Science and “mere opinion”: a dilemma wrapped in an irony

The call for evidenced-based policy is a dilemma wrapped in an irony. I’m not sure if it is necessarily like that but it seems that way on the evidence so far. It was certainly evident in the recent New Statesman piece written by Brian Cox and Robin Ince. The Twitter discussion that followed was lively though unfruitful. “Positivist” accused one. “Nobber” was the witty reply.

For my part it was probably just bad timing. I was in the middle of marking a pile of essays when I read the piece and so naturally was in the frame of mind of wondering what grade to give it: top marks for communication and argument, poor marks for research and evidence.

And here is the irony, where was the evidence in their argument for evidenced-based policy?

For example, in pointing to the past century of unparalleled medical, technological and intellectual advances where was their evidence for their claim that “the foundation for these changes is the scientific method”? Historians of science and technology will tell you it’s not quite as simple as that. Technology often leads to science rather than the other way round (maybe more often than not). The classic example is steam engines leading to the development of thermodynamics rather than theories of heat leading to the development of steam engines.

Indeed, where is the evidence for them to speak of “the” scientific method in the first place? Science is littered with failed attempts at defining itself and it is doubtful whether a single method can unify such diverse practices as theoretical physics, palaeontology and genetics.

There is also an element of having your cake and eating it when scientists claim credit for technological progress. In the nineteenth century scientists had distanced themselves from the public as part of a process of self-definition, a necessary separation to establish their own realm of professional expertise. By the mid-twentieth century (after the technological horrors of two world wars) such distancing became a convenient tactic for avoiding accountability by separating the “pure” practice of science (done by scientists) from the “abuse” done to it by others.

However, avoiding accountability should also mean relinquishing the credit for those developments which the public would highlight for praise. If scientists wish to be associated with what others do with science then they should take the rough with the smooth. Alternatively, we could drop the language of use and abuse, and simply say this is what science is and what science does whether we like it or not.

I should stress that this is NOT an argument against evidence-based policy. If the evidence is there we should not ignore it and if we don’t have the evidence available then we should do our best to find it. I have a naïve belief that if I wish to interact with the world then I would like to know something about it before I do. I say “naïve” because, like most things, this belief gets more complicated on closer inspection. This is not only because it raises the question of what counts as “evidence”, but also who gets to say what counts as evidence and on what grounds.

And here is where the irony conceals the dilemma.

One of the points in Mark Henderson’s Geek Manifesto is: “We want science and critical thinking to become central to the national conversation” (p247). It is in that spirit that we should apply our critical thinking to science itself.

But who should we turn to for that critical thinking? It is an epistemological version of the “Quis custodiet” problem (ie. “who should guard the guardians”) – who should investigate the investigators?

I have written elsewhere of the difficulties of using science to explain science. Would we seriously want to use chemistry and physics to explain what a chemist and physicist does? Presumably we would need to go outside science (or at least into the social “sciences”) for our critical thinking.

If we are to turn our critical eye upon science itself then what is it that we will be examining? This brings us back to the piece by Cox and Ince who tell us that

science is a process, a series of structures that allow us, in as unbiased a way as possible, to test our assertions against Nature.

And that,

Science is the only way we have of exploring nature, and nature exists outside of human structures

This, I think, is very revealing, because it seems to claim a sense of objectivity by association: nature is “outside human structures”, science tests against nature, therefore…..the implied suggestion being that science itself is somehow outside human structures.

“No, no”, might be the reply, “It doesn’t say that, or imply that.” In which case let us put science firmly back inside “human structures” and examine it as another example of what humans do individually, collectively, socially, politically, culturally, historically. And for that type of investigation who you gonna call? Or to paraphrase the Bill Murray character: “Back off. I’m a historian.”

So, to take our cue once more from Cox and Ince, let us say that “science is a process, a series of structures”. What kind of “process” and “structures” do they mean? Is that a social process? A series of epistemological structures?

If it is a social process then let’s open it up to sociologists. If it is a structure of ideas then let the philosophers take a peek. If the structures are a set of values then that is something historians have some expertise in examining. In fact, philosophers, sociologists and historians could look at each of these.

Unfortunately the History and Philosophy of Science, together with the Social Studies of Science is just the kind of expertise that all too often gets rejected by those who wish to defend science. The critical thinking that was demanded in the Geek Manifesto is rejected when the object of scrutiny is science itself.

