Land & Res L
Rev at 17–18 (cited in note 16) (noting that opponents of regulation also call for “sound
science”).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
548 The University of Chicago Law Review [79:471
are likely to backfire because they will lead to a fundamental
suspicion of scientists and the work they produce by large sections of
the public. For instance, in the wake of the deposition of
radioactivity by the Chernobyl catastrophe on the Lake District of
England, government scientists assured local sheep farmers and the
larger public that there would not be long-term impacts on the soil or
the sheep. Once the scientists were forced to admit the uncertainties
in their assessments, backtrack on their assurances, and reveal that
assumptions they had made in their earlier assessments were
inaccurate, the farmers became extremely suspicious and cynical
towards scientists in general.325 Efforts to disguise values and
assumptions and to rely on boilerplate rhetoric in which doubters are
called “anti-science” and “irrational” can instead eliminate trust in
the policy making and scientific processes, and harden opponents’
positions.326
Explicit recognition by scientists of their values, preferences,
and interests and how those might shape the research they conduct
and the conclusions they draw can reduce, rather than exacerbate,
that lack of trust. The public in the Lake District already interpreted
the information provided by the scientists through this lens—
acknowledgment of that possibility and efforts to respond to those
concerns (rather than denial of them) might have improved the
receptiveness of the audience to the relevant information.327
325 Brian Wynne, Misunderstood Misunderstandings: Social Identities and Public Uptake
of Science, in Alan Irwin and Brian Wynne, eds, Misunderstanding Science? The Public
Reconstruction of Science and Technology 19, 21–44 (Cambridge 1996); Bocking, Nature’s
Experts at 8–9, 173 (cited in note 53) (describing how overly assuring statements or scientific
cover-ups of politically sensitive issues may undermine the public’s ability to trust science as a
whole).
326 See Freudenburg, 72 Reliab Eng & Sys Safety at 125, 127 (cited in note 45) (noting the
inherent nature of policy choices in technical decision making, and arguing that use of
“irrationality” or “anti-science” arguments destroys trust, hardens opposition, and eliminates
“common ground” on which discussion might occur). See also Kristin Shrader-Frechette and
Earl D. McCoy, Molecular Systematics, Ethics and Biological Decision Making under
Uncertainty, 13 Conservation Bio 1008, 1012 (1999) (noting that if scientists are not transparent
about how policy is affecting their conclusions, “the public will find out, and the
misrepresentations will not be successful anyway”); Fisher, 20 Oxford J Legal Stud at 121
(cited in note 21) (noting a lack of trust in Britain of risk assessment process in wake of the
“mad cow” disease crisis, where assurances of safety by government officials proved incorrect).
327 See Wynne, Misunderstood Misunderstandings at 38–39, 42–44 (cited in note 325);
Bocking, Nature’s Experts at 154, 169–74 (cited in note 53) (arguing that the public perceives
risk information provided by scientists through a lens that is shaped by the institutional and
political context within which the scientists work, the interests that they might have in shaping
the information, and the degree to which the public trusts the scientists); Doremus, 34 Envir L
at 447 (cited in note 15).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 549

The transparent, explicit recognition that scientists have values
and perspectives, and that these will shape the information they
produce in a world filled with uncertainty, dynamism, and
complexity, does not mean that we fall into the trap of complete
relativism, abandoning the effort to try and develop information
based on the reality in the natural world. Instead, it simply
recognizes the necessary imperfections in all human endeavors,
including science, and seeks to provide for a better decision-making
process by recognizing, compensating for, or even building on those
imperfections.
