Sceptical covid-19 research and sceptical polar bear science: is there a difference?

This essay about medical researchers having trouble getting their papers published because the results don’t support the official pandemic narrative has disturbing parallels with my experience trying to inject some balance into the official polar bear conservation narrative.1 Especially poignant is the mention of models built on assumptions sold as ‘facts’ that fail once data (i.e. evidence) become available – which of course is the entire point of my latest book, The Polar Bear Catastrophe That Never Happened.

Read the commentary below, copied from (6 September 2020). Bold in original, link added to the story to which this is a response, and brief notes and links added as footnotes for parallels with polar bear conservation science.

Thanks for the ongoing sanity that is Lockdown Sceptics. I read the piece yesterday about how the scientific community is slowly starting to wake up to the fact that we have been significantly underestimating the level of immunity in the population (something that LS has been saying for months). I was really struck by these lines:

“Unfortunately, not all scientists are so timid with their views. Could it be the silence of too many sceptical scientists that has allowed more confident scientists like Neil Ferguson to become so influential?”

As sceptical scientist myself, this point hit home, but the reasons for the silence of the sceptical scientific voice are not just to do with lack of confidence.

Firstly, it is important for a scientific argument to have data. Without data you’re just expressing an opinion which, of course, can still carry weight depending on who is expressing it.3 However, there are real issues both with the data we have around COVID-19 and its reporting.

It is a well-known problem in science that the “negative results” are rarely published and so the literature is heavily weighted towards positive findings. 4 This can lead to a false perception of what is happening. So for sceptical scientists wanting to make arguments, the data may simply not be there as it was a “negative result”.

Scientists also tend to want to publish interesting findings. As a result, the COVID-19 literature tends to be biased towards the serious or rare cases as these are by definition “interesting”. 5

Here’s an example of the title and the first few lines of a case report in the New England Journal of Medicine from April, which illustrates this point:

Coagulopathy and Antiphospholipid Antibodies in Patients with COVID-19

“We describe a patient with Covid-19 and clinically significant coagulopathy, antiphospholipid antibodies, and multiple infarcts. He was one of three patients with these findings in an intensive care unit designated for patients with COVID-19….”

There is nothing wrong with this paper, it is a typical case report. However notice that the title gives no qualification of the fact that the patients are in the intensive care unit and as such are not representative of the vast number of patients with COVID-19. If you just read the title you could erroneously infer that ALL patients with COVID-19 have issues with their blood coagulating and their immune system going haywire. That’s the problem, a report of a rare finding, designed to alert clinicians in the ICU of potential complications, can feed confirmation bias in a lot of the media (and the public) that COVID-19 is the new plague that will kill you as soon as look at you.

Unfortunately you cannot publish the balancing paper:

Mild cough in Patients with COVID-19

“We describe a patient with Covid-19 and a mild dry cough that resolved itself in a few weeks…”

It is uninteresting. Although ironically it would be interesting (and probably publishable) if COVID-19 was actually causing all patients to have major complications!

Finally as you reported today in your article about Prof. Gupta, there is also further worrying bias in the COVID-19 literature with editors scared to publish “dangerous” ideas that could “impact our response to COVID-19”. Limiting publication of such finding in “lesser journals” (essential ones that aren’t so widely read), is an effective way of burying the findings as they may appear less “valuable” than a publication in Nature.6

This literature bias makes addressing the major issue facing the sceptical scientist even more daunting. This issue is that they need to overturn established orthodoxy around COVID-19 and our responses to it.

The advantage that modellers had at the start of the COVID-19 pandemic is that they did not much real world data because they could run their models built on assumptions.7 So it’s not surprising that the modellers got in first. It is only now that we have the actual data can we look at what the modelling predictions and point out how inaccurate these were and start to see where the assumptions were wrong.

The problem is that the models and modellers created and established “facts” and you require a lot more data to overcome an established “fact” than was needed to create that “fact” in the first place.8

This was compounded by the fact that we then implemented solutions with assumed efficacy (e.g. wearing face coverings, lockdowns) and the use of these solutions have now become more articles of faith rather than scientific hypotheses.9 So to overcome such solutions will require large amounts of evidence to achieve a shift amongst the scientific community, many of whom have been active advocates of these very solutions. Imagine what data you would actually need to persuade Nicola Sturgeon that mask wearing has no benefit or Matt Hancock that lockdown is not the answer? I’d wager it would be almost impossible and will be all the more impossible if don’t allow the publication of “dangerous data” in the first place.10

Finally I think it import to also understand that science is a professional industry and that most scientists work for businesses and institutions. Most of these businesses and institutions will have implemented COVID-19 based policies, supported by senior leadership who, even if they don’t believe in the policies, will need to be seen to be “doing the right thing”.11 Scientists working in these organisations will also have contractual obligations that will limit their ability to publish without permission or produce communication that could be deemed to be detrimental to their place of work.12

Imagine if you worked for one of the companies working on developing a vaccine and wanted to publish something saying that “vaccines are a waste of time and money because everyone will be basically immune through infection before they get to the clinic”? This effectively means that the vast majority of scientists are in environments that require a level of collective “self-censorship” and so, with a few exception, most of us have to bite our tongues or run the genuine risk of “blow back” on careers.13 We are not in the position of having a comfortable academic chair from which to cast our pearls of wisdom.14

Despite this, science is built on data and so ultimately I have to believe that we can get to a point where we stop treating COVID-19 as a special case and recognize it as just another disease to go alongside all the other risks we face in being alive. I am greatly encourage by the fact that we’re seeing journals like the BMJ publish “sceptical” opinion pieces as it shows that this shift may be starting to occur although today’s article about Prof. Gupta shows that we may have a lot further to go.15


  1. Of course, this analogy applies also to the experiences of many scientists sceptical of climate science narratives: I am not alone in this regard, nor am I alone in my experience of challenging the dominant narrative of polar bear conservation science. Mitch Taylor, Peter Ridd, Tim Ball, Judith Curry, Roger Pielke Jr. and a host of other scientists could write a similar list of parallels.
  2. cf. polar bears are thriving
  3. cf. Ian Stirling’s opinion carried significant weight early on
  4. cf. ‘negative results’ for polar bears is evidence of bears not starving due to reduced sea ice (or population increases), such as in the Beaufort Sea and Barents Sea
  5. cf. cannibalism blamed on climate change
  6. cf. or refuse to publish at all
  7. cf. the 2007 polar bear extinction model
  8. cf. the importance of summer sea ice to polar bears
  9. cf. Ian Stirling interview 2016
  10. cf. Six good years in a row for Western Hudson Bay polar bears
  11. cf. IUCN Polar Bear Specialist Group expelled Mitch Taylor
  12. cf. my expulsion from the University of Victoria
  13. cf. BioScience attack on my scientific credentials and integrity
  14. cf. Mitch Taylor on accountability in polar bear science
  15. cf. 2016 paper on status of Canadian polar bears

Comments are closed.