Peter Vickers on future-proof facts and public policy


The Sun orbits the Earth. Light travels through space via a luminiferous ether. All diseases are transmitted by bad air. I think it’s fair to say some ‘facts’ from the history of science have aged like milk (which can, incidentally, definitely transmit disease). In this context, how can we determine which of today’s scientific claims will last forever, and which we should take with pinch (or few) of salt?

Peter Vickers, Professor of Philosophy of Science at Durham University and author of the recent book Identifying Future Proof Science published by Oxford University Press, thinks he’s found the answer – and it doesn’t involve looking at the primary evidence in scientific journals, attending conferences or scouring reports.

Identifying Future Proof Science is, in his own words, “an attempt to identify scientific claims that are established scientific facts in the sense that they’re future proof. They’re not going to be overturned by scientific revolution in the next 100 years, or even in the next 1,000 years”.

Vickers argues that a claim is future-proof if at least 95% of relevant experts in a large, international and diverse scientific community would agree that it is a ‘scientific fact’. I’ll leave it as an exercise to come up with counter examples, though the book already debunks many of the most obvious ones.

The diversity criterion in particular is doing a lot of heavy lifting. Many apparent counter examples from the history of science, namely the dogmatic belief in classical mechanics in the 19th century, may meet the 95% criteria but fail on diversity.

“The physics community in the 19th century was nearly all, you know, middle to upper class white men. So, it’s sort of a cheeky way to rule that case out.” Lord Kelvin, who reportedly stated at the turn of the 20th century ‘There is nothing new to be discovered in physics now’ definitely fit that description.

Humanity doesn’t have a good way to measure scientific community opinion

But while Vickers argues that some claims are future-proof, that doesn’t mean that nothing is going to change over the next centuries, or even millennia. “You might look back 1,000 years and say, look at what’s changed, won’t loads change in the next 1,000 years? And I think it will.” Nevertheless, Vickers claims that “some scientific claims [are] are so firmly established that we can have full confidence that they will still be believed in the future.”

What kind of claims does Vickers believe are established scientific facts? “The claims that are being made are […] not ambitious claims, they’re just things that basically everybody accepts and has accepted for years and have been established over a period of many decades with a huge amount of research”.

“Or even something more banal than that, like the water cycle. Like, why doesn’t the sky run out of rain? You know, things like that. I think we know there’s not going to be a new theory of the rain anytime, even in 1,000 years or 5,000 years”.

But the work isn’t just relevant to academically establishing whether simple facts can be thought to be “beyond reasonable doubt” or not. Scientific consensus is also key to implementing appropriate policy, with climate change and public health pertinent areas.

“IPCC [Intergovernmental Panel on Climate Change] authors do have to write after their statements, ‘low confidence’ or ‘medium confidence’ or ‘high confidence’ or sometimes they use ‘very high confidence.’ One of the things that inspired me was reading this article where they interviewed some IPCC report authors and asked them about their opinions on these qualifications, and the person replied that they need a way to distinguish high confidence from scientific fact, and they don’t really have one”.

Vickers also cites a Nature paper by researchers from the Czech Republic, which showed that properly informing the public of the medical consensus on the safety of Covid-19 vaccines increased vaccine uptake, as inspiration. “In the back in the back of my mind there’s this idea that you can influence public opinion, which in turn influences what policymakers can actually bring in without getting voted out of the next election”.

The problem is “humanity doesn’t have a good way to measure scientific community opinion.” Vickers is keen to fix this and is involved in a pilot study for a future institute (“I’m calling it an institute, but it’s not an institute yet.”).

People want to make up their own minds about things, otherwise it’s too Big Brother

The aim of the institute would be “to ask tens of thousands of scientists a question” and “to very quickly, within a couple of weeks, get a response back on the scale of, you know, many thousands or tens of thousands, where we can say ‘this reflects scientific community opinion’.” That would allow the IPCC to more rigorously apply confidence ratings and public health bodies to produce more effective messaging.

But in the provisional institute’s provisional name – the Institute for Ascertaining Scientific Consensus (IASC) – there’s something missing compared to the rest of Vickers’s work: the word ‘fact’. While there’s a “core” of the team that would support using the term ‘fact’, it was (aptly?) difficult to achieve consensus amongst such a “big international team”. It seems Vickers’s criteria might not have passed its own test, yet.

