An Illustration: Science Advice During the Fukushima Emergency
The date is 2011, the Great East Japan Earthquake and tsunami have struck and more than 15,000 people will die. Meanwhile, the Fukushima Daiichi nuclear plant is in a critical condition. The outcomes will alter, amongst other things, the future of energy policy in Europe. The question from the UK Prime Minister to the Government Chief Scientific Adviser (GCSA) is whether British nationals should be advised to stay in Japan, or to leave (UK Government, 2012).
The GCSA, Sir John Beddington, invoked the pre-arranged mechanism characterised as the Scientific Advisory Group in Emergencies (SAGE) and over a period of about 48 hours gathered scientists from disciplines that included nuclear science, engineering, meteorology and medicine. They reviewed the available evidence and advised the Cabinet Office Briefing Rooms (COBR) Emergency Committee on the balance of risks to people staying or going.
Based on SAGE’s advice about the risks the Prime Minister concluded there was no need to evacuate British Citizens outside the exclusion zone recommended by the Japanese government. Some national governments took a different view, and advised their citizens to leave. Sir John meanwhile carried out telephone conferences with open question and answer sessions at Embassy locations, to which Japanese officials were also invited.
The advice given during the Fukushima emergency shows in acute form some of the characteristics of the provision of science advice in policy more generally. From a practitioner’s point of view perhaps the most important are first, that the science does not determine the outcome; and second, that any significant policy or political question requires insights from multiple disciplines. In an emergency, the adviser will need not only rapidly to decide which disciplines are most relevant but also who to bring in to represent them. Urgent advice depends strongly on knowledge embedded in scientists. In that case they must not only be good scientists, but also willing to work flexibly and communicate succinctly. They need to be willing to use their judgement based on the immediate available information about the particular incident, and they need to be tolerant of other disciplines’ mental models, even notions of what constitutes good science or scholarship.
A large part of the discussions amongst the experts during the emergency was about how to define and describe the level of risk. Remember that there was little time to refine or define, and the important point was communicating specialised information about risk and degrees of uncertainty to non-scientists such as Ministers, who would in turn have to defend their decisions to the public and in the media. Chapter 2 explores engaging with risk.
While the direct death toll from the Fukushima emergency is low, the wider effects are highly significant. One author states that the Fukushima incident is “extremely unlikely to result in a single death” directly from radioactivity (Thomas, 2011), as does the IAEA’s report (International Atomic Energy Agency, 2015). But concern about the direct effects of radiation may deflect attention from the huge systemic effects such as human displacement and the effects on mental health. Chapter 3 explores ways for thinking about systemic effects and plausible futures from the perspective of decision-makers.
At the time there were also reports that some in the West, applying bounded rationality to make sense of the extensive news coverage, may have associated the high death toll with the nuclear emergency rather than the earthquake and tsunami. Sir John’s willingness to discuss the situation publicly when it was still extremely uncertain opened up opportunities for engagement with the changing levels of knowledge and confidence. Chapter 4 explores modes of public engagement and the use of different lenses in public reasoning.
The experts grappled hard with the challenge of summarising degrees of certainty and confidence in their evidence. They settled on defining what they called the Reasonable Worst Case scenario, and outlining the risks associated with that. At the moment of Ministerial judgement, cognitive discussions about what risk such a phrase might represent take their place in a wider context of experience, professional relationships and the emotional dynamics of human decision-making. The Minister is really asking the scientist the very human question: “Are you sure? Are you sure you’re sure?” As Chap. 5 explores, curating these moments is at the heart of the craft skill of enabling policy-makers and scientists to get the best from each other.
Framing the Moment
Whether looking at the near or the very long term, the first challenge for the decision-maker and the Adviser is to agree on how to frame the question and the system. Fukushima is an example of what Roger Pielke Jr. calls “Tornado politics” (Pielke Jr., 2007). Decision-makers know they want scientific advice. A relatively small number of people are directly involved in the process of making the decision. Although under scrutiny, the values that inform it are not much contested at the time.
Pielke contrasts “Tornado politics” with the “Abortion politics” that apply when the science is relatively well established and understood, and political resolution depends primarily on judgements about values.
There is perhaps a third category, that of Ecstasy politics. Here there is not yet a reasonable consensus about which science is most relevant, as in Tornado politics, nor are there established value-based or political framings, as in Abortion politics. This category takes its name from the dispute in the UK that followed advice from the Advisory Council on the Misuse of Drugs to the Home Secretary. In line with its statutory duties, the Council considered the harms and benefits of the drug Ecstasy. It came to the view that those harms were insufficient to justify it being classified and controlled under the relevant legislation.
