The campaign for the ‘public understanding of science’ has focused on
the image of the scientist as a producer of new knowledge. There is perhaps
an equally urgent need for the public to understand the scientist as ‘expert’,
delivering advice for regulation on issues of personal and environmental
safety. For it is in ‘regulatory science’ that the scientist is most visible
and most immediately responsible for our individual and collective welfare.

Regulatory science is also important for bringing our understanding
of science up to date, because that is where the old positivistic image
of science as the simple objective search for truth is most blatantly incorrect
and misleading. All the problems of scientific practice that scholars and
journalists have recently uncovered – such as maintaining standards of excellence
and containing unethical behaviour – are more visible and are best documented
in the regulatory field.

The American style of regulation is adversarial and rule-governed but,
in principle, public. The British style is consensual and informal but still
largely secret. Styles of scholarship correspond: Americans tend to be empirical,
Europeans (including to some extent the English) theoretical. Sheila Jasanoff
applies a European subtlety of interpretation to the publicly available
American documentary sources, and so produces a synthesis of special quality.

The bulk of her book consists of case studies that illustrate her analysis
of regulatory science. Although these are all American, and so might have
features that are unfamiliar in other countries, the cases are important
in elucidating the complex structure of regulatory science.

The main lesson is that nothing can be taken for granted as unproblematic,
at least not in the American scene. Debate will rage over the procedures
for obtaining and interpreting data, along with the principles of setting
regulatory standards. Legislators, administrators and even the judiciary
share the action with the scientists. The process of ‘validation’, previously
assumed by philosophers to be logically complex but socially unproblematic,
is now seen to present pitfalls at every point.

Although Jasanoff’s analysis is so illuminating on the problems of quality
assurance of scientific information, one perennial issue of principle remains
elusive: the objectivity of science, and of the scientists. There is a
contradiction here that must be resolved if regulatory science is to be
effective scientifically and trusted by the public.

For it is now abundantly clear that the separation of ‘facts’ from ‘policy’
is impossible. ÐÓ°ÉÔ­´´-advisers inevitably negotiate their scientific
conclusions in the light of policy concerns, sectional or common. Yet the
scientists are themselves not clear about this duality, and the public even
less so.

The problem is revealed clearly in the tactic which Jasanoff describes
as ‘boundary work’. This means erecting conceptual and political barriers
which deter the layperson from criticising the scientific justification
for policy conclusions.

To some this might appear to be politics of a yet more subtle and devious
sort, all the more important for being unselfconscious. But there is a real
constitutional problem here. ÐÓ°ÉÔ­´´s’ legitimacy extends only to their
technical expertise. But they are now becoming, as in Jasanoff’s title,
a ‘fifth branch’ of government, and as such not elected and not accountable.

Jasanoff sees the goal of regulatory science as its smooth operation.
She does not discuss the possibility that some American-style bruising debate
might occasionally be a good thing for preventing an overly cosy regulation
game. Yet she is aware of the tendencies to bias in the system. As she remarks,
public-interest groups are less well represented than industry at the scientific
discussions, and are often excluded as proper participants.

The penetration of interest-driven politics into the heart of the scientific
advisory process is a question she deals with only lightly. So when she
considers conflicts of interests – she can only remark lamely (or perhaps
ironically) that ‘this is a structural problem for which there are no easy
²õ´Ç±ô³Ü³Ù¾±´Ç²Ô²õ’.

Those with a public-interest or activist bias might find Jasanoff almost
English in her assumption that harmony is the overriding value in the regulatory
process. But this is a matter of interpretation; scholars and the interested
public alike have much to learn from her work.

Jerry Ravetz is director of the Research Methods Consultancy specialising
in quality assurance of scientific information.

For centuries we have been taught and conditioned to assume that science
is certainty. If not today, then tomorrow, scientists would make the discoveries
that would remove our worries about disease, hunger and even our social
affairs. It is hard to find a pronouncement about science where this faith
is not reinforced, implicitly or explicitly. The distinguished lecturers
at the recent British Association meeting in Swansea did nothing to challenge
the old verities. It is equally hard to find an academic syllabus or examination
where anything but incontrovertible facts are involved.

