GENET archive


9-Misc: The dialogue between precaution and risk

genet-news mailing list

-------------------------------- GENET-news --------------------------------

TITLE:  The Dialogue Between Precaution and Risk
SOURCE: Nature Biotechnology 20 (11): 1076-1078, by Lillian Auberson-Huang
        sent by AgBioView, USA
DATE:   Nov 2002

------------------ archive: ------------------

The Dialogue Between Precaution and Risk

"To realize that you do not understand is a virtue Not to realize that you 
do not understand is a defect. The reason why the sage has no defects Is 
because he treats defects as defects. Thus, he has no defects. - Tao Te 

The "precautionary approach" or "precautionary principle" is based on the 
premise that activities threatening harm to human health or the environment 
require precautionary measures, even if some cause-and-effect relationships 
are not fully established scientifically. This concept is gaining 
popularity in environmental policy and law as applied to the commercial use 
of genetically modified organisms (GMOs). Many opponents of the 
precautionary principle argue that its definition and goals are vague and 
its adoption by regulators potentially stifles trade and limits innovation. 
However, I argue here that the precautionary approach also fills a void 
created by the growing disenchantment with the limitations of classical 
risk-assessment methodology for evaluating the wide range of potential 
impacts from technology applications, especially in agricultural 

Classic limitations

Risk assessment for the introduction of GMOs into the environment is based 
on four steps: first, hazard identification; second, evaluation of 
potential consequences; third, estimation of damage potential, if the 
hazard materializes; and fourth, implementation of risk-mitigation 
strategies. Regulations for transgenic crops require risk assessments to be 
performed with the "best available science" and call for ethical behavior 
in decision-making, based on the prima facie principles of minimizing risk 
and reducing the damage potential from potential hazards.

The classic approach of assessing hazards individually, based on likelihood 
of occurrence and extent of damage, has proven too limited to adequately 
describe the full range of technological, societal, and environmental 
impacts from the deliberate release of GMOs. Experience also has shown that 
the interpretation of the final residual risk by various stakeholders can 
be highly dependent on context and all too easily politicized or 
misconstrued. Cases in point are the assessments of the risks of horizontal 
gene transfer of antibiotic marker genes in transgenic plants, and the 
risks of Bacillus thuringiensis (Bt) proteins expressed in transgenic corn 
pollen and their potential effect, under field conditions, on the mortality 
of the monarch caterpillar. The difficulties in interpreting the meaning of 
residual risk arise, in part, from the various value systems that are 
applied in our definition, regulation, and acceptance of plant genetic 
engineering in crop breeding. Scientists, regulators, and consumers alike 
are all struggling with essential questions regarding the nature of 
transgenic plants. In regulations, this gives rise to various regulatory 
triggers and various portals of entry for transgenic plants into the 
regulatory machinery1.

In other words, societal values and subjectivity inevitably provide the 
"framing assumptions" for the risk assessment. To give an example of 
framing assumptions, one can consider the case of corn expressing Bt toxin, 
which is lethal to lepidoptera. Both the European corn borer and the 
monarch caterpillar belong to this class of insect, but the first is 
considered an agricultural pest, whose fate carries little emotional 
impact, whereas the second is culturally important and high in both color 
and emotional impact. From a purely objective stance, Bt toxin is neither 
"good" nor "bad" and the European corn borer and the monarch caterpillar 
are neither "better" nor "worse" than one another. Risk assessment, when 
used as a tool, can provide some guidance to the potential damage of a 
particular hazard (such as the consequences from the presence of Bt in the 
environment), but no answers regarding whether it is "right" or "wrong" to 
allow the hazard to exist. It thus becomes necessary to consult civil 
society regarding risks, benefits, and alternative options. The appearance 
of the "precautionary approach" in environmental policy-making further 
reminds us that different sets of values, interests, and priorities need to 
be integrated into the overall assessment of risk. Stirling et al.2 
attribute the shortcoming of the classical risk-assessment approach for 
evaluating the commercial use of transgenic crops to the observation that 
the resulting technological risk exists on many levels of decision-making 
all at once: environmental, human health, agricultural management, 
economical, societal, and ethical. In the regulatory appraisal of risk, it 
is therefore difficult to properly deal with the "multidimensionality" 
(many levels of risk) and "incommensurability" (no single units of 
performance available) of risk. This has to do with the fact that the 
classical risk-assessment methodologies were developed to assess closed 
systems where there was a direct hazard, such as toxicity, pathogenicity, 
or flammability3. These hazards are correlated to potential exposure 
effects, for a final characterization of the residual risk in terms of a 
given unit, such as permissible concentrations of chemicals or organisms in 
the environment. During the risk-management stage, a combination of 
technical upkeep and minimization of human error through information, 
training, and attention to the physical environment of installations are 
ways of actively maintaining the residual risk at as low a level as 