The main point of dissent within the team seems to be that ‘facts’ could be seen as too paternalistic. “People want to make up their own minds about things, otherwise it’s too Big Brother.”

“So what you can say is ‘look, you know, we’ve measured opinion, and here’s the result’ and then hopefully people would think ‘well if 75% of the experts think it’s inevitable to reach 1.5 degrees, that’s quite high, and the consequences of reaching more than 1.5 degrees are really bad, so maybe we should act”.

Vickers hopes that there are some cases where “you would sway community opinion and that would help policymakers to bring in policies which […] the public would support because we’ve got this information about where the policy is coming from”.

Let’s consider some more modern scientific claims. Smoking causes cancer. The Universe is 13.8 billion years old. Humans evolved from fish.

To some, they will all be obviously true. To others, the currency of science is models, with facts being largely irrelevant. Many would further argue that given the turbulent history of science its hubristic to talk of scientific ‘facts’ at all.

Vickers points to the influence of the titans of 20th century philosophy of science – Thomas Kuhn and Karl Popper – for why many in the philosophy of science community distrust the idea of established facts, and potentially for why an institute like the IASC hasn’t been founded before.

That’s why in the book at some point I say ‘I don’t want scientists to read this book

To simplify, Kuhn believed in scientific revolutions; there are “periods of stability in science”, during which science advances within a paradigm, “which are then punctuated by big scientific revolutions” or ‘paradigm-shifts’, which are accompanied by a new set of ‘weak’ facts. “For Kuhn, facts were relative to a paradigm, and we might have new facts in 50 or 100 years”.

Karl Popper on the other hand “disagreed with confirmation entirely. He was a falsificationist. He thought the scientific method was all about trying to prove things wrong”. His view of science didn’t leave much room – or any – for scientific facts.

Vickers also calls out the scientific anti-realists who are sceptical about science, often based on the historical record (“the idea that you know scientific ideas have always been overturned and they always will be”), calling it “sort of a simplistic symmetry argument”.

With his criteria, Vickers almost creates a reflection of this reasoning, arguing based on the historical record that claims meeting his criteria haven’t been overturned, therefore modern-day claims also meeting the criteria also won’t be overturned.

But unlike scientific anti-realism, this produces a practical approach that’s more in line with how we actually experience the world and act within it. “With something like smoking causes cancer, I want to say it would be ridiculous to say that in 50 years people will look back and say ‘look, people used to believe that smoking causes cancer, now we know differently’ or ‘now we have a different theory’”.

But what if this is all just a simulation? Vickers paraphrases the philosopher of science Carl Hoefer: “If you push the sceptic to the limit where they’re actually a Cartesian sceptic, then you’ve won.” Vickers distinguishes between reasonable scepticism and radical scepticism, and argues that we should ultimately ignore the radical sceptics, and the Cartesian sceptics who believe “this might all be a dream or there might be an evil demon inventing my thoughts”.

“The idea behind trying to find the 95% threshold is that usually that threshold is such that the remaining sceptics in the less than 5% are usually being unreasonable. That’s sort of the idea. I don’t think I say that explicitly, but I suppose that’s the idea, that, you know, once you reach a really solid international scientific consensus, then whoever is left not believing in it is probably being unreasonable.

“There will always be scientists like that, very professional scientists, affiliated with an academic department, with a PhD, who for whatever reason have a radical view about some theory that everyone else takes for granted. But they often think of themselves as the lone voice of reason or something, I think that’s something that, you know, effects scientists as well as, you know, the people who we sometimes see on the TV – maybe in America especially!”

Curiously, Vickers’s criteria implies that if you want to know if a statement is a fact you might not need to do science at all, you might instead just need to ask enough scientists. There’s an interesting contradiction in the power of Vickers’s method being in the secondary evidence of scientific consensus, while scientists themselves are busy looking at the primary evidence to make their decisions.

“It is [interesting], and it’s potentially delicate. That’s why in the book at some point I say ‘I don’t want scientists to read this book’, because I want scientists to keep looking at the first order evidence, and not think ‘oh, the best way to truth is to look at the second order evidence, I’ll do that’ because if they all did that the whole thing would collapse”.

Image: Peter Vickers

Leave a Reply

Your email address will not be published. Required fields are marked *


This site uses Akismet to reduce spam. Learn how your comment data is processed.