3 Comments

My title is the title of Prof. Peter Gluckman’s guest lecture to the Wolfram Institute, Durham University (UK) on January 22nd, available online as a PDF file.

As a teaser to encourage readers to explore it, I’ve excerpted some passages below and in some cases offering my own thoughts.

There’s much there that readers and fellow scibloggers will see echoes or want to ponder.

Gluckman opens by explaining his roles as Chief Science Advisor to the Prime Minister. Without listing all five bulleted points he offers and subsequent observations it might be condensed to:

  • To present to science-related issues to the Prime Minister. (There are several aspects to this – read them for a better understanding of his position; these elements are not my specific interest here.)
  • To encourage better use of evidence in policy.

On the former, I’m cautiously optimistic. My main concern is that his description has him delivering mainly, even exclusively, to the Prime Minister, not to the wider parliament or the public (including media). My own views are that this must limit how much he can achieve. I worry a little about how the information is used, given it is (nominally, at least) only delivered to the one person. Wouldn’t delivering the advice to all, open to the public record, be more likely to ensure that the advice is used soundly? (Gluckman makes the good point that the advisor cannot get involved with the public opinion aspects, but it seem to me that this falls under his ‘honest broker’ model, covered later, rather than specifically if the information is open to all or not.)

On the latter – I’m certainly with this. I actually wrote advocating this only a short time prior to his position being formed; point there being that my opinion on this does not follow his, but is something I have independently come to believe.

In particular he goes on to say,

My view is clear — the use of high quality information and evidence should be at the core of good decision-making for good outcomes. Decisions made in the absence of scientific knowledge can only be made on the basis of either anecdotal experience or belief and dogma — they are the only other sources of knowledge.

then briefly elaborate on anecdote and experience, something readers here will know I and others at sciblogs have expounded upon frequently. (A key point, of course, is that you really have to test things before you can know what the story actually is.)

Relevant to readers here, too, he notes ’Another emotive example — one that I suspect I will soon have to engage with — is that of health claims associated with alternative medicines and herbal-type products.’ We cover our fair share of this on this at sciblogs and I have to admit I am left wondering precisely what he is anticipating having to deal with. Possibly this relates to the Natural Health Products Bill?

In this context he cites from Tim Minchin’s now famous beat poem, Storm. If you haven’t already listened to this, do – it is excellent. (So much so, that I paused here to listen to it again.)

Moving on, Gluckman offers a definition of science that appeals to him – what do readers think? -

I like a definition put forward by Jonathan Marks in his book Why I Am Not a Scientist (here I should point out that Jonathan is indeed a well published scientist and was a molecular and evolutionary biologist who later turned to social anthropology). His definition is ’Science is the production of convincing knowledge in modern society’.

It is worth parsing this definition:

Science is not passive — it is the product of a process. It requires an active process of observation by a primed observer and therefore the production of science is highly context specific.

Convincing — there is a social process beyond mere discovery or fact production. The fact has to become accepted in order for other scientists to incorporate and build upon it.

Modern society — while these words may seem controversial, I think it is now pretty well accepted that science as we know it now started with the ideas, values and social practices that arose in Europe and its colonies in the enlightenment, which ‘separated’ the natural world from the spiritual and moral worlds and moved from a Platonic to what we now might call a Baconian and Popperian approach: namely an iterative process of experiment or observation, hypothesis testing and reformation until knowledge considered to be reliable is developed. The key point is that science is not the facts themselves, science is a process by which we make our best efforts to understand what is going on in the universe, in the natural and social world, and in ourselves.

I largely agree with this, although I’m not sure I would have parsed or interpreted the phrase exactly the same way myself. (The points he makes in the parsing are good; it is the particular interpretation of the phrase that I’m not sure I agree with. For another of many accounts of ‘what science is’, if perhaps less formal, you could try Dr. Harriet Halls’ What is Science?)

This he follows with a caution that science-based knowledge is not, in itself, sufficient for policy, using New Zealand’s folate supplementation of bread debate as an example. This particular example, as he relates it, seems to political strategy, but nevertheless there are issues beyond the science itself as he points out. (For what it’s worth, some breads appear to be supplemented with folate despite the opposition to this and not necessarily promoted as such either. Anecdotal, but it strikes me as possibly an illustration that some of the opposition claims don’t hold up, e.g. that it would cost the bread industry too much, affect the taste badly, etc.)

In particular, Gluckman raises the point that science of complex, non-linear, systems changes the game as things get expressed in likelihoods rather than certainties.[1] A key issue is that this runs into the interpretation of risk and what is considered ‘sufficient’ (lack of) risk. There is quite a bit to this and I would encourage readers to read his own take on this – there is little point in me repeating it here.

Finally he turns to how science should deal with ’the body politic’. I agree with his point that ’Advisors or advisory committees must not be advocates for anything other than the knowledge base, but must act as honest brokers’ (this was part of idea behind my concept of a NZ Science Party). His discussion distinguishes between ’the issues advocate and that of being the honest broker’.[2]

Many here will have thoughts on this remark,

As a science advisor, clarity of communication is essential even when dealing with concepts of risk and probability. But this is occurring against the background of scientists being poorly trained to communicate, a public with variable scientific literacy, and a new range of media which is increasingly unfiltered and no longer takes on its traditional role of discerning reliable from non-reliable information.

There’s so much here I could say I might be best to not even start, or, alternatively, set it in another article exploring just this one thing. One thing that I will say here is that science blogs have a role in the latter, teasing apart for others reliable and unreliable information.

I’ll leave you with his penultimate paragraph,

Thus the nature of scientific advice has evolved — first from a ‘decisionist’ model in which politicians decide the goals and scientists provide the tools to achieve them, then to an authoritarian ‘technocratic’ model originating from the absolutism of mid-20th century science which we have discussed. Then, in turn, to a more complex model, termed by Millstone as ‘co-productionist’, in which it is accepted that there is an intertwining of science and technology in the way people live their lives, that for that reason they have a legitimate interest in science-based decisions, and that for policy makers science defines the boundaries of uncertainty within complex systems.

Footnotes

1. I have an article in draft that touches on this in a different context, so-called ‘personalised’ medicine.

2. I have to admit here that as a consultant I see echoes of the issues of advising clients here.


Other articles on Code for life:

Time for disclaimers on remedies?, ’alternative’ or not

Elections — time for policies to be deposited in advance?

When is a scientific paper political campaigning?

Why (some) people don’t trust science

What should be taught in science communication courses?