Sense about scientific advisers part 2: three recommendations

January 19, 2015 at 7:56 pm | Posted in Feature Articles | Leave a comment
Tags: ,

Three recommendations for science advice in the EU

Part 2 of our series on scientific advice in the European Union. Here we present three lessons we believe should be learned from the controversy about the CSA role, before making three recommendations as to how the on-going debate should be framed. (Click here for part 1 of this series.) Click here to download a printer-friendly PDF of this article.

Thinking - maure - stockxchng - email


On 14 November last year, media outlets reported that the position of Chief Scientific Adviser (CSA) to the President of the European Commission would be scrapped, a move met with almost universal condemnation from leading figures in the scientific establishment (click here for a catalogue of media reactions and here for the Science Media Centre’s grab-bag of quotes).

What is interesting about the reactions is how they reveal a great deal of diversity in conceptions of what a CSA is for. Things which commentators variously thought was lost in scrapping the role included:

  • a source of impartial, independent advice;
  • a voice for science;
  • insulation from influence by people with mixed motives and vested interests;
  • a conduit from the world of research to the world of policy-making;
  • a facilitator of international coordination and collaborative efforts in research;
  • a champion or catalyst of scientific advice from across the research spectrum;
  • a symbol of the value of science in Europe;
  • an organiser of dispersed scientific information;
  • an evidence-based approach to policy at the heart of the Commission’s processes;
  • a (sometimes even the) source of scientific input to the policy process;
  • a budget champion;
  • a voice for the research community;
  • an issue expert;
  • an arbitrator in disputes;
  • a translator of complexity;
  • a builder of scientific institutions.

None of these appear unreasonable as expectations of the function and/or objectives of who (or what) is tasked with integrating science into policy making in the EU. Overall, however, there is a shortage of articulation of how exactly a scientific advisory body is expected to deliver on these expectations, in particular how the appointment specifically of a CSA rather than some other approach guarantees these various desirable outcomes.

In the hope of identifying common ground between the various parties in the debate about whether a CSA is an optimal strategy for integrating evidence into EU policy-making, in the following we discuss some of the unifying issues behind the above expectations, try to anticipate some of the pitfalls in achieving the various objectives with which a scientific advisory body might be tasked, and present three considerations for framing the discussions going forward, so as much as possible of what is desirable in the above list may be preserved in the debate.

To do this, we present three lessons we believe should have been learned from the discussion so far, before making three recommendations for how on-going debate should be framed. The recommendations are not intended to be exhaustive (we do not discuss the role of a CSA as a potential budget champion, and only touch on how scientific advisers can act as options czars); the purpose is to challenge some of the assumptions being made about how the appointment of a CSA contributes to science-based policy-making.

Holding policy-makers accountable to the evidence

An often-raised problem is politicians withholding or ignoring awkward research which undermines the case for their pet policy objectives. Examples of this happening are many; in the UK they include research into minimum pricing for alcohol, the increased use of food banks, the effects of immigration on the labour market and the effect of regulation on drug use, all of which was conducted to inform policy yet was apparently delayed or blocked from publication because their results contradicted the policy positions of the UK Government (Sense About Science 2014).

Encouraging governments to acknowledge and respond to awkward evidence instead of ignoring it appears an obvious objective for a scientific advisory body which in some sense “represents the science”. Unfortunately, while intuitively appealing it is not clear if or how the appointment of a scientific adviser actually achieves this.

One problematic outcome would be if a scientific adviser had enough power to overrule government process on the grounds that it does not sufficiently follow the evidence. This would put science on top rather than on tap and, by making government answerable to the adviser, creates the threat that elected government becomes usurped by a shadowy technocracy.

Even the option of giving a scientific adviser less power, in a role of advising and bearing witness should a politician defy the evidence, is riddled with problems. For one thing, when advisers publicly oppose their bosses’ ideas, they tend to get sacked and lose all say in subsequent decision-making –precisely the opposite of integrating science into policy-making.

