Student Essay
October 25, 2005


Nina Therkildsen

The precautionary principle is increasingly being invoked in debates
over environmental and public health policy, and throughout the world
it has been incorporated into legislation at both national and
international levels. Although there is no verbatim definition of the
precautionary principle, it is clear that its basic premise is
anticipatory action: where there is a potential threat to humans or
the environment, action should be taken to prevent damage -- even if
there are still scientific uncertainties about cause-effect
relationships (e.g., Sandin 1999, Bodansky 1991, Harremoes et al.

While many believe that the precautionary principle is the key to
dealing effectively with global environmental problems, others see it
as a threat to science-based management and antagonistic to human
development. Henry I. Miller and Gregory Conko have recently published
a series of articles challenging the legitimacy of precaution as a
guiding principle in environmental policy-making (e.g., Miller and
Conko 2001a, 2001b, 2001c, Conko 2003). They argue that the
precautionary principle promotes arbitrary decisions that serve to
paralyze industrial progress, inhibiting development of risk-reducing
technologies so that, in effect, excessive precaution may end up
increasing the overall level of harm that society experiences.

A main thrust of their criticism is that the precautionary principle
is not clearly defined and that it means different things to different
people. Yet, ironically, instead of considering and challenging
several possible interpretations of the precautionary principle, they
structure their entire argument around demolishing the logic behind a
highly simplified and very narrow definition the precautionary
principle -- a definition that is widely different from the broad
interpretation promoted by most supporters of the precautionary
principle (such as the European Science and Technology Observatory
(ESTO) 1999, CEC 2000, Harremoes et al. 2002). This incongruence
arises mainly because Miller and Conko focus exclusively on one aspect
of the precautionary principle -- the preventive action in face of
uncertainty -- whereas observers both in academia, governments, and
NGO's generally agree that the precautionary principle encompasses
additional components such as acknowledging, rather than concealing,
scientific uncertainty and ignorance, exploring a range of
alternatives to potentially harmful actions, and making decisions in
an open, informed, and democratic way with increased public
participation (e.g. CEC 2000, Raffensperger and Tickner 1999, ESTO
1999, Kriebel et al. 2001, Harremoes et al. 2002).

In this paper I argue that when those who are trying to operationalize
and implement the precautionary principle envision it as being
multidimensional and broad, Miller and Conko's attack on a narrow and
incomplete interpretation is misguided: instead of contributing to the
overall debate about the merits and shortcomings of the precautionary
principle, Miller and Conko are fundamentally talking at cross
purposes with supporters of the principle. Miller and Conko do not
articulate a relevant criticism of how the precautionary principle is
incorporated into environmental policy, but simply erect a "straw man"
that they then shred to pieces.

Rather than providing a comprehensive response to their argument, I
try to demonstrate my point by focusing on two of Miller and Conko's
important postulations: 1) that the precautionary principle makes
government regulators less accountable because the lack of a clear
definition allows them more discretion and provides them with a
pretext to justify arbitrary decisions; and 2) that the precautionary
principle infringes on people's personal freedom. For both these
points, we see that only in the context of Miller and Conko's narrow
definition of the precautionary principle do these claims make sense;
when viewed within the framework of policy changes that proponents of
the precautionary principle suggest, these postulations are easily

In terms of the first postulation, Miller and Conko are clearly
correct in claiming that the lack of a clear definition makes it
difficult to operationalize the precautionary principle; there can be
no dispute that when there is no legal consensus on how to interpret
the principle, it is impossible to develop a formulaic standard
procedure for its implementation (Bodansky 1991, Sandin 1999).

However, Miller and Conko suggest that the inevitable consequence of
not having a standard procedure for applying the precautionary
principle is that government regulators become less accountable than
they are under the current management regime. This view is strongly at
odds with the basic premises of a broad definition of the
precautionary principle. According to the logic of the ESTO (1999),
Harremoes et al. (2002) and many others, the precautionary principle
should actually minimize the personal discretion granted policy-makers
and scientific experts and to a greater extent leave important
decisions about risk assessment and risk management in the hands of
the general public.