The political context in the UK, as in many other countries, includes deep-seated moral and political viewpoints about the significance of, and ways to manage harm from, drugs in general. The government announced that it would classify the drug. In protest at what he considered to be the government ignoring the advice, a leading expert, Professor David Nutt, resigned from the Council. He vividly compared the harms from Ecstasy to those of the risky pursuit of horse-riding (Nutt, 2009).
This very public dispute, although possibly damaging to the perceptions both of scientists and of politicians to each other at the time, prompted reflection that led to greater clarity about what should be expected of both. The Government Chief Scientific Adviser led the development of principles that, to paraphrase, pointed to the importance of scientists providing the best evidence; of Ministers listening to it; and of Ministers explaining the bases of their decisions. It equally pointed to the importance of scientists respecting the democratic authority of Ministers in making those decisions, including the requirements on them to take account of factors additional to the scientific evidence (UK Government, 2010).
For those concerned with the proper conduct of science in government the counter-example to the Ecstasy debate was the sequence of events around the government’s decision in a similar instance, that of the drug, khat. The Council advised that khat was not sufficiently harmful to justify classification. In line with the new guidelines, the Home Secretary of the day met the Council’s Chair and demonstrated that she had understood the basis of their advice. When she nevertheless decided to classify the drug, she wrote to him explaining the other factors that informed her decision, such as the objective of supporting international co-operation between police forces (May, 2013). This form of public reasoning better satisfied those who wished to ensure the scientific evidence was heard, while acknowledging the wider range of factors that inform political decisions in a democracy.
From the perspective of providing evidence to inform the decision, Ecstasy-style disputes are often implicitly disputes about the definitions of the system. Disputes about the definition of the system affect what science is relevant, and therefore also which stakeholder voices have a say. Taking khat as an example, the scientific system might be the human body and the drug’s effect on that, the economic and social effects on the user’s family, on their community, on national health or security budgets and objectives or, as in this case, on international systems of policing whose objectives go far beyond policing khat.
In Tornado politics, those dealing with the crisis will naturally focus on the lives and wellbeing immediately at stake. But what happens during a crisis can also affect what happens when the situation returns to normal, or a new normal. So Tornado politics influence longer term systemic politics around which there is no single decision or pre-defined sequence of decisions. The British response on Fukushima may have informed Japanese willingness subsequently to invest in British nuclear power new build at Hinkley Point, which in turn was an important moment in the evolution of policy on nuclear energy in the UK. Meanwhile, the emergency was also linked to the change in German energy policy that led to the decision to phase out nuclear power.
Decision-makers and their advisers often face the choice between a reasonably certain answer to a partial question and a very uncertain answer to a “better” and more comprehensive question. Ultimately the natural and physical systems considered to surround any policy decision could be extended indefinitely outwards to encompass just about everything on the planet (or extended inwards to quarks, probably). It is also inevitable that sometimes the policy “system” is, at key moments such as in Cabinet government or during budget negotiations, so broad that the decision on one area of policy is made as part of a set of negotiations or trade-offs against a completely unrelated set of policies. It can be hard for those involved in developing specific policy option, whether scientists, officials, citizens or even sometimes Ministers, to see these trade-offs happen. Being resilient to such moments is often a requirement for staying at the interface of science and policy.
Staying with the more traditional physical and temporal scales of policy debate, it may be helpful for scientists to make more explicit why and how they are defining their systems. To caricature: for many scientific disciplines, abstracting elements of a natural system and building models of it in order to enhance understanding is so much at the heart of how they work that it can feel like a moral imperative. The system may be defined by what can be modelled or considered traditionally within a discipline or a group of disciplines used to working together.
In the physical sciences, therefore, there is little reason for the system being studied to coincide with the with policy question. Arguably, one of the challenges of climate science and policy is the fact the evolution of atmospheric and other earth sciences, based on the forms of observational evidence and physical theories available, led to models for which a key indicator of outcomes was the measure of global average surface temperatures. This measure is not well designed to inform public debate and public decision-making, at least at the regional, national or local levels which are where most policy levers lie. In its extreme form the mismatch between what needs to be modelled for a particular purpose and what can most easily be modelled scientifically takes the form of the old joke amongst physicists that they can model anything, provided it can be assumed to be spherical and in a vacuum and so taking the simplest form possible for the purposes of applying the laws of physics. Economists’ assumptions of equilibrium, perfect information and human rationality are a less extreme form of the same fundamental approach to the challenges of understanding, explan...