The Intergovernmental Committee for the Cartagena Protocol on Biosafety
meets this week (October 1-5, 2001) in Nairobi, Kenya. A key element of the
protocol is the 'precautionary approach'. The following paper outlines
three core elements of the precautionary principle, and suggests how
precaution can be applied to the transboundary movement and release of
'living modified organisms' (LMOs) within the context of the protocol.
THE PRECAUTIONARY PRINCIPLE and the CARTAGENA PROTOCOL ON BIOSAFETY
The Cartagena Protocol on Biosafety aims to protect biological diversity
from adverse impacts of LMOs. The protocol contains direct reference to the
precautionary approach (in the Preamble) and also includes precautionary
language in the binding text. Articles 10 and 11 state that parties to the
protocol are permitted to take precautionary measures to avoid harm caused
by LMOs, even when there is lack of scientific certainty regarding the
extent of harm that might occur. The Intergovernmental Committee for the
Cartagena Protocol on Biosafety (ICCP) must now determine how the
precautionary approach will be implemented in a non-arbitrary manner.
Specifically, importing countries must have clear guidelines for applying a
precautionary approach to particular LMOs.
APPLYING THE PRECAUTIONARY APPROACH TO LMOS: POINTS TO CONSIDER
1. Definitions of Adverse Effects
Under the precautionary principle, harm or "adverse effects" must be
defined broadly enough to encompass a full range of potential impacts. The
following questions should be considered:
(i) What is the nature and extent of potential harm?
Adverse effects include direct and indirect impacts on individual
organisms, populations and ecosystems and should also consider impacts on
social systems (e.g. changes in agricultural practices or threats to
biological resources of cultural significance). A precautionary approach
should pay particular attention to impacts that are widespread, long-term,
not reversible, and/or accumulative.
(ii) What standards are used to measure harm?
Conclusions about the safety or risks of LMOs will depend on the standards
against which such impacts are measured. For example, several regulatory
policies measure the potential impacts of modified organisms relative to
the impacts of large-scale, high-input agriculture. By this standard, LMOs
are considered hazardous only if they pose a greater threat than intensive
agricultural practices. A different standard of comparison (e.g. organic or
low-input agriculture) will likely result in different conclusions about
the impacts of LMOs. Under the Cartagena Protocol on Biosafety, such
standards must be explicit, and must be sufficiently flexible to
accommodate higher levels of environmental protection.
2. Recognition of Uncertainty
The Cartagena Protocol on Biosafety suggests that parties may err on the
side of caution when there is lack of scientific certainty about the
impacts of LMOs and until there is evidence to demonstrate adequate safety.
Implementing these provisions will require analysis of the causes and
extent of uncertainty, and careful evaluation of the evidence used to
demonstrate safety. The following factors should be considered:
(i) Error bias
In most cases, detection and evaluation of the adverse effects of LMOs on
biodiversity will require experiments and monitoring procedures that are
designed specifically for this purpose. Tests designed for other purposes
(such as evaluating agronomic traits) or poorly designed trials will likely
show that LMOs have 'no effect'. Such conclusions may indicate, or result
in, a bias toward showing safety; that is, a bias toward concluding LMOs
pose no adverse effects when in fact they may. In these situations, it is
important to look carefully at how experiments were designed. What was the
sample size? What was the geographic range and time scale of the tests?
What control or comparison studies were used? Careful attention to these
questions will determine if tests are sensitive and robust enough to detect
adverse effects, and will help to shift error bias toward caution.
(ii) Weight of evidence
When dealing with complex biological and ecological systems, experimental
evidence gathered on a single, isolated variable may be a poor indicator of
how the system functions as a whole. While laboratory experiments and
controlled field trials will tell us something about potential impacts of
LMOs, we cannot expect these methods to predict accurately the effects of
unconfined, global release over long periods of time. We must therefore
gather and weigh a broader range of evidence including: interdisciplinary
investigations (e.g. combining ecology, evolutionary biology, sociology,
ethics and economics); local knowledge (e.g. traditional ecological and
agricultural knowledge); case studies (e.g. documented experiences of
people who have used the technology); and correlation to other similar
technologies or activities (e.g. release of non-indigenous organisms).
(iii) Participation and transparency
Despite attempts to acknowledge and reduce uncertainty, conclusions about
the potential effects of LMOs and decisions about their use will always
involve an element of informed judgement. Such judgements will be better
informed if all stages of the research and decision-making process are open
and transparent. Participatory procedures are not only more democratic, but
are also likely to yield more robust and appropriate evidence upon which to
base decisions.
3. Precautionary Action
The Cartagena Protocol on Biosafety endorses the right of parties to make
an "appropriate" decision to avoid or minimise the potential adverse
effects of LMOs on biodiversity. Yet there are many forms of precautionary
action. The most appropriate form should be a function of the level of
identified harm, the extent of uncertainty, the availability of alternative
technologies, as well as the agricultural, social, environmental and
economic goals of individual parties. Examples of precautionary action
include:
(i) Bans on new LMOs or phasing out of existing LMOs
Bans and phase-outs may be appropriate if the stakes are high (e.g. if
there is evidence that hazards are serious) and/or the level of uncertainty
is high and/or alternative, less hazardous, less uncertain technologies are
available.
(ii) Moratoria on further development and commercialisation
Based on existing evidence and uncertainties, a temporary ban may be
appropriate. This measure must include a commitment to continued research
on the potential hazards and benefits of LMOs as well as development of
alternatives.
(iii) Conditional approvals with monitoring
Conditional approvals may be warranted if well-designed, peer-reviewed
testing and open decision-making procedures indicate low potential for
adverse effects and a low degree of uncertainty. Conditions applied to the
approval of LMOs may include restricted time-frames, restricted geographic
ranges and/or restricted commercial uses. Such approvals must also be
coupled to careful monitoring for adverse impacts, and effective mechanisms
for labelling and tracing LMOs. This option, however, rests on the
assumption that adverse effects can and will be detected through
monitoring, and further, that once adverse effects are detected, they can
be reversed or controlled. Proponents and decision-makers must acknowledge
and take responsibility for the repercussions of such assumptions.
For a full version of this paper
