In the context of the continuing coronavirus pandemic and related crises – which are rapidly intensifying in much of the northern hemisphere as their second wave worsens – I want to reflect on last year and try to begin to identify some possible lesson learned. 2020 was a year we won’t soon forget and for me personally it was an absolute shocker of a year with lasting consequences. Processing, and attempting to reflect more deeply on, some of these events and my behaviour offers one way of beginning to move forward.
In this post I’ll start with the personal – my own behaviour – then briefly consider current events elsewhere in the northern hemisphere (I’m based in Melbourne, Australia) that relate to my own pandemic behaviour. I’ll then discuss some possible lessons.
Early on in the pandemic I started to wonder whether the form and scale of some of the public health measures being implemented were appropriate and may turn out to be ill-considered. I don’t recall exactly why I began to wonder this, or how skeptical I was at first, but I do recall listening closely to some preeminent clinical epidemiologists (e.g. Professor John Ioannidis) who asserted that we had both a poor understanding of the level of threat posed by COVID-19 and a poor understanding of whether the intended benefits of lockdowns (and other pandemic suppression measures) that were being implemented would exceed their potential harms and associated adverse second- and third-order effects. Ioannidis argued that blanket lockdowns constituted “medieval medicine” with unknown ‘side-effects’ and suggested that more targeted, evidence-based measures focussed on at-risk population segments should be preferred. I also recall noting the suggestion that the SARS-CoV-2 pandemic may be a repeat of the swine flu pandemic (of 2009-10) and others in which early threat assessments were greatly exaggerated. What if COVID-19 was a repeat of this? Additionally, other epidemiologists commented that non-pharmaceutical interventions (NPIs) generally referred to as ‘lockdowns’ – such as mandatory stay-at-home orders, and additional border closures, school closures, etc – may not improve aggregate health outcomes and could even turn out to be counter-productive. Overall, some public health experts suggested disruptive NPIs may just delay the eventual epidemic outcomes (including delaying the reaching of herd immunity) at great cost to societal functioning and the lives affected by those interventions. Furthermore, some forecast the potential for vicious cycles related to the ‘lockdown’ measures, such as, for example, the potential for recurring ‘lockdown cycles’ which have escalating costs with each new iteration. Related to this, uncertainties about vaccines loomed quite large back in March 2020.
Over recent years I had also become more immersed in what sociologist Harry Collins has called the “scientific form of life”. Collins notes that some of the constitutive features of this form-of-life include “debate and ‘organized skepticism'”, with a skeptical approach to claims and evidence drilled into members of this collectivity. It initially appeared to me that such norms were being inadequately followed in the context of COVID-19, and others also seemed puzzled by the lack of debate about related policies. (Later contrarians would also contend that core traditions of open inquiry, reasoned debate and agonistic discussion had broken down in the context of the pandemic, whilst others, in contrast, emphasised the need for unified public health messaging during a public health crisis [e.g. link]).
Other experts countered that we knew enough to act and that with novel public health threats like a pandemic you must act before you have perfect data and be willing to risk an ‘over-response’ (i.e., move fast and apologise later – if necessary). The basic thinking was that the Chinese government wouldn’t have locked-down the city of Wuhan for nothing, and this was sufficient signal to take this very seriously. I recall being alarmed by the events that were reported in the media but also wondering if the level of alarm was justified.
I also recall the tone and tenor of discussion on social media and in other media during much of 2020. My sense at the time – which I now recognise as largely reflecting my own biases – was that such discussions were often overly ‘panicked’ and had an uncritical bias towards sharing ‘bad news’ and related reporting which I believed may turn out to be sensationalised or somewhat unbalanced. The assessments of many credible experts placed far greater emphasis on uncertainty (e.g., knowledge gaps and limitations), and appeared to me to be more measured and skeptical than what I was seeing on social media.