Here I am drawing on Sandra Harding’s concept of “strong
objectivity,”328 in which “by recognizing the societal and disciplinary
cultures in which each of us is positioned, and that therefore cannot
help but mold our scholarship, we can take steps to becoming more
objective, . . . more self-aware,” and potentially even more powerful
epistemologically.329 Harding’s contention is that we have allowed the
ideal of neutrality in science to obscure the inevitable biases in the
scientific process,330 “includ[ing] the judgments scientists make about
interpretation or reliability of data” and “the decisions scientists
make about which problems to pursue or when to conduct an
investigation.”331
In the end, Harding argues that “[s]cience is politics by other
means, and it also generates reliable information about the empirical
world.”332 If that is the case, and if the goal is to achieve as much truth
as possible as quickly as possible, then scientists, decision makers,
and institutional and legal designers should not demand perfect
neutrality from scientists—surely an impossible, perhaps even
undesirable, task—but instead should draw on as wide a range of
perspectives in “economic, political, and cultural diversity that is
necessary to enable those who count as peers to detect the . . . values
and interests” that may be otherwise dominant in the relevant
scientific community.333 Being cognizant of the standpoints that
328 See note 7.
329 Takacs, The Idea of Biodiversity at 43, 122 (cited in note 54) (applying Harding’s
theory of “strong objectivity” to conservation biology). See also Bocking, Nature’s Experts
at 43 (cited in note 53) (noting that “it can be most productive to acknowledge explicitly the
political nature of knowledge”); N. Katherine Hayles, Searching for Common Ground, in
Michael E. Soulé and Gary Lease, eds, Reinventing Nature? Responses to Postmodern
Deconstruction 47, 48 (Island 1995) (calling for understanding “positionality” of scientists in
trying to provide an understanding of reality).
330 Harding, 59 Soc Rsrch at 568–69 (cited in note 7).
331 Id at 578.
332 Harding, Whose Science? Whose Knowledge? at 10 (cited in note 7).
333 Harding, 59 Soc Rsrch at 578 (cited in note 7).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
550 The University of Chicago Law Review [79:471
researchers come from can make it easier to understand their
assumptions and biases, ones that might otherwise remain hidden,
and thereby allow us to better approach truth.334
CONCLUSION
This Article has focused on the role that the different
perspectives of natural scientific disciplines have and could play in
shaping environmental law, but the issues I have raised are not
limited to either natural science or environmental law. While there
may be a number of ways in which social science disciplines might
differ from natural science disciplines, the underdetermination of
theories from data, and the resulting importance of perspectives in
shaping information, is surely not among them. Economists certainly
have their own perspectives, such as emphasizing the importance of
efficient use of resources, and there are significant variations of
perspectives within economics.335 Moreover, there are important
examples of the use of social science disciplines as design tools within
environmental law.
For instance, in the early 1980s, the ESA was amended to
require that certain agency decisions under the Act could be made
“solely” on the best available science; the provision was intended by
Congress, and has been applied by the courts, to prevent the
consideration of economic costs in making relevant implementation
decisions, a form of insulation of particular agency decisions from
the field of economics.336
In contrast, the role of regulatory review by the White House
Office of Management and Budget (OMB) shows the use of
economics as a discipline to shape decision making in favor of the
consideration of economic costs. Beginning with the Carter
Administration and waxing and waning with other Presidents
depending on the ideological perspective, the OMB has had the
ability to at least delay, and sometimes veto, the issuance of major
regulations by environmental agencies in order to require an analysis
of the costs and benefits of the regulations. The result has been a
334 See id at 580–81, 583 (“[S]tandpoint theory demands acknowledgment of the
sociological relativism that is the fate of all human enterprises including knowledge claims, but
rejects epistomological relativism.”). See also Oreskes, 7 Envir Sci & Pol at 380 (cited in
note 45) (arguing that science can never be invoked, in making political decisions, as providing
one concrete and universally accepted answer).
335 See, for example, Peter A. Hall, Policy Paradigms, Social Learning, and the State: The
Case of Economic Policymaking in Britain, 25 Comp Polit 275, 288 (1993) (describing the
distinction between Keynesian and monetarist macroeconomists).
336 See Doremus, 75 Wash U L Q at 1049–56 (cited in note 55).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
2012] Which Science? Whose Science? 551

greater weight on economic costs in environmental regulation. The
relevant office in the OMB is heavily staffed by economists, and the
process can be seen as an effort to privilege economists (this time in
an internal regulatory review program) in order to ensure that their
perspective helps shape the regulatory process.337
Likewise, there are many other areas of law that depend on
scientific information in the context of high levels of uncertainty and
underdetermination, and where the perspectives and values of the
scientific practitioners might significantly shape the information that
is produced.
