JOHN RIDGWAY
The UK’s Government Office for Science really does seem to have this thing about the rise of misinformation (due to the so-called ‘infodemic’). Not long ago, they got together to discuss how it is supposed to be damaging the prospects for tackling climate change. Now they have commissioned the publication of a paper that claims to summarise everything we know about its wider implications and what we can do about it. The paper is heavily influenced by the works of Messrs Cook, Lewandowsky and van der Linden, to the extent that it amounts to little more than an academic panegyric, issued for the benefit of three egos that are already in full bloom and really don’t need any more watering.
Given the paper’s main expert influence, and how those experts emphasise the benefits of seeing everything through a behavioural scientist’s eyes, you can imagine that it has attracted quite some criticism. This is mainly because of its advocacy of Inoculation Theory and the use of ‘prebunking’ as the first line of defence when trying to deal with misinformation. I am tempted to add my own voice to these concerns but I think I have probably already made my thoughts clear enough regarding this most invidious of ideas. Besides which, when I started reading the paper I quickly came across a problem which I thought perhaps more worthy of my attention today – a problem that seems to be, if anything, more fundamental.
For me, the warning bells started to sound as early as the Executive Summary, in which one reads:
There are conceptual and policy challenges associated with how terms to describe false and misleading information are used and interpreted, but debates in the literature around these should not become a distraction from action to address harms.
Really? Are they actually saying it is more important to take action to address harms than it is to first establish how one may objectively define the basis for positing such harms? Are they really acknowledging a lack of agreed definition within the literature but then choosing to ignore the problem regardless?
Basically, the answer is yes, as becomes evident when reading section 4.1, ‘Definitions and Terminology’, in which it is stated:
Rather than providing detailed definitions of each term, this section focuses on the broader conceptual and policy challenges associated with how such terminology is used and interpreted. This reflects our consultation with experts, who emphasised that an excessive focus on definitional distinctions can detract from understanding how false and misleading information operates in practice, the impacts it has, and how it can be addressed at scale [My emphasis].
I am not at all surprised to find that the experts consulted were not that interested in pinning down definitions, considering how fluid and malleable the term ‘misinformation’ has become in their practical hands. In fact, a hint of this can be found in the paper’s own insistence in referring to ‘false and misleading information’ rather than ‘misinformation’. Gone are the days when misinformation was just false information, and hence something that was objectively verifiable. Now we can also have information that simply misleads, whatever that means.
The reality is that they can’t be comfortable with a firm definition of ‘misinformation’, because that would require a definitive basis for determining what is and is not misleading, as opposed to simply leading. And none of those who purport to be tackling an ‘infodemic’ (but are actually seeking to clamp down on stuff they don’t agree with) would want ‘misleading’ to be well-defined. Much better to just get on with the clamping and not worry too much about whether the right things are being clamped. No wonder the experts see the definition of one’s terms as an unnecessary distraction.
Troubled by this blasé attitude towards basic definitions, I searched amongst the cited papers to find a good example of misinformation that was not factually wrong but nevertheless misleading. A good example would at least reassure me that the experts know ‘misleading’ when they see it. This is what I found in a van der Linden paper:
So, as far as I can see, a headline draws attention to a correlation between two things and points out that the CDC has become involved to determine if there is a causal relationship. There’s nothing misleading about that. But this is what van der Linden goes on to say:
However, there was no evidence at the time that the doctor died because of the COVID-19 vaccine, yet the headline falsely implied causation where there was only correlation (van der Linden, 2022)
Huh?
Having got nowhere by consulting the experts, I turned to AI. I was then reliably informed that my work is considered factually correct but nevertheless misleading for three reasons (always three reasons; that’s AI for you):
- It undermines confidence in the institutions and pathologizes the consensus.
- It reframes climate mitigation policies as dangerous.
- It contributes to the ‘discourse of delay’, particularly by emphasizing risk and uncertainties.
Remember, misleading information is deemed by the Government Office for Science to be misinformation even when factually correct. Hence, it is something to be clamped down upon. The implication is that if you raise issues that would legitimately require one to lose confidence in an institution, then you will still be clamped. If one challenges a consensus, one must prepare for more clamping. One cannot draw attention to risks and uncertainties within the solution domain without being clamped. And one cannot advocate a sensibly robust approach to the management of transition risk without being accused of engaging in a ‘discourse of delay’, followed by more joys of the clamp. Such clamping is supposed to be a legitimate response to misinformation but I put it to you that branding my concerns in this way represents an abuse of language designed to protect a specific policy (in this case Net Zero) from democratic accountability and sceptical scrutiny.
I can’t be doing with such a highly politicized and tendentious approach to what should be an exercise in objectivity. This is an institutional abuse of terminology that casually flicks my concerns into the category of ‘misinformation’ or ‘delayism’. The only historically and logically legitimate definition of misinformation is the dissemination of verifiably false data, fabricated evidence, or objectively disproven factual claims. The rest is politicking and should not be given the thin scientific varnish that the likes of Cook, Lewandowsky and van der Linden seek to apply. The Government Office for Science and its supporting experts should either define their basic terms without equivocation, or make way for those who are prepared to do so. It’s bad enough that they should advocate for prebunking, but it is simply unforgivable that they also inadequately define their terminology, quite knowingly, in order to enable an entirely subjective basis for deciding what needs to be prebunked.
••••
••••
••••
••••
(UKR)