Senior scientists are also very eminent figures. Even without formal power they can exercise a tremendous amount of influence. This might be a good thing, and may even be an effective way of balancing the on-tap/on-top dilemma in holding politicians to account when they stubbornly chase policy goals which won’t work.

However, in the circumstance of voicing opinion we would want assurance that the statements of the eminent adviser were themselves evidence-based. Since the special function of the science adviser is to introduce respect for evidence into the policy-making process, there needs to be a demonstration that the opinion is scientific rather than personal (we return to the importance of this point in the next section).

A further issue to bear in mind is that, when science and policy appear to be at loggerheads, we need to resist the convenient myth that a body of evidence somehow determines the policy which should be implemented. Policy decisions are messy compromises aimed at achieving desired outcomes which are (in some manner, depending on your political theory) an expression of how the people involved in the decisions want to live.

In most real-world situations there are more possible choices and reasons for choosing them than research can fully evaluate. Besides which, scientific evidence says little about which outcomes are desirable to society and which are not: for example, evidence might show that harm from heroin being available on prescription would be lower than prohibition of heroin, but that won’t necessarily matter if society thinks it is morally wrong to take drugs. The evidence can inform the debate about drug policy but it cannot determine it.

Attempts to resolve values debates such as appropriate drugs policy by simply claiming the scientific high ground confuse the mechanisms for resolving factual debates with the mechanisms for resolving values debates. This confounds the democratic process by obscuring political bartering behind a shield of supposedly scientific debate, reducing the political legitimacy of the outcome at the same time as muddying the contribution and role of science in society.

Lesson 1: In encouraging decision-makers to be accountable to evidence, be cautious that the influence of a scientific advisory body (whether that influence is formal or informal) does not circumvent proper democratic process.

Arbitrating in disputes

Evidence of uncertain validity and relevance is just one weapon deployed in the complex battles fought by vested interests in securing desired policy outcomes. The need for independent evaluation of this evidence and its bearing on potential policy decisions is clear; to those in government, the appeal of having a go-to adviser to perform this task is obvious.

The problem is, it takes a lot more to resolve a scientific debate than simply appointing an expert. Regardless of an apparently instinctive belief that science is hierarchical and an expert by their position has sufficient command of a field to “know” the state of the evidence, the truth is that scientific knowledge is secured not by the status of the individual but instead by the methods they follow. To put it another way: science is a matter of how, not who.

In medicine, we have learned the hard way that expert opinion is unpredictably fallible and that what really counts is the process one follows in finding and evaluating the relevant evidence, with the gold-standard method being the systematic review. So when we ask for an adviser who can “represent the evidence”, what we probably want is not a person to whom we can turn when we want an opinion, but rather someone who can make sure good processes for finding and evaluating evidence are being developed and followed.

Implementing an effective mechanism for integrating scientific evidence into policy-making is far from a straightforward process and much needs to be done in terms of developing reliable methods for distilling what is known out of the uncertainty and discord which characterises scientific debate. Scientific advisers may have a place in implementing these review processes and ensuring they are functional, but this is a governance role which should not be confused with an advisory one.

Lesson 2: Ensuring that evidence is being properly generated and reviewed in the course of decision-making is a governance role, not an advisory one; mandates and oversight mechanisms should be established accordingly.

A conduit from the world of research to the world of policy

A view often stated is that policy-makers are detached from the researchers and institutions which produce so much of the evidence which politicians end up using or ignoring, be it food and agricultural technology, advances in medical care, the ability to anticipate and evaluate risks to health and the environment from economic activity, and so forth.

The extent to which this is the case is difficult to evaluate. Some talk as if there is no link at all between the work of politicians and scientists, particularly when discussing the activities of the European Commission, as if a CSA is a major (or even the only) connection between the Commission’s work and the scientific evidence. This must be an exaggeration: the machinery of the European Union is littered with scientific committees and institutions responsible for appraising evidence for policy purposes. While they may not function as effectively as everyone would like, they do exist.