During 2020 I often pushed back against what I saw on social media by sharing information that countered or questioned concerns and beliefs. I aimed to help alleviate anxieties where these may be exaggerated, whether these be related to estimates of the ‘infection fatality ratio’ (versus the ‘case fatality rate’), immune system responses, heterogeneous susceptibility to serious disease, and myriad other aspects. I pointed to the differences between the current pandemic and historical pandemics – for example, epidemiologists pointed out that in a pandemic young people tend to be disproportionately affected whereas in the current coronavirus pandemic in most countries the average age of deaths from COVID-19 is close to average life expectancy. (Others felt such messages lacked the necessary empathy or compassion for people affected by the pandemic – e.g. those who’d lost a loved one – a valid criticism I wish I’d given far greater consideration).
Important questions for my own reflection include the following: why did I feel compelled to address perceived biases and alleviate anxiety? I can partially rationalise it in terms of prior work experiences in which I became aware of how risk perceptions often poorly reflect actual hazards and in which I had learned about factors which promote these misjudgments such as novelty. My concern that panic often impairs rational thinking, good policy and decision-making is also genuine – I’ve often seen this play out. However, I have no public health expertise, nor am I a medical expert, and, thus, I have no expertise regarding the pandemic and its management. Arguably I should’ve ‘stayed in my lane’.
I wouldn’t go so far as to say that I felt I was a ‘voice of reason’ in a time of inflamed anxieties though some issues like those listed above (e.g. risk perceptions vs actual hazards) have this flavour to them. I need to reflect on such potential motivations. I certainly don’t see myself as more rational than others – quite the opposite, in fact. My own ongoing struggles with reason and rationality partly motivate my interest in these issues.
A further factor I need to reflect on is my concern about catastrophising impulses – what others have called the “extreme claims problem”. My preliminary conclusion is that my reactions to such potential impulses can be too reactionary and unthinking.
A further central question: why did I take alternative views departing from the emerging expert consensus so seriously? I found prominent ‘contrarian’ scholars like John Ioannidis (from Stanford University) and Carl Heneghan (from the Centre for Evidence-Based Medicine at Oxford University) to be compelling, along with other epidemiologists questioning the prevailing covid policy consensus.
Whatever the reasons (like those listed above), from a certain point – probably from around mid-April during the first lockdown in Melbourne, Australia – I began to think more critically about some pandemic policies and wondered whether ‘following the science’ necessarily entailed what we were told it did. For starters I knew from earlier STS training that policy rarely follows directly and unambiguously from the relevant science (or other forms of expert knowledge). Rather, knowledge utilisation processes can be highly complex and involve myriad judgements which are rarely made clear when politicians and others glibly claim that what they’re doing is ‘science-based’ or ‘evidence-based’. But more than anything I began wondering whether the unprecedented policy actions that were being undertaken were justified by our emerging epidemiological knowledge about COVID-19.
However, rather than question my questioning – and reflect on the fallibility of experts whose heterodox views often turned out to be wrong – I tended to gravitate towards the evidence and experts which supported such questioning. For example, I noted and emphasised papers by epidemiologists which argued that COVID-19 is far less deadly than we initially thought and raised questioned about related policies. I noted emerging research that may indicate pre-existing immunity had been underestimated (related to other circulating coronaviruses [see this recent article in Science magazine]). And I noted many other kinds of ‘good news’ which I felt weren’t getting adequate attention.
Though I stuck close to the scientific literature and steered clear of covid denialists – people who argue that COVID-19 is “just the flu” and/or who argue that it’s, overall, only about as deadly as influenza, or those who dismissively claim covid is a ‘scamdemic’ – it’s probably fair to say I had a partial focus on reviewing and sharing literature and ideas that reduced the feeling I’d been mistaken in my own doubts.
In hindsight I suspect that the thought that I might have been mistaken felt like a threat to my fragile self-concept (as an early career researcher whose ideals include rational thinking and critical thinking skills) and my already poor self-esteem which was worsening in the context of career issues (and other personal crises). Consistent with some psychological theory discussed below I think it’s fair to say that I became motivated to identify evidence that would provide ‘ammunition’ in social media debates and which may help to alleviate my own anxiety that I’d been mistaken in adopting a skeptical, questioning attitude.