If we were to allow the critical gaze to fall on science then we would see how problematic is Cox and Ince’s attempt to separate science from politics. “There must be a place where science stops and politics begins,” they say and in that “must” I can’t help but hear an accompanying “please let there be”. If only if it were that simple. They do add that “this border is an extremely complex and uncomfortable one” but the complexity and discomfort is all on the policy side with no acknowledgement how politics (in all its macro and micro varieties) can be complex and uncomfortable within science.

Finally, the ghost at the banquet is the spectre of “anti-science” only this time presented as a “lifestyle choice” to abandon evidence in favour of “humbug and charlatanism”. The threat is that “mere opinion” is elevated above science. However, by devaluing opinion in this way they expose the asymmetry in what is supposed to be a new dialogue between science and public: science has facts, the public has opinions.

Sadly it is not the science-public dialogue that some of us hoped for.


For a thoughtful piece on the Cox/Ince article and the Twitter debate that followed see the post by Rebekah Higgitt in the Guardian.

A similarly well-considered piece on the Cox/Ince artice and the relationship between science and politics can be found in Jack Stilgoe’s post in the Guardian.

The two posts by Higgitt and Stilgoe say much of what I would like to have said here especially where they relate to the idea that HPS and STS each have evidence that Cox and Ince ignore (thus underlining the irony and the dilemma I tried to point out).

For a good piece on the Cox and Ince article and why the “geek movement” is bad for science see the post by Haralambos Dayantis.

And a post from Martin Robbins on why the Dayantis piece is not very good at all

A post from Ken Perrott defends the Cox/Ince article with respect to some of the issues raised by Stilgoe, Higgit and Dayantis.

A lovely insight into how science really works can be found on Twitter through the hashtag #overlyhonestmethods. Not only do the tweets underline the HPS/STS perspective of science-as-human-activity, but they are also very funny For a taste of the tweets and how they relate to STS see “Science and its #overlyhonestmethods

For relationship of science and technology (and relationship of historians of science and historians of technology) see the post by Patrick McCray

On the importance of history of science and the need for historians to communicate it properly see this post from Emily Winterburn.

Communicating history of science was something we used to do quite well. Here’s James Burke in his series “The Day the Universe Changed” where he explains that “The view of things at the time controls what science does at every level”. Raw data is not very raw at all. We only get what we are looking for (about 24 mins in is a good place to start) See also his final thoughts on science, values and different ways of seeing the world taken from same series. The kind of popular HPS/STS that would be valuable now.

For some sensible ideas about the whole discussion and Twitterspat generated by the Cox/Ince editorial see post by Jon Butterworth in the Guardian

24 thoughts on “Science and “mere opinion”: a dilemma wrapped in an irony

  1. “Unfortunately the History and Philosophy of Science, together with the Social Studies of Science is just the kind of expertise that all too often gets rejected by those who wish to defend science.”

    In the somewhat heated twitter discussion Cox explicitly did reject these disciplines claiming that “70% of physicist do so (with right!).

  2. Really enjoyed your post, but I can’t help thinking this debate is a series of ever decreasing circles. What I found most interesting yesterday was the level of agreement between people when talking directly.

    As for the geek thing, who and where are they?!

    1. Agreed it often seems like both sides occupying the same ground but determined to face in opposite directions. Having said that I think there are real differences in thinking about what science is – something which HPS and STS are quite good at examining!.

      The geeks? Tend to be self-proclaimed, taken as a badge of honour.

  3. Having read quite a lot the various articles around this issue, I generally thought the criticism of Cox and Ince was, largely, missing the point. Maybe I’m giving them too much credit, but I just felt that they were arguing for more evidence based policy and were assuming that the “scientific method” would eventually result in suitable evidence. You do, however, make a valid point about “who should investigate the investigators”. Naively, one might expect science to – over time – converge towards some kind of result that matches reality/nature. There would be theoretical calculations, computer models, experiments, observations and the experiments/observations would rule out all but the most suitable models (my terminology may not strictly correct, but you probably get the idea). In many cases, I suspect that this would be true.

