A new pro-lockdown study has been doing the rounds on social media. In a Twitter thread, one of the authors claims that it “confirms the tragic consequences of delaying the UK’s first lockdown”. He argues that, if lockdown had started just one week earlier, there would have been up to “35k fewer deaths”.
Although the thread went viral (as many pro-lockdown threads do), the study was not without its critics. One of these was Philippe Lemoine, whose work I’ve discussed several times here on the Daily Sceptic.
In a Twitter thread of his own, Lemoine retorted that the study “doesn’t confirm jackshit” and merely exemplifies the “ridiculous methods that pass as counterfactual analysis in the field of epidemiology”. He went on to say that drawing strong conclusions about the “tragic consequences” of delaying lockdown is “intellectually dishonest”.
Profanity aside, the criticisms Lemoine proceeds to outline are well taken. As he points out, the latest pro-lockdown study is based – yet again – on the assumption that epidemics keep growing exponentially unless the government decides to do something. This assumption is not merely questionable, but false.
We know from examples like South Dakota – whose libertarian governor Kristi Noem did basically nothing – that infections start falling long before the herd immunity threshold is reached, even if there’s no lockdown. (There are at least eight other places where infections fell from a peak in the absence of both business closures and stay-at-home orders.)
Armed with the assumption that the only thing capable of arresting epidemic growth is lockdown, the authors conclude that Britain’s first lockdown had a large effect – one that would have been even larger if it had been imposed a week earlier.
Of course, there’s ample evidence to suggest this isn’t true: infections peaked around the same time in no-lockdown Sweden; reconstructions of Britain’s epidemic curve show cases peaking before the first lockdown; and Chris Whitty himself told MPs that “R went below one well before, or to some extent before, March 23”.
So, another pro-lockdown modelling study based on assumptions that we know are wrong. (Note: I’m not saying the lockdown had absolutely no effect; just that you can’t claim it had a large effect.) However, the story doesn’t end there.
The author of the original Twitter thread didn’t take kindly to Lemoine’s criticisms. After demanding to know “who specifically” Lemoine was accusing of intellectual dishonesty, he asked him to remove the “libellous” tweet and “desist from further public defamation”.
While Lemoine (a Frenchman) could have perhaps been politer, resorting to accusations of “libel” when faced with criticism isn’t a ‘good look’ for a scientist. It suggests you’re more concerned with social status than with finding out the truth. Why not just ignore the Twitter digs, and answer the man’s criticisms?
While this little dispute hardly matters, it doesn’t show ‘The Science’ of lockdown in a very favourable light.