The phenomenon of bolstering that I wrote about in my PhD thesis also seems relevant. In my thesis I cited the ideas of cognitive scientists (Hugo Mercier and Dan Sperber) who contend that “according to the argumentative theory [of human reason], reasoning should be even more biased once the reasoner has already stated her opinion, thereby increasing the pressure on her to justify it rather than moving away from it”. That is, such biased reasoning and related behaviours are primarily aimed at justifying a publicly stated opinion (and militates against critical reflection).
I now feel that much of the above behaviour constitutes a significant personal failure to act appropriately and live up to my own highest ideals. As deaths continued to pile up during the winter in the northern hemisphere as the second wave of the pandemic intensified, and as Australia enjoyed a largely ‘COVID-normal’ summer following successful aggressive suppression efforts, I began to reflect further my own behaviour during the pandemic. Though much of it was well-intentioned I don’t feel good about it.
Additionally, I’ve begun to think about my behaviour in relation to the unfolding situation in Europe and the USA and the approach of some experts. Why, for example, are Swedish medical doctors still asking questions like ‘Why did the world react so hysterically to covid?’ and asserting that the response of governments reflect “a state of collective mass hysteria”? Why are many actors and commentators still asserting that there is no evidence that lockdowns work (e.g. link, link) and some scholars continuing to assert that lockdown harms greatly exceed their benefits? Why do dissenters continue to claim that regressive social pressures are reinforcing adherence to the “covid consensus”? And, most centrally, why have many of the most public proponents of such views not modified their beliefs?
Related observations stand out. Firstly, something that’s striking is that many of the people who expressed such views about COVID-19 adopted their core positions early in the pandemic and, if anything, they often appear to be even more committed to such ideas. Such ideational lock-in seems impervious to conflicting evidence, criticism and personal reflection. Secondly, divisions between groups of actors seem to be becoming further entrenched between those who believe the “covid consensus” reflects “hysteria” (e.g. link) and others who defend, and advocate for, these pandemic suppression policies.
Such dynamics can be related to my own behaviour during the pandemic. A key issue, perhaps my primary sin in all of this, is rather than proactively question my own thinking and behaviour, during the more restrictive second lockdown in Melbourne (which lasted 112 days from early July to the 24th October) I became progressively more critical of what governments were doing in Australia and more interested in alternative policies. Though I felt it would be immoral to simply “let it rip” (as covid denialists may prefer), I did wonder if our improved understanding of the epidemiology of COVID-19 could justify alternative policies that are far less socially disruptive whilst protecting public health.
Additionally, as the second lockdown in Melbourne dragged on into its third and fourth months, rather than thinking things like “this will be worth it in the end” and “we’ve got no alternative – we just have to do this”, I began to ask different questions like “is there a better way?” and “is this really the wisest approach?” and more actively explored alternative policy ideas. I also found the sense of shifting goalposts and ambiguous policy objectives frustrating, as this shifted from ‘flattening the curve’ to ‘aggressive suppression’ to achieving elimination of the virus (though the latter objective was never explicit).
My reflections on my behaviour and the pandemic are ongoing. I think there are a number of sociology of knowledge-related lessons (see Collins, 2019). In addition, a few more specific themes and related potential lessons learned currently stand out:
1) Cognitive dissonance and self-justification: one aspect I’m interested in is the potential influence of a motivational mechanism termed cognitive dissonance which psychologists argue often “underlies the reluctance to admit mistakes or accept scientific findings”. Dissonance theory emphasises “the discomfort people feel when two cognitions, or a cognition and a behavior, contradict each other” (link). An illustrative example is a smoker who knows (or later learns) that smoking causes cancer – they are likely to experience intense cognitive dissonance if they continue to smoke. The theory further argues they will try to reduce this discomfort. Or, in the context of COVID-19, dissonance may be experienced by someone who publicly opposed a particular public health measure (e.g. strict lockdowns) and later is confronted with strong evidence of their benefits, particularly if they maintain their opposition. Similarly, the belief thatI’m a critical thinker whose beliefs are evidence-based and expert-informed would be dissonant with increasing awareness that their beliefs are inconsistent with much relevant evidence and/or the views of credible experts. Some social psychologists such as Carol Tavris and Elliot Aronson argue that related self-justification processes condition responses to cognitive dissonance.