    However, there are areas where there is concern, one of which is the pharmaceutical industry (others might say climate change, but I would disagree). There is quite a strong drive (by Ben Goldacre for example) to put pressure on the pharmaceutical industry to publish more of their results so as to properly show the effectiveness of new treatments. There’s also a similar push by others in the medical profession to properly test medical screenings. This may be an example of an attempt to investigate the investigators, but I’ve always seen this as consistent with what Cox and Ince (for example) are trying to do. The criticism is always that they’re not properly following the scientific method. For example, the tests are poorly designed or not all the data is published. It seems, to me at least, that a fundamental aspect of what is happening at the moment is pressure to properly use the scientific method which was developed in the past (I believe) by philosophers and other social scientists. The assumption they are making (I believe) is that the scientific method is now well established. It now just has to be used properly to inform policy (whether that be by the government, by the medical profession or any other organisation for which this might be appropriate). Maybe there is still a role for philosophers/social scientists, but it’s not quite clear – to me at least – what this role would be. There’s always a role in these discussions, but I can’t quite see how a philosopher (for example) could check that some experiment was properly set and that the data was suitably analysed.

    I guess the argument I’m making is that the scientific method exists and if followed properly will provide evidence to inform policy decisions. The main thing we have to be careful of is situations where some can apply this poorly so as to provide evidence that suits their agenda and I guess some form of regulation might be needed to combat this, but it doesn’t – in my view at least – suggest that we shoudn’t be aiming for evidence-based policy making. Sorry, this is rather longer than I had intended. Hope it makes some kind of sense.

    1. Thanks for the comment and there is a lot of sense in what you say. I agree with you about the need for evidenced-based policy (or at least evidence-informed policy). I also think that there are several methods that can help supply that evidence (such as double-blind trials etc). And the work of Ben Goldacre in exposing the deficiencies in trials and policies is admirable. All of this is consistent with what you argue and what Cox and Ince say. So what am I quibbling about?

      I think it comes down to the question of evidence and expertise. If I want to know about the physics of something then I’ll ask a physicist. If I want to know the history of something then I’ll ask a historian. Unfortunately Cox and Ince don’t seem to have consulted any historians to support what they say. Similarly, if I want to know science as a social practice, or as a system of objects, behaviours and practices then I think it best consult the people who have the expertise in this (eg from Social Studies of Science). And again, STS is the kind of expertise that they seem not just to ignore but to dismiss.

      Think of a comparable situation. Someone who doesn’t know physics makes statements that sound like New Age crap to a physicist. It’s quite common for those statements not only to be dismissed as wrong but also as somehow to be “anti-science”. Now turn the situation back round again. Someone who doesn’t know history makes statements that sound naive to a historian. Should we be surprised if the historian dismisses those statements as wrong? They might even shriek about “anti-history” (but they don’t).

      The problem of “investigating the investigators” isn’t just one of holding people to account. You are quite right in highlighting the need in holding science to account (and actually most of the time science is pretty good and doing it itself). The problem I tried to highlight was simply the one of who has the expertise to study science as something that is done by people in particular social, cultural, political, economic and historical contexts. I think the kind of expertise we would look for comes from HPS and STS.

      Hmm, quite a long reply. Hope it makes sense. Maybe I should do another post 🙂

      1. Thanks for replying. Interesting, I hadn’t quite thought of it in that way. I would probably agree that there will be areas of history/sociology and other social sciences that many scientists don’t understand well (and may sound ill-informed if they discuss things in these areas). In fact, we all talk about “the scientific method” but few (myself included) will have read Hume. I don’t see this as a major issue since most learn the scientific method as they go through their careers (from lecturers, PhD supervisors, colleagues). I can well imagine that historians and others in the social sciences could play a crucial role in developing our society so as to accept evidence-informed policy making.

        What I would say though is that – having followed various twitter conversations today by some of those involved – some of the problem is due to the difference between communicating science and doing science. I think that Brian Cox, Robin Ince and others really see themselves as science communicators (even though Brian Cox is an active scientist). They see themselves as telling the story and engaging with the public/policy makers. They may not go into all the details and all the history; they just want to get the basic message across. I think there is some merit in this. If the message starts to be taken seriously others can get involved to develop it properly and thoroughly. I think we do have to be careful about getting bogged down in details when what we’re trying to do now is to simply get the basic message out that evidence-informed policy making is what we should be aiming for.

  4. Again much to agree on but I would add that the “story” could be so much better if it included what historians have to say (and that doesn’t have to be the pedantic details). The converse is that it could undermine the message if it’s apparent that they don’t know what they are talking about and if they dismiss the efforts to help them get it right. Others might be better able to judge the extent to which this has happened but it clearly is a danger.