Risk analysis and regulators

All regulatory frameworks around the world require a thorough risk 
assessment of transgenic crops, case by case, before commercial release. 
The science of risk assessment concurs on the essential point that both the 
genotype and phenotype of a transgenic organism need to be evaluated. For 
the genotype, this means the molecular characterization of the gene 
cassette introduced, consisting of promoter, structural gene, and 
terminator, as well as the number of copies actually integrated into the 
plant genome. However, it is through the phenotype, or the expression of 
plant characters, that the plant interacts with the environment and can 
potentially affect human health.

In Annex II of the Cartagena Protocol4, the procedure of risk assessment is 
described in a stepwise fashion, consisting of hazard identification in 
association with the phenotype, estimation of the likelihood that the 
hazard will occur, description of the consequences should the hazards 
occur, and determination of the acceptability of overall risks. In 
addition, Article 15 on Risk Assessment states that "recognized risk 
assessment techniques" should be taken into account. 	 For closed 
technological applications, insights gained while carrying out a risk 
assessment can be extremely useful for the design of proper containment 
measures to prevent the inadvertent release of harmful chemicals or 
dangerous organisms into the environment. Although the risk-assessment 
methodology for the release of GMOs follows a stepwise procedure similar to 
that for closed systems, an entirely different approach is actually 
required. During the risk assessment for the deliberate release of GMOs, 
one needs to ask the question the other way around: once transgenic crops 
are growing in the environment, what would be the estimated damage 
potential from the spread of the genes from the transformed 
plant?antibiotic marker, structural, or promoter?for wild flora, soil 
composition and microbial community, ecology, and human health. This 
requires pre-existing knowledge of baseline values for tolerated risk, such 
as the impact of antibiotics from medicinal and veterinary uses in the 
environment, gene flow from conventionally grown crops, the natural 
evolution of plant pests and viral populations?keeping in mind that the 
baseline has a dynamic quality and needs to be periodically redefined.

Many of the environmental concerns related to the deliberate release of 
transgenic crops are important issues that need to be addressed directly, 
but without losing sight of why we care in the first place or of how this 
sense of ethics can be extended to other urgent environmental issues, such 
as the overuse of antibiotics for prophylaxis in animal husbandry or the 
effects of pesticide use on insect communities. For example, the manure of 
livestock treated with antibiotics has been shown to contain high levels of 
resistant bacteria. By recycling the manure as fertilizer on cropland, 
farmers are inadvertently inoculating the soil with resistant bacteria 
carrying antibiotic-resistance genes. In examining the damage potential 
from the transfer of antibiotic marker genes from transgenic plants to soil 
microorganisms, it will be necessary to understand what gene transfers are 
already occurring and thus to re-examine the desirability of other 
agricultural impacts on the environment.

Precaution informs risk analysis 

The popularity of the precautionary approach bespeaks an ideal in the area 
of environmental policy, leaving the door intentionally open to making 
decisions regarding technological applications as a matter of "choice," in 
the absence of scientific certainty. Environmental processes are highly 
complex; ecological awareness is synonymous with the ability to perceive 
the interdependence of all forms of life and life processes. Yet, the 
concept of interdependence has vast dimensions, beyond the experience of 
one individual, beyond science, and beyond economics.