Tavris and Aronson’s book Mistakes Were Made (But Not by Me) presents a few additional arguments which all appear highly relevant:
- Firstly, that cognitive dissonance (a so-called ‘negative drive state’) often leads to biases in evidence evaluation and/or production in order to reduce the discomfort that is being experienced (e.g. dismissing dissonant data). They emphasise the influence of related self-protective biases aimed at self-justification;
- Secondly, they argue that cognitive dissonance is a particularly strong motivational mechanism whenever a person’s self-concept is threatened – that is, they argue that cognitive dissonance is “most painful when evidence strikes at the heart of how we see ourselves” (link); and
- Third, they theorise the processes by which a person can gradually move from initial ambivalence towards dogmatic certainty via some initial decisions and/or small cognitive steps which unleash a subsequent need for self-justification and associated forms of rationalisation. They theorise these processes in terms of the ‘pyramid of choice’ (also see here) and ‘entrapment’.
Entrapment is further described (in the book Mistakes Were Made (But Not by Me)) as entailing “action, justification, further action” sequences which have the effect of increasing “our intensity and commitment” to a specific belief or choice. They suggest that related unconscious self-protective biases can be triggered which promote a move from initial ambivalence to dogmatic certainty. Tavris further contends that the initial decisions we make are extremely important such as where we jump to a particular conclusion. In the COVID-19 context this might be adopting the belief that public health policies are disproportionate to the actual hazards of COVID-19, or believing that the pandemic is likely to be a repeat of the swine flu pandemic experience. Tavris argues such actions/decisions, no matter how tentative, trigger unconscious mental processes and biases.
When I read about dissonance theory I reflected on whether (and how) such social psychological processes may have influenced my own behaviour during the pandemic and other behaviour I’ve observed. Related lessons include: 1) the need to be very careful when jumping to conclusions because this theory predicts that as soon as we make an initial decision we are likely to bias our consideration (and/or production) of evidence in conformance with this action or belief; and 2) the need to be cognisant of the potential influence of unconscious biases in our consideration of evidence. Tavris and Aronson argue that the theory of cognitive dissonance has “exploded the self-flattering idea that we humans, being Homo sapiens, process information logically”.
2) Frames and framing processes: a further theme (which is more abstract and complex) concerns how both interaction within and between groups along with the consideration of facts can be structured by frames – that is, intersubjective constructs (which therefore must be maintained in ongoing interactions) defined by sociologist William Gamson as “a central organizing idea for making sense of relevant events and suggesting what is at issue”. Similarly, Goffman invoked the notion of frames to theorise cultural constructs which govern the subjective meanings given to events and which give order and meaning to experience. For example, we might consider central organising ideas like ‘living with the virus’, or perhaps ‘COVID zero’ (which is oriented to eliminating viral transmission and fighting a collective ‘war against the virus’), as cultural frames or we could consider broader ideological frames (Sovocool et al 2016) which argue, for example, that the COVID-19 pandemic is ‘nature’s wake-up call’ and/or a ‘warning from nature’. Some sociologists argue that frames have broader effects like “orienting people to certain kinds of facts while leading them to ignore others” (link) and constraining what can be discussed and understood.
For example, we could consider the meanings given to the emergence of a new variant of the coronavirus in the UK which may be significantly more transmissible. If this variant is more contagious this will make it more difficult to suppress viral transmission. So, where ‘COVID zero’ is the primary frame we’d expect such aspects to get most attention. On the other hand, virologists have argued that the some of the mutations in this variant are likely to mean it is less virulent (i.e., less likely to cause severe disease). Reduced virulence would be a good thing if we need to ‘live with the virus’ (to some extent). The related transmission-virulence trade-off