Yet now some doubts creep in. The new global environmental problems
are vast. The phenomena are poorly recorded and even less well understood.
Our theories and models are constantly being modified by unexpected, significant
factors. Clearly, we are far from having definitive conclusions about these
problems, of the sort that science traditionally offers. Moreover, it is
very hard to say when this instability will give way to a consensus. Typically,
we find that facts are uncertain, values in dispute, stakes high and decisions
urgent; and the framing of the problem involves politics and values as much
as science.

As any scientist knows, a phase of indecision and debate is quite common
in pioneering fields of science. But it is normally expected to be brief,
because researchers would not enter a new field unless they sensed that
it was ripe for development to a consensus. However, global environmental
issues are not chosen for study because of their intrinsic scientific interest.
They are thrust upon us because of their practical urgency, regardless of
whether they can be solved by science. Paradoxically, we now see an inversion
of the old distinction between hard facts and soft values. We face decisions
that are hard in every way, where the scientific inputs are irremediably
soft.

Of course, the decision-making process does not require classic scientific
certainty; politics is all about the management of uncertainty along with
the reconciliation of conflicting interests. But we cannot simply reduce
science to policies and let everyone fight it out. For one thing, environmental
issues are only partly defined in terms of special interests and advantages;
in many of them, it is the welfare of the planet as a whole that is at risk.
Also, science up to now has been concerned with the world in an objective
way. Vested interests, of individuals or of groups, have been rigidly controlled
lest they corrupt the processes of debate and evaluation.

This new mixed area of politics and science, involving what we might
call research driven by agenda, is not a comfortable in which to work. But
that is where the major decisions will be taken, and our survival will depend
on them. We need a code of debate that is appropriate to these new conditions.

The need is not to remove uncertainty (for that is impossible), but
to make it open and positive, rather than covert and manipulative. In conditions
of severe scientific uncertainty, assigning the burden of proof can be decisive
for the outcome of a decision process. Unless the issues are argued explicitly,
decision making will be biased to the advantage of the side with the most
resources and skill, prejudicing the whole debate.

The recognition of irremediable scientific uncertainty in these issues
would entail changes all around. As a start, decision makers would have
to recognise that the buck stops with them. No scientist can honestly cite
a three-digit number as the accurate prediction of some future environmental
state. On their side, scientists would have to recognise the legitimacy
and usefulness of non-expert participants in the process. For whenever particular
socio-environmental problems are discussed (to say nothing of those involving
ethics), the theoretical training and specialised experience of scientists
does not provide them with a commanding expertise.

Indeed, in dealing with such complex problems the scientific experts
are, in their own way, amateurs. They bring essential skills and information.
But their contributions are to a dialogue of exploration and consensus,
rather than to a rigid demonstration which conclusively proves something.
The critical assessment of their information, in the context of its use,
is a task for all participants, and not merely those with narrowly defined,
technical expertise.

For their part, environmental groups will need to appreciate the complexity
of their task. On them falls (perhaps unfairly) the burden of adhering to
scientific standards of argument while debating in a political forum. Political
effectiveness and scholarly integrity are a novel combination in natural
science (though not social science), which is the challenge of our time.
We would all do well to study that cautionary tale, Ibsen’s Enemy of the
People.

In such cases we need an extended peer community, performing a participatory
peer review. Its members would be chosen by their competence in some relevant
area, and also by their common commitment to reach a solution; they could
include scientists, teachers, housewives and investigative journalists.
They would contribute ‘extended’ facts, including field studies, case histories,
leaked information and Parliamentary questions.

Precedents for this principle already exist, in forms appropriate to
national cultures. In Britain it has so far mainly been done through lay
members on Royal Commissions and regulatory quangos. Elsewhere there is
recognition and support for local and environmental representatives in regulation.
This is the way to achieve the right sort of public involvement in, and
understanding of, science in the environmental age.

Science cannot deliver certainty in knowledge on the global environmental
issues, any more than it can deliver certainty on the moral issues of reproduction
engineering. If we are to have the broad social commitment that will be
necessary for people in our societies to adjust their lifestyles, the governing
of these problems cannot be done in the absence of public consent. The management
of scientific uncertainty has become too big a task for the technical experts
alone.

Jerry Ravetz is chairman of the Council for Science and Society. He
has recently published The Merger of Knowledge with Power (Cassell); his
joint book with S O Funtowicz, Uncertainty and Quality in Science for Policy,
will be published by Kluwar later this year.