What the notion of a conduit may more subtly and usefully allude to are the less tangible benefits of the existence of a chief scientific adviser, of how close communication between a member of the research community and a member of government can elevate the profile of science in general, shaping a context in which science is more familiar to and better-valued by a political elite which rarely comes from a scientific background. The use of science advisers could open up institutions such as the EU Commission to a more sympathetic view to funding for research and make them more receptive to using science in policy making, by virtue of familiarity with scientific interests.

The overall gains from this relationship are difficult to qualify; perhaps some more work could be done to articulate this rather than leave the understanding implicit. Heading off any threats to effective function of the position in terms of accountability (it would surely need to be transparent) and also preventing the sorts of conflict between scientist and regulator which undermine the relationship we would be trying to nurture (politicians get to decide their advisory structures and won’t put up with scientists who embarrass them, however correct they are to do so), would be a necessary and difficult balancing act.

Some might argue that the new Commission’s eagerness to dispense with the role of CSA, coupled with the absence of proposals for integrating scientific research into policy-making, is just another manifestation of the low value placed by politicians on the role of science in policy-making. One can only speculate. If so, however, it should be noted that shifting the cultural attitude of politicians is a big ask for anyone and there should be no assumptions made that simply having a CSA contributes very much at all to achieving it.

This leads us to a general point about not confusing the means to an end with the end itself. A number of commentators describe the position of CSA as being symbolic of a commitment to integrating science into policy, but there is no point in having the symbol if the commitment is not there – in a way, it is better if it is not, because then the Commission is not posturing without genuine intent.

Lesson 3: Elevating the value of science in political decision-making is a challenging cultural shift; the mere appointment of a scientific adviser is no guarantee this shift is being made; without appropriate oversight mechanisms and mandate for the position it could do more harm than good.

Next steps

It must be possible for a sensible mandate and institutional setting for integration of science into policy to be worked out; although the individual concerns about what the loss of the CSA role means for science in EU policy-making are varied and incomplete, they do seem to add up to a general picture of what it would mean for science to be integrated into policy-making.

That said, it is not sufficient for a partial view of the problem to result in ad hoc and ill-considered introduction of a novel position with vague mandate, insufficient budget, wide powers and influence etc. in the face of a lack of evidence or even rationale for thinking that the position will do more good than harm. The sooner constructive discussion between the various parties with an interest in scientific advice can begin, the better.

The following objectives seem to follow from the above discussion. In no case need it be assumed that appointment of a scientific adviser is necessary or appropriate to achieving the outcome; other mechanisms for achieving the desired results will exist and may be preferable (for example, cultures of understanding may be best fostered by outreach strategies from learned societies). As such, form should always follow function.

1. Find the right balance between making policy-makers accountable in their use of scientific research and providing protections for due democratic process in decision-making.

In this discussion we are concerned with ensuring that science remains on tap but not on top, while policy-makers are nonetheless discouraged from simply cherry-picking their way through the evidence for their positions or framing research questions in such a way as to produce policy-based evidence rather than evidence-based policy.

2. Promote a culture of understanding of the processes and value of scientific research within the policy-making community, and likewise policy-making in the scientific community.

Here, we are concerned with discouraging policy stakeholders from scientising values debates, and creating a political culture in which the value of robust scientific research is more fully appreciated.

3. Give close consideration to how a body might coordinate and provide oversight of research conducted for policy purposes, in a transparent, objective and comprehensive manner, with an emphasis on process, understanding this as a governance rather than advisory role.

This is to ensure best practice is defined and followed when evidence is gathered and analysed, so research can be commissioned and reviewed in a timely and transparent manner, producing valid results of maximum utility. This is an essential precondition for scientific agencies to arbitrate in factual disputes and for insulation of scientific advice from vested interests. The challenge is in ensuring that good practice is followed without regulatory processes becoming bogged down in disputes about scientific processes.

Leave a Comment »

RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Blog at
Entries and comments feeds.

%d bloggers like this: