by concealing its exercise, which highlights the importance of transparency and
accountability in the construction of a global ‘sustainability science’.
US legislation on the freedom of information entailed that much of the scientific
evidence on which policies were based entered the public domain, with the
consequence that many of the uncertainties that characterized the scientific basis
of policy-making could no longer be concealed. Once uncertainties became
unconcealable, the tactics had to change. The most widely adopted option was to
diminish the emphasis on claims that policies were based on, and only on, (sound)
science in favour of a narrative asserting that policy deliberations start from valuefree scientific assessments of the expected environmental impacts of technological
products or processes, which are delivered to policy-makers in the form of
quantitative thresholds or targets, along with estimates of such residual uncertainties
as may remain. Policy-makers then decide how to achieve the targets, goals or
standards that the experts had recommended. For example, scientific panels might
specify figures for concepts such as a ‘threshold limit value’ or a ‘maximum residue
limit’, and then policy-makers might set rules to try to ensure that those figures
were not exceeded, or that they were not often exceeded.
The practice of portraying assessments of the risks and/or benefits of technologies, provided by officially appointed panels or committees of experts, as if
they were entirely uncontaminated by non-scientific considerations, least of all
political or economic considerations, is widespread and entirely orthodox; it is,
however, invariably misleading. Representations of such putative risks and/or
benefits of familiar or innovative technologies are inevitably constructed by hybrid -
izing normative with empirical considerations. When scientists working for
industrial interests are asked if they can provide sufficient data to enable their
employers to negotiate their way over regulatory hurdles, they are likely to con -
duct different studies from those that public-interest scientists could conduct when
asked to search for evidence that those products might pose some risks. Not only
may those two groups conduct different kinds of studies, they can also be expected
to interpret the available data in different ways and to judge them against different
criteria.
In a project called Late Lessons from Early Warnings, the European Environment
Agency (EEA) documented numerous examples in which incumbent authorities
had, in the name of science, asserted the absence of environmental risks or hazards,
only for evidence subsequently to emerge showing that the initial reassuring
narratives had exaggerated the reliability of apparently reassuring evidence, and
discounted evidence indicating problems (Gee et al., 2001; EEA, 2013a). Examples
included fish stocks, radioactivity, benzene, asbestos, halocarbons, diethylstilbestrol,
tributyltin and vinyl chloride.
That pattern highlights a key type of policy judgement – namely, how much
of which kinds of evidence are variously deemed necessary or sufficient to
recommend or decide to permit, restrict or forbid some technology or practice?
One of the devices that has contributed to the persistence of such flawed regimes
has been the practice of discreetly leaving it to scientific advisers to decide how
Invoking ‘science’ in green transformations 47
accountability in the construction of a global ‘sustainability science’.
US legislation on the freedom of information entailed that much of the scientific
evidence on which policies were based entered the public domain, with the
consequence that many of the uncertainties that characterized the scientific basis
of policy-making could no longer be concealed. Once uncertainties became
unconcealable, the tactics had to change. The most widely adopted option was to
diminish the emphasis on claims that policies were based on, and only on, (sound)
science in favour of a narrative asserting that policy deliberations start from valuefree scientific assessments of the expected environmental impacts of technological
products or processes, which are delivered to policy-makers in the form of
quantitative thresholds or targets, along with estimates of such residual uncertainties
as may remain. Policy-makers then decide how to achieve the targets, goals or
standards that the experts had recommended. For example, scientific panels might
specify figures for concepts such as a ‘threshold limit value’ or a ‘maximum residue
limit’, and then policy-makers might set rules to try to ensure that those figures
were not exceeded, or that they were not often exceeded.
The practice of portraying assessments of the risks and/or benefits of technologies, provided by officially appointed panels or committees of experts, as if
they were entirely uncontaminated by non-scientific considerations, least of all
political or economic considerations, is widespread and entirely orthodox; it is,
however, invariably misleading. Representations of such putative risks and/or
benefits of familiar or innovative technologies are inevitably constructed by hybrid -
izing normative with empirical considerations. When scientists working for
industrial interests are asked if they can provide sufficient data to enable their
employers to negotiate their way over regulatory hurdles, they are likely to con -
duct different studies from those that public-interest scientists could conduct when
asked to search for evidence that those products might pose some risks. Not only
may those two groups conduct different kinds of studies, they can also be expected
to interpret the available data in different ways and to judge them against different
criteria.
In a project called Late Lessons from Early Warnings, the European Environment
Agency (EEA) documented numerous examples in which incumbent authorities
had, in the name of science, asserted the absence of environmental risks or hazards,
only for evidence subsequently to emerge showing that the initial reassuring
narratives had exaggerated the reliability of apparently reassuring evidence, and
discounted evidence indicating problems (Gee et al., 2001; EEA, 2013a). Examples
included fish stocks, radioactivity, benzene, asbestos, halocarbons, diethylstilbestrol,
tributyltin and vinyl chloride.
That pattern highlights a key type of policy judgement – namely, how much
of which kinds of evidence are variously deemed necessary or sufficient to
recommend or decide to permit, restrict or forbid some technology or practice?
One of the devices that has contributed to the persistence of such flawed regimes
has been the practice of discreetly leaving it to scientific advisers to decide how
Invoking ‘science’ in green transformations 47