In order to develop this argument, we must look more closely at the
reasoning behind Miller and Conko's assertions. Throughout their
papers they claim that precaution is not necessary for effective
environmental regulation because current methods used in risk
assessment and management realistically portray the magnitude and
severity of risk faced by society. They argue that the assessment of
risks is scientifically sound because it is based on standardized
methods that in a completely neutral and value-free manner yield
unambiguous numerical indices which easily and consistently can be
translated into prescribed policy actions.

This claim clearly relies on the basic assumption that scientific
assessments are accurate and objective and that science can provide
all the input that is needed to make good policy. However, experience
and empirical evidence is accumulating to demonstrate that none of
these assumptions are true (Santillo et al. 2000, Lackey 1994,
Johnston et al. 1996).

First of all, there are inherent limitations to the knowledge we can
obtain through scientific methods. At the most basic level, our
findings are limited by what questions we ask; we simply do not know
what we do not know (Wynne 1992). Furthermore, the experimental design
typically used in risk assessment studies leave a number of
uncertainties and indeterminacies that are typically not accounted for
and not explicitly identified (Santillo et al. 2000, Power and Adams

In order to come up with a numerical indicator in face of this
uncertainty, risk assessors are forced to make value judgments about
which data to collect, how to simplify available facts into simple
models, how to extrapolate because of unknowns, how to select the
statistical tests to be used, how to select sample size, and which
exposure-response model to employ (Power and Adams 1997).

When different risk assessors make different choices about these
issues and accept different assumptions to overcome uncertainties,
results can vary widely. For example, when eleven European
governments, in an exercise, established teams of experts to predict
the probability of accidental release of ammonia, the teams came up
with eleven different risk estimates ranging from 1 in 400 to 1 in 10
million (Contini et al. 1991). Another example is provided by ESTO
(1999) who compared values obtained over a number of years by industry
and government-sponsored studies in industrialized countries of the
risks and environmental impacts associated with modern coal-fired
electricity generating technologies (expressed as monetary external
costs). The results ranged from 20$ per kilowatt hour to less than
four hundredth of a cent per kilowatt hour -- a difference of more
than 4 orders of magnitude.

When different studies based on the same dataset come up with such
varying results, it is painfully obvious that risk assessments are not
objective; in fact, it appears that the results can often be entirely
dependent upon the assumptions made by the risk assessors (Santillo et
al. 2000, Power and Adams 1997). Under the risk assessment system,
setting these assumptions is seen as a scientific matter, for
scientists alone to resolve, but really it is a highly subjective
process filled with social and political implications that require a
wider debate (Wynne and Mayer 1993). As a result, government
regulators under the risk assessment regime are actually not basing
their decisions on objective knowledge -- rather they justify their
actions through information that carries a veneer of scientific
soundness, but is really a result of subjective judgments by experts.

Contrary to what Miller and Conko argue, the current risk management
system gives an enormous amount of discretion to risk assessors and
managers because it distorts the relationship between value judgment
and expert opinion (Wynne 1992). When scientists are granted mandate
to make decisions that have no basis in science, the public is
disenfranchised with regards to issues that have great implications
for society as a whole.

Under the precautionary principle, policy makers should not assume
that science can answer all questions needed for risk management, but
instead recognize that a large component of risk assessment and
management relies on value-judgments (Kriebel et al. 2001, Power and
Adams 1997). When the uncertainties, ignorance, and indeterminacies
inherent to scientific knowledge are made explicit and are emphasized,
rather than hidden or downplayed, regulators must acknowledge the
limitations to their knowledge and can turn decision-making in to a
more open process where there is room for public input at all steps of
the process. This means that regulators under the precautionary
principle are more accountable because they have to justify their
actions based on value judgments made by society and cannot hide
behind a guise of supposedly objective information that in an ideal
world would give a simple answer to difficult policy questions.

Miller and Conko's claim that the lack of a standard procedure makes
it impossible to implement the precautionary principle in a consistent
way is misguided, because a central tenet in the precautionary
principle is that risk is multidimensional and that knowledge cannot
be reduced to simple numericals or yes/no answers (ESTO 1999). The
level of proof that is appropriate for a particular issue depends on a
number of complex factors including the size and nature of the
potential harm, the claimed benefits, the available alternatives, and
the potential costs of being wrong in both directions, i.e., of acting
or not acting in the context of uncertainty, ignorance and high stakes
(Harremoes et al. 2002). Each situation must be considered
individually, and there can be no analytical fix for the problems
encountered in the social appraisal of risk because the relative
priority attached to different dimensions of risk is intrinsically a
matter of subjective value judgment (ESTO 1999, Harremoes et al.

