Epistemic prudence, Noakes, and the limits of authority

Wittgenstein said “Whereof one cannot speak, thereof one must be silent”, and that quote seems as good a place as any to kick off a post on appeals to authority, the death of expertise, and the boundaries of disciplines. As I argued in a 2012 column, agnosticism is often the most reasonable position on any issue that you’re not an expert in (with “agnosticism” here meaning the absence of conviction, not necessarily the absence of an opinion).

However, we like to have strong opinions – and with the rise of social media, we like to flaunt them. Furthermore, the widespread availability of information via the Internet has arguably “democratised” expertise in the sense that no matter what your point of view, you can find a community that shares it and provides you with evidence to buttress it – no matter how lunatic that point of view is.

Combine that with the politically correct notion of “respect for the opinions of others”, and we quickly end up drinking too deeply from the well of postmodernism, where truth takes a back seat to sensation. (For example, Sandy Hook: while you might think it crazy enough to suggest that it was the work of the US Government, some people are crazy enough to say that it never happened at all.)

As per the title of this post, two remedies to the proliferation of “opinion dressed (or interpreted) as fact” suggest themselves. These two remedies by no means exhaust the set of possibilities, but are simply the two that have been on my mind over the past few days.

First, epistemic prudence: This was the topic of the agnosticism post linked above, and also the subject of a post about misleading claims made by Prof. Tim Noakes and others about health and diet (among other posts), but I want to briefly add a couple of points emerging from a discussion I attended during the Franschhoek literary festival last weekend.

The discussion was between Prof. Noakes and Prof. Lionel Opie (who was, I believe, Noakes’s Ph.D. supervisor back in the day), and moderated by Judge Dennis Davis. You can listen to the whole thing on Soundcloud (embedded at the end of this post) if you like, but I’m going to highlight three moments that illustrate the key concern I’ve been expressing across all the Noakes posts I’ve made – namely, that of the poor example he sets regarding the difference between hypotheses and facts, and how to engage with evidence and arguments.

At 31:27, when asked where his evidence is (after a few minutes of dismissing anecdotes as meaningless, and that association doesn’t prove causation – nice to hear, but standards he doesn’t seem to apply to his own anecdata), he refers to a Volek et. al. book resulting from research commissioned by Dr. Atkins. That’s the Dr. Atkins who was recommending a quite-similar diet to Noakes, and who was operating a Foundation devoted to promoting that view.

This doesn’t mean that the research is necessarily poor, at all – but whenever I or others have pointed Noakes to contrary evidence, he’s never passed up an opportunity to dismiss the research on the grounds of perceived bias. If this is a good standard to apply to research (it’s not, at least not reliably), then an honest broker should apply it across the board, not simply to research with inconvenient findings.

53.16 – “If you’re insulin resistant, you do not have to get any disease whatsoever. If you eat a high-fat diet all your life, you will not develop diabetes, you will not get cancer, you will not get dementia. That I can guarantee you”.

Judge Davis (correctly) notes “that’s a very extravagant claim”, to which Noakes responds, “those diseases did not exist until we started eating highly-processed foods”. But it was in 1906 – long before we took the dietary mis-step that Noakes claims we did – that Alois Alzheimer first observed the characteristic neuropathology of the dementia (Alzheimer’s) that would end up bearing his name.

Noakes has certainly become adept at the just-so story and the business of salesmanship, but again – whether his conclusions are right or wrong – people are seeing a bad example of science at work, and a scientist (including Noakes) should ideally care about promoting good scientific practice, alongside promoting good health. Both of these goals can be pursued simultaneously.

At 41:37, Davis asks Noakes how he can trust that in 5 years time, Noakes won’t change his mind again (and therefore why he should believe him now), after Noakes’s about-turn on carbohydrates (he used to advocate carbo-loading for athletes, just in case you’re unaware).

(41.50)”Because I’ve said one thing, and now I’m saying the exact opposite. And they can’t both be wrong. And that’s key”, replies Noakes in a quite emphatic tone. The audience laughs, while Opie (in vain) protests “but you could be wrong now”. Opie is right, Noakes very wrong, and if the audience was laughing at the same thing I was cringing at, I’d be happy to hear it.

I was cringing inwardly at hearing such shocking logic from an A-rated Professor at my university, as Noakes seems unaware that he is expressing a straightforward false choice (or false dilemma, or false dichotomy) fallacy, that takes this form:

  1. Either claim X (carbs bad, fat super-good) is true or claim Y (carbs good, fat super-bad) is true (when X and Y could both be false).
  2. Claim Y (carbs good, fat super-bad) is false.
  3. Therefore claim X (carbs bad, fat super-good) is true.

You can hopefully see the problem, but to be clear: if there’s any intermediate position available to us between those two extremes (and there is, namely that fat is less bad than we thought at one one, and carbs less good, but that saturated fat is still bad, etc.), then a choice presented by Noakes is a false choice. The options he presents for consideration don’t exhaust the set of possible options, and he could end up being wrong on both occasions.

This might be either a demonstration of logical ignorance, or the cynical manipulation of an audience to make his case appear stronger than it is. Neither option is attractive, and neither option helps the cause of good scientific reasoning or communication, regardless of how thin you might be, or your quality of life. So, I hope that there is a third possibility here too.

Second, and in brief as I’ve gone on for longer than I expected to, the limits of authority: we can all develop resources for separating unjustified claims from justified ones, and for knowing what good arguments look like, in general.

Just because Alain de Botton is a bestselling author, it doesn’t follow that he always makes sense. Just because Dr. Oz is a renowned cardiologist, doesn’t mean he can’t sometimes act like a quack.

In general, let’s remember that no matter how right (or at least, qualified) you might be in one field, you might be just as susceptible as the rest of us to error in all others. We know that we’re bad at spotting our own blind-spots, and this makes it all the more crucial that we devote significant attention to remembering how easy it is to make mistakes, and trying to avoid encouraging others to make mistakes along with us.

Also published on Medium.

  • albie_cilliers

    A very well written piece, as always Jacques.

  • Michael Osborne

    Jacques, I can’t agree when you say that “agnosticism is often the most reasonable position on any issue that you’re not an expert in (with “agnosticism” here meaning the absence of conviction, not necessarily the absence of an opinion).” For all practical purposes, I am personally convinced, not agnostic, that evolution happened, that continental drift is occurring, that Henry VIII once reigned, and that mosquitoes carry malaria – despite the fact that I am no expert in any of the respective fields. When I use the word “convinced” in this sense I mean something stronger than merely having an opinion. That is not to say that these beliefs are utterly beyond question. Just that she who wishes to convince me otherwise bears a very heaven burden of persuasion.

    • As I say, ‘often the most reasonable position’. Sometimes, not.

  • Michael Osborne

    Jacques, I do note that you use the qualifier “often” in your second sentence” I would suggest amending that to read “may be in certain instances.”