Article 1 of the Cartagena Protocol calls for states to integrate the 
precautionary approach in action taken to "ensure an adequate level of 
protection in the field of safe transfer, handling, and use of living 
modified organisms resulting from modern biotechnology..." Article 2 
emphasizes the right of states to enforce more stringent protection goals 
for the "conservation and sustainable use of biological diversity than 
called for in this Protocol".

The purpose of regulations is to establish the limits of tolerability and 
acceptability for technological risk and to establish how these limits can 
be assessed and maintained in technological applications. With societal 
health and welfare at stake, regulations should aim for a balance between 
permitting and restricting, with the proper dose of precaution in analyzing 
the risks and benefits.

Risk assessment is the centerpiece in the regulation of transgenic crops. 
Its formality, procedure, and scientific methods are rigorous and well 
established. The call for precaution is neither in contradiction with the 
science of risk assessment nor does it set a requirement for zero risk. 
Instead, the dialogue between precaution and risk invites us to find 
perspective on the risk assessment itself by pondering the following:
- Variations that could occur in measured characteristics or properties
- Possible flaws in the methodology used to assess effects
- The position of the baseline level of significance or insignificance.

This is an interactive process, where the element of precaution challenges 
the science of risk assessment to develop better methodologies and clearer 
definitions of baseline levels of risk. On the other hand, the call for 
precaution could also imply that the weight of ignorance or uncertainty is 
so great as to require that a decision be based on choice alone. There is 
wisdom in acknowledging the unknown, but this principle can be easily 
abused, especially if supported by an aggressive political agenda. There 
should therefore be a provision for verifying the legitimacy of the choice 
and the agency of proxies, with an audit and revocation of the decision 
possible at any time.

The dialogue between precaution and risk has the characteristics of the 
risk-analysis approach described by Wolt and Peterson5 for the evaluation 
of agricultural biotechnology applications. Risk analysis is the 
integration of the science-based risk assessment with inputs from public 
policy, thus integrating science into a wider context of social, cultural, 
political, and economic determinants. Whereas risk assessment is highly 
formalized in detail and procedure and is carried out by technical experts 
("the process of recursive characterization mostly focus(ing) on exposure 
refinement and proceed(ing) by tiers"), risk analysis can be characterized 
as an "analytic-deliberative process, seeking input from stakeholders 
regarding how uncertainties should be addressed." On a societal level, this 
could be input provided from expert panels, citizen juries, depth surveys, 
and focus groups.

Opening the dialogue between precaution and risk with civil society is a 
move toward a systems approach, where all variables are admitted 
simultaneously for the characterization, framing, management, and 
communication of risks. A dialogue between precaution and risk is a 
dialogue between science and society. The depth and sincerity of this 
dialogue will be decisive in deepening the social-learning component 
related to the introduction of new technology. It can also help to prevent 
the science-based risk assessment from being split away from its societal 
context, as has happened in the introduction of transgenic crops into 

The interface of risk assessment and society is a delicate one. A recent 
example of this was the dispute between the United States and European 
Union (EU) at the World Trade Organization (WTO; Geneva) concerning the 
EU's position on hormones in meat and meat products6. In his review of this 
case and its wider implications for the settlement of science-based trade 
disputes, Christoforou7 writes: "Issues of science transcend borders, 
cultures, and traditions, affecting citizens' fundamental rights to life 
and death." Therefore, clarity and proper intentions among the actors 
involved in conflict settlement are desirable qualities for integrating 
scientific uncertainty during judicial review. Christoforou7 points out the 
difference between decisions based on "reason" (risk assessment) and those 
based on "choice" (precautionary approach), indicating that only the latter 
is amenable to judicial review by the Appellate Body. According to him, the 
task of the panel members of the dispute-settlement system (two trade 
diplomats and a lawyer) should be to decide "whether a contested measure is 
based on a risk assessment, not whether the scientific theory upon which 
the conclusions of the risk assessment are based is scientifically correct 
and acceptable."