Hence, the precautionary principle gives consistency a new meaning. In
the risk assessment regime, consistency is about following
standardized procedures that give an appearance of objectivity, but
actually produces subjective, arbitrary, and incomplete numericals to
be translated into policy actions. Under the precautionary principle,
on the other hand, consistency is about looking carefully at what we
know and don't know in each situation, taking into account notions of
uncertainty and ignorance, and then make decisions that are compatible
with the level of risk society finds acceptable.

Conko reveals his misunderstanding of this fundamental aspect of the
precautionary principle by saying that "it is not a risk assessment
tool; our debate is about risk management" (Conko 2003, p. 640). This
way he assumes that the precautionary principle would not change the
way we look at science, only the way we translate the results of
typical risk assessments into policy decisions. This is clearly a
faulty interpretation; risk assessment and risk management are
integrated processes and the precautionary principle represents an
overarching way of thinking about that process (Santillo et al. 2000,
ESTO 1999). The dimension of ignorance and indeterminacy cannot simply
be added to the standard risk assessment procedure. Hence, the
precautionary principle is not simply about taking awkward decisions
in face of uncertainty and ignorance; it is as much a function of
styles of government, patterns and power, and changing interpretations
of participation and value in a complex world and changing society
(HarremoŽs et al. 2002, Wynne 1992).

When viewed through this broad interpretation, it is clear that the
precautionary principle makes government regulators more accountable
because environmental policy-making should to a greater extent become
a public process reflecting the values of society, instead of being
governed by isolated experts who potentially could abuse public trust
by claiming to be objective when they are actually making highly
subjective value-judgments on issues that affect the entire society.

If the public is granted a more central role in risk assessment and
management under the precautionary principle, we can expect that their
freedom to choose what risks they are willing to accept would be
increased. Therefore, it seems somewhat paradoxical that Miller and
Conko claim that the precautionary principle infringes on individual
freedom; they argue that this loss of freedom could occur at two
distinct levels: 1) corporations could be prohibited from developing
new products (because they would have to show that they were "safe"
before they were marketed) and 2) consumers could have less freedom to
choose between goods (because they would not have the option of
choosing those products that were deemed potentially harmful and hence
removed from the market).

When following the logic of the interpretation of the precautionary
principle promoted in ESTO (1999), Harremoes et al. (2002) and many
others, these claims make little sense. The consumers that Miller and
Conko claim will lose their freedom to choose goods are the very same
"general public" that is given more input in the policy-making process
under the precautionary principle. The precautionary principle is not
about achieving zero risk, as such a condition is rarely if ever
possible (CEC 2000). Hence, people (the "consumers" and "general
public") have the opportunity to make conscious decisions about what
products they want access to, taking into account their potential
risk. This way, the precautionary principle does not take the freedom
to choose away from consumers, it only allows them to make these
choices at an earlier stage in technology development.

In addressing Miller and Conko's concern about loss of freedom at the
corporate level, it is important to stress that the precautionary
principle does not call for a complete ban on all industrial progress
and does not hamper all industrial activity (Raffensperger and Tickner
1999). Surely, it may prevent corporations from marketing certain
products that are deemed to confer a potential threat, but that loss
of freedom suffered by a relatively small number of individuals is
more than outweighed by the increased freedom the general public gets
by not having to absorb the costs if the threat materialized -- it is
a question of equitable cost and benefit sharing because the
precautionary principle is about finding the best way to meet
society's needs. If freedom could be assessed per capita, then we are
likely to increase the overall level of freedom in society by
minimizing activities that can provide great profit to a few people
but the externalities of which can confer great costs upon society --
exactly what the anticipatory action prescribed in precautionary
principle is supposed to achieve.

