Confuse science with politics at your peril

When analysis and policy collide the results are often messy, sometimes awful

One of the frequent refrains from politicians during the pandemic was that they were “following The Science”. The implication, that they were being offered scientific advice which led to clear, direct policy decisions - almost as if no human judgement was required between the gathering of data and the implementation of government edicts.

In October 2020, the scientific journal Nature caused a stir by explicitly endorsing Joe Biden for the US presidential elections. Fast-forward 2 years, and a randomised study (published in Nature Human Behaviour) has demonstrated that this decision led to no change in opinion as to who people should vote for, but led to a substantial decrease in trust in the journal from US Republicans and a decrease in trust in science in general. A damning result, and one would think worthy of serious reflection.

Instead, Nature published an accompanying editorial which amounts to a long winded version of, “Sure the study says what we did had several harms and no obvious up-sides, but we had to do it because Trump was really bad”.

We need to talk about science and politics.

What is “following the science?”

This problem is not new, but has been at the forefront of many peoples minds since the start of the pandemic. There has been not so much a blurring of the line between science and politics as total confabulation about their relationship.

This has been a particularly acute issue from random, self selected “independent” scientific advisory groups which have sprung up in the UK, Australia and the US. The hallmark of these groups has been a covert (or sometimes overt) political affiliation of the members, with suppression (or even exclusion) of dissenting voices, offering strong opinions on policy which are dressed up as simply being “The Science™”. Policy IS politics.

Science will not and cannot tell you what policy decisions should be made.

Science can generate data. Scientists can interpret the meaning of the data and the implications of it. Science can estimate what the potential outcomes of different policy choices might be, and can help establish what choices might optimise different outcomes. Policy choices however, have to weigh up intangible things such as values.

These values may differ between individuals, regions and cultures, and the idea of having democratic systems is to allow these values to be represented by elected officials. Take the way right or left leaning politics value the freedoms and responsibility of inviduals vs the rights and responsibilities of the collective, respectively. Science has nothing to say about which of these values is right or wrong, better or worse. There may be situations where balancing these differing values leads to greatly differing outcomes. Science can help inform us as to these trade offs, but it cannot make a choice.

Make no mistake. There is no such thing as “following the science”, and hearing these words is a giant red flag.

What’s the problem?

Scientists of course are allowed political opinions. These opinions on what policies should or should not be enacted will be informed by their knowledge of science. But it will also be informed by their worldview, which not everyone will share. It is by definition the opinion of a human; albeit an opinion which may be more informed on the science relevant to the issue. This is why scientists with an equal grasp on the subject matter may have wildly differing opinions on the policy implications.

Whilst there is nothing wrong with this, it is dangerous to be ignorant of this issue. As we have seen from the Nature study, when scientists nakedly weight into politics without making clear the distinction between personal opinion and the scientific endeavour, they create confusion (seemingly the scientists themselves may be confused about this distinction) and foster distrust of science itself.

This is dangerous. We live in an age where social media has made it easier than ever to sow distrust in scientific institutions. Trust in science (and health care) are vital to improving the health of populations and moving society forward. That trust does not come cheap and must be earned. It is infinitely harder to restore than it is to break. Scientific journals should not be weighing in on personal political matters, especially when we have clear evidence it amounts to no tangible benefits and obvious harms.

I was troubled by comments on twitter from the editor of the journal “Science”, on the aforementioned articles in Nature:

This gives people the permission to say things like "climate change may be real, but I don't think we should have government regulation to deal with it," which is unacceptable. We can't concede that by letting people pick and choose.

This is a classic confusion of the role of scientists in politics. Of course, a scientist is perfectly allowed to have an opinion on the policy implications of climate change, as is anyone else. Climate scientists are more informed on the scientific elements of this than most others. But when it comes to the values or where and how the government can and should intervene, science has no say. A libertarian and a communist will not agree on the right policies on carbon emissions, even if they are in 100% agreement on the science informing the issue.

Where there is a scientific case to be made for or against the claims of a policy, journals are the place to make it. If reducing carbon emission will reduce global temperatures by X amount, science can inform us on this. But science cannot make decisions on the role of government in imposing restrictions on private businesses, tax breaks, incentives etc to achieve this.

If you the editor of a scientific journal and want to lobby on behalf of your preferred policy, go ahead. Write to a newspaper. Write to your political representative. Make the scientific case for the outcomes of your policy choice.

But be clear in your own mind and when making your case that this is your personal opinion informed by science. Not some great ordained political truth divined directly from the mouths of the great ethereal data Gods. Most importantly, do not print your opinion on the front page of the scientific journal which you edit.

Closing thoughts

All of this is before we even consider that people with an equal grasp of a scientific subject may not even agree on how to interpret it, making a mockery of considering “The Science™” as any singular viewpoint. There are very few areas of science that are anywhere near settled, and a lack of acknowledgement of uncertainty is a red flag in its own right. The pandemic gave us some of the best examples of these and we would be wise to learn from them.

A final thought: also be aware of people who claim that science itself is inherently and unavoidably political. Whilst it is true that science has to be done by humans and so will always involve decisions which are political by nature, this is a flaw in the scientific endeavour which should be minimised, not celebrated. Stuart Richie has written an excellent piece on this topic which I would recommend.

Science is political - and that's a bad thing

Summary

Science can inform policy choices, but cannot make them for us. These choices need to consider the different values we hold as humans, which is the role of our elected representatives. Scientists who are ignorant of these differences cause serious harm to trust in science by confusing their scientific and political messaging. There is no role for personal politics in scientific journals.

I will leave you with a comment on the matter from one of my heroes: David Spiegelhalter,

“Stop epidemiologists from giving public health advice. Epidemiology has nothing to do with what people should do”