When it enters into force, the Cartagena Protocol on Biosafety will be the 
first legally binding international agreement to honor the precautionary 
approach as a policy instrument for upholding collective ethical values, 
such as the conservation and sustainable use of biological diversity and 
the protection of human health, both relevant to the transboundary movement 
of GMOs. Risk assessment remains the primary tool under the Protocol for 
competent authorities to "identify and evaluate the potential adverse 
effects of living modified organisms on the conservation and sustainable 
use of biological diversity in the likely potential receiving environment, 
taking also into account risks to human health." Although the Cartagena 
Protocol is clear on the importance of both aspects of informed decision-
making?precaution and risk?it leaves wide open the terms of their 

This is the space in which the dialogue between precaution and risk should 
be made audible. Are the precautionary approach and risk assessment 
compatible? The answer is yes, provided that risk assessment can provide 
some indication as to the proper degree of precaution that is warranted and 
precaution can be used to audit and refine risk-assessment assumptions. 
Given the controversial nature of GMOs, it will be important to promptly 
initiate this dialogue between precaution and risk, between science and 
civil society, to achieve the higher goal of the Protocol, which is greater 
environmental protection in an era of free-market trade.

As the terms of reference between the application of the precautionary 
approach and risk assessment in decision-making are missing from the 
Protocol, this also leaves open the way in which potential trade disputes 
linked to the transboundary movement of GMOs will be settled. Trade issues 
come under the jurisdiction of the WTO agreements, and as Matthee and 
Vermersch8 point out, "It shall not be clear whether a country applies the 
precautionary principle as a means to protect the environment or as a 
disguised restriction on trade." This is a result of the fact that 
"justification or necessity of the adoption of precautionary measures 
cannot be scientifically proven." Some recommendations for improving the 
situation could include emphasizing the importance of small-scale field 
trials, under local environmental conditions, of GMOs intended for import, 
and a clarification of the role of science in dispute settlement7. What is 
certain is that the manner in which competent authorities make decisions at 
the delicate interface between precaution and risk will have a substantial 
impact on consumer perception of issues.

1. Auberson-Huang, L. Proceedings of the Third Congress of the European 
Society for Agricultural and Food Ethics, Florence, Italy, October 3-5, 
2001 119-122 (A&Q, Milan, Italy, 2001).
2. Stirling, A. et al. On Science and Precaution in the Management of 
Technological Risk: A Synthesis Report of Case Studies, Vol. 1 (Institute 
for Prospective Technological Studies, Sevilla, 1999).
3. Kaeppeli O. & Auberson L. Chimia 52, 137-142 (1998).
5. Wolt, J.D. & Peterson, R.K.D. AgBioForum 3, 291-298 (2000).
6. Anonymous. WTO Dispute Panel Report on EC Measures Concerning Meat and 
Meat Products (Hormones). Aug. 18, 1997, WT/DS26/R/USA (World Trade 
Organization, Geneva, 1997).
7. Christoforou, T. Environ. Law J. 8, 622-648 (2000).
8. Matthee, M. & Vermersch D. J. Agric. Environ. Ethics 12, 59-70 (2000

ACKNOWLEDGMENTS. The author wishes to warmly thank Jeff Wolt, Marielle 
Matthee, Pia Malnoe and Josef Syfrig for the many stimulating discussions 
on risk assessment and the precautionary approach.


|                   GENET                     |
| European NGO Network on Genetic Engineering |
|                                             |
|             Hartmut MEYER (Mr)              |
|               Kleine Wiese 6                |
|           D - 38116 Braunschweig            |
|                 Germany                     |
|                                             |
| phone:  +49-531-5168746                     |
| fax:    +49-531-5168747                     |                      
| mobile: +49-162-1054755                     |
| email:                   |