It is interesting that some of the central points in Miller and
Conko's argument against the precautionary principle can be echoed
back by proponents of the precautionary principle as criticisms of the
current environmental regulation regime. Miller and Conko claim that
the precautionary principle would make government regulators less
accountable; proponents of the precautionary principle argue that
regulators face a low level of accountability under the current system
because the general public is entirely disenfranchised in the
decision-making process. Similarly, Miller and Conko argue that the
precautionary principle infringes on personal freedom both at the
level of corporations and consumers; proponents of the precautionary
principle would argue that it can increase the overall level of
freedom in society by promoting cost and benefit sharing.

When opponents can use identical accusations against each other, there
are clearly some fundamental misunderstandings that make this argument
static. It is clear that there is significant incongruence between
what Miller and Conko are criticizing and what people who are working
towards operationalizing and implementing the precautionary principle
are really promoting (e.g. ESTO 1999, Harremoes 2002). Either the
proponents of the precautionary principle have failed to communicate
what they stand for, or Miller and Conko have failed to listen. In any
case, talking at cross purposes is not productive and therefore Miller
and Conko's contribution will hardly help bring forward the general
debate surrounding the precautionary principle. Their publications may
work as adequate propaganda to win over people uninformed about the
true issues in this debate, but if they want their voice to be taken
seriously by a broader audience, they must attune their argument to
attacking the interpretation of the precautionary principle that is
being promoted by its supporters rather than focus on an
oversimplified caricature.


Bodansky, D. 1991. Scientific uncertainty and the precautionary
principle. Environment 33:4-5, 43-44.

Commission of the European Communities (CEC). 2000. Communication from
the Commission on the precautionary principle. Commission of the EC
COM(2000)1. Brussels.

Conko, G. 2003. Safety, risk and the precautionary principle:
rethinking precautionary approaches to the regulation of transgenic
plants. Transgenic Research 12:639-647.

Contini, M., A. Amendola and I. Ziomas. 1991. Benchmark exercise on
major hazard analysis. Commission of the European Communities EUR
13386 EN. Luxembourg.

European Science and Technology Observatory (ESTO). 1999. On science
and precaution in the management of technological risk. Institute for
Prospective Technological Studies. Sevilla, Spain.

Harremoes, P., D. Gee, M. MacGarvin, A. Stirling, J. Keys, B. Wynne,
and S. G. Vaz (eds.). 2002. The precautionary principle in the 20th
century: late lessons from early warnings. Earthscan. London, UK.

Johnston, P., D. Santillo, and R. Stringer. 1996. Risk assessment and
reality: recognizing the limitations. pp. 223-239 in Quint, M. et al.
(eds.). Environmental impact of chemicals: assessment and control.
Royal Society of Chemistry. Cambridge, UK.

Kriebel, D., J. Tickner, P. Epstein, J. Lemons, R. Levins, E. L.
Loechler, M. Quinn, R. Rudel, T. Schettler, and M. Stoto. 2001. The
precautionary principle in environmental science. Environmental Health
Perspectives 109:871-876.

Lackey, R.T. 1994. Ecological risk assessment. Fisheries 19:14-18.

Miller, H.I. and G. Conko. 2001. Precaution without principle. Nature
Biotechnology 19:302-303.

Miller, H.I. and G. Conko. 2001. The perils of precaution. Policy
Review 107 (Accessed 6
February 2004).

Miller, H.I. and G. Conko. 2001. Precaution (of a sort) without
principle. Priorities for Health 13
(Accessed 6 February 2004).

Power, M and S.M. Adams. 1997. Charlatan or sage? A dichotomy of views
on ecological risk assessment. Environmental Management 21:803-805.

Raffensperger, C. and J. Tickner (eds.). 1999. Protecting public
health & the environment: implementing the precautionary principle.
Island Press. Washington, D.C.

Sandin, P. 1999. Dimensions of the precautionary principle. Human and
Ecological Risk Assessment 5:889-907.

Santillo, D., P. Johnston, and R. Stringer. 2000. Management of
chemical exposure: the limitations of a risk-based approach.
International Journal of Risk Assessment and Management 1:160-180.

Thornton, J. 2000. Pandora's poison: chlorine, health and a new
environmental strategy. MIT Press. Cambridge, MA.

Wynne, B. 1992. Uncertainty and environmental learning. Global
Environmental Change (June):111-127.

Wynne, B. and S. Mayer. 1993. How science fails the environment. New
Scientist (5 June):33-35.