As an example, consider the controversial history of the forensic
sciences, in particular the disciplines (such as fingerprint
identification, voice identification, and handwriting identification) in
which investigators purport to be able to identify individuals from
physical characteristics.338 These disciplines are incredibly central to
much of modern criminal law, as prosecutors often depend heavily
on identification from forensic scientists in building a case against a
criminal defendant at trial or in deciding which suspect to pursue in
an investigation.339 But criminal law scholars have noted the
weaknesses of the disciplines: There is in fact significant uncertainty
in the identifications that are made, although when forensic scientists
testify as witnesses, they usually frame their answers in terms of
absolute certainty. The clients of forensic scientists are almost
uniformly prosecutors and police investigators, creating at least the
possibility of external pressure to reach outcomes that are supportive
of the goals of those groups. Institutionally, forensic scientists often
work for law enforcement organizations, producing similar pressures.
And, it is at least plausible that individuals who work in these
disciplines do so at least in part because of a personal motivation to
help combat crime.340 All of these factors might lead to a particular
perspective on the part of these disciplines, focused on identifying
337 See Biber, 33 Harv Envir L Rev at 49 (cited in note 239).
338 For an overview of the fields, see Michael J. Saks, Merlin and Solomon: Lessons from
the Law’s Formative Encounters with Forensic Identification Science, 49 Hastings L J 1069,
1080–94 (1998).
339 Id at 1090–94.
340 See id at 1091–94; Risinger, 52 Vill L Rev at 712 (cited in note 70); Michael J. Saks,
The Past and Future of Forensic Science and the Courts, 93 Judicature 94, 95–97 (2009); Susan
Haack, What’s Wrong with Litigation-Driven Science? An Essay in Legal Epistemology,
38 Seton Hall L Rev 1053, 1078–80 (2008); D. Michael Risinger and Michael J. Saks,
Rationality, Research and Leviathan: Law Enforcement-Sponsored Research and the Criminal
Process, 2003 Mich St L Rev 1023, 1037–40; D. Michael Risinger, et al, The Daubert/Kumho
Implications of Observer Effects in Forensic Science: Hidden Problems of Expectation and
Suggestion, 90 Cal L Rev 1, 27–41 (2002).
01 BIBER ART (DO NOT DELETE) 6/13/2012 2:48 PM
552 The University of Chicago Law Review [79:471
criminals, helping to solve cases, and ensuring that the clients or
colleagues succeed in their goals of identifying a suspect and
obtaining a successful conviction in an adversarial trial. As a result, it
is no surprise that the methodologies of forensic scientists often seem
better suited to winning an argument than conducting a (relatively)
dispassionate analysis. As noted above, the disciplines appear to
completely ignore the inevitable uncertainty in their conclusions;341
there has been little or no effort to apply standard scientific
techniques to validate the methodologies used. Indeed, open
hostility has often been the response to those who question the
disciplines (although that may be changing).342
None of this would come as a surprise to anyone who
understands that disciplines are shaped by their external
environment and internal perspectives and values. And seen that
way, a range of solutions becomes clear: perhaps we should create
more institutional separation between forensic scientists and law
enforcement, in hopes of changing the external pressures on the
discipline; perhaps we should broaden the client base, so that
scientists work as much for defense attorneys as they do for law
enforcement; or perhaps courts should take a more skeptical eye
towards the information produced by forensic scientists through
legal doctrines such as exclusion under Daubert v Merrell Dow
Pharmaceuticals, Inc.343 Maybe we might do nothing at all, but even if
that is our option, it should be a choice that is made with an
understanding of the perspectives of the relevant disciplines and how
that shapes the information they are providing to decision makers.
341 See notes 338–40.
342 See, for example, Risinger, 52 Vill L Rev at 712 (cited in note 70); Haack, 38 Seton
Hall L Rev at 1078–80 (cited in note 340); Saks, 49 Hastings L J at 1080–90 (cited in note 338);
Risinger, et al, 90 Cal L Rev at 27–41 (cited in note 340).
343 509 US 579 (1993). For explications of many of these proposals, see the sources cited
in note 340