    1. I agree with you here. There’s no reason why the story couldn’t be expanded to include what historians and other have to say and this could clearly be done without getting bogged down in details. I’ve never been a big fan of scientists dismissing the contribution that has been made (or could be made) by the social sciences. It does come across as unnecessarily arrogant.

    2. Adding more stuff to a story makes it harder to get across, so that’s one reason why, as a writer, I hesitate to ‘expand’ stories. A lot of this debate from the science history/philosophy side seems to be “we want more of our stuff in what you’re doing.” Well, okay, so explain to me in language I understand why that is. Many of these arguments seem to presuppose that adding more facts and information to a narrative make it more persuasive; and comments like the above seem to act as though it *must* be useful “because history/philosophy”. None of this is really likely to bring people onside.

      They may sound like simple questions, but

      1. It’s not a question of adding more stuff to stories. It’s about getting the stories right in the first place. And I have to disagree when you think that history will not bring people onside. History is extremely popular in all media (written, broadcast, digital). History of science can also be done extremely well (personally I like Jim Al-Khalili’s stuff and like many of my generation was inspired by Bronowski). In fact, history naturally lends itself to story telling (it IS story telling). I know historians can quibble over details in popular narrative histories but I’m quite happy for them to do that off-screen (and so with no harmful side-effects for the public).

        What is a problem is alienating supporters by dismissing their expertise.

      2. To a certain extent this is what I was suggesting in an earlier comment. There is a skill to communicating science and it does require an element of telling a story without getting bogged down in details. Having said that, there has to be some care taken that what is communicated doesn’t dismiss the role other disciplines have played and could still play. Of course, people from those disciplines could get involved and communicate their own story. I tend to agree with what I think you are saying. Good science communicators shouldn’t be criticised simply because they didn’t include what someone else – often an expert who doesn’t necessarily appreciate what public engagement is about – thinks is important.

      3. Yes, I’m sure history is very popular – I like it myself – but not everything is about history, least of all the New Statesman article in question, which is a rallying call for progress aimed at a politically progressive audience. I’m all for accuracy (and the last bit of their 2nd paragraph grated on me slightly until I reread it and understood what it was shorthand for), but pedantry can be the enemy of good rhetoric, and beyond accuracy a lot of this seems to be a case of “when you have a hammer, everything looks like a nail.” And when you talk about ‘alienating supporters’, what you actually mean is ‘alienating me,’ which is fine, but don’t assume that you’re the intended audience, or that an editorial in a political magazine is an academic exercise.

  5. Martin, are you arguing the piece is political/campaigning rhetoric and therefore can legitimately be economical with the actualité? That’s an astonishing argument to make in defence of a piece urging better attention to evidence. Maybe that’s not what you are saying, but it reads a lot like that. Are you surprised some of us fear an unstated hierarchy of expertise lies behind this vspat?

  6. Martin, we all agree on the main point of the article (something that I know Rebekah Higgit has tried to point out as well) and I do accept that pedantic accuracy can get in the way of good rhetoric. But there is also a related problem that inaccuracy can get in the way of good rhetoric. By this I mean not that any inaccuracies will get aged historians coughing into their sherry but it can set the wrong tone. The NS article was a good strong argument (I say that at the start of my post), but the tone of “outside science is mere opinion” and that somehow science is separate from politics could sound arrogant even to people not versed in STS.

    I think I might have to disappoint you about my feeling alienated. I think I’m too long out of the game for that. But I do know it is the kind of thing that does piss people off, in the same way that New Agers probably piss off physicists.

  7. I think we should not forget issues of language and genre in all this. We are talking about an editorial, not an academic article. This genre demands a bit of shortness and pithiness, so to speak. So that’s one constraint. The other constraint is the language we have at our disposition to say things with. In that language we have two words, science and politics. When we talk about making political decisions based on science, we separate science from politics, although in reality they may not be as separable as our language makes us believe. And of course there is a lot of historical depth to all of this, but again, that’s not easy to convey in a short piece of text.two words

    1. Yes, agreed and I must confess to using a few pithy statements in my time as a union branch chair. Hopefully the overall impression I conveyed was both accurate and helped to further the cause. My worry is that the overall impression in the NS piece might not be helpful. Yes, it provides a good rallying call (and I think many of us have agreed on that) but even with a “progressive audience” the tone might put some people off.

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 )

Google photo

You are commenting using your Google 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 )

Connecting to %s