Communicating Un/Certainty

[opinion] We aren’t supposed to claim that scientific findings were “The Truth”. But to say that every scientific statement was uncertain also is usually mostly wrong.

The blanket statement “everything in science is provisional” is used by doubt-spreaders to undermine scientific consensus in the public. It’s the basis for all denial of expert opinion. “Oh, these scientists don’t really know anything, they say so themselves. Hence, my fringe opinions are as valid as a scientific consensus that emerged from centuries-worth of scientific inquiries”. When a layperson heard this false argument before, they will be reminded of it, whenever you use such blanket statements about science being uncertain. This is where the mindsets of researchers and communicators differ immensely from the mindsets of everybody else.

And it also usually isn’t true considering the context. Let’s take Classical mechanics as an example. Physicists, high-tech engineers, and other researchers and developers would call classical mechanics only conditionally valid, unfinished, or incomplete. But in the real-life experience of virtually every other human being, the projections made using classical mechanics are more than just sufficiently reliable. So, if we report on things that are within the realm of classical mechanics, adding “well actually, classical mechanics will probably be replaced by a better theory at some point” isn’t helping.

Pointing out that the scientific consensus is preliminary when in the context of the subject matter it has proven to be absolutely valid, invites doubt. This may not be so tragic with the example of mechanics (unless someone tries to quantum-tunnel through walls). But with the big ecological catastrophes, this is clearly a problem.

Of course, you should not present the worst-case scenario in your climate model as our certain future. But, you should also make it clear that the best-case scenario will already be extremely bad, and also will almost certainly not happen. And, if you choose to talk about the worst-case-scenario, make clear it is because in your opinion it is prescient to prepare for the worst.

When you say “there is uncertainty”, it sounds like you are saying “everything is equally possible” to anyone who’d wish your inconvenient truth to be false – even those who actually do care are prone to this bias.

No sentence in your piece should make a lazy blanket statement about un/certainty.

A good place to start is probably to rigorously delete phrases like “everything in science is preliminary” or “science is always changing” and replace them with content. Explain the context and clearly communicate what is scientific consensus, where exactly are the uncertainties, and how big are they.

The danger with communicating the necessary context, of course, is to overwhelm the audience with details. We need the cognitive load to be manageable for our audience. Otherwise they can’t possibly follow your argument.

A lot of the craft of science communication is to understand and convey the current state of scientific inquiry. And a big part of it is to identify the relevant uncertainties and provide the context appropriately.

Science can provide certainty to a high degree in most everyday situations. And that is exactly how it should be communicated. Relevant uncertainties, however, can also be tolerated by the audience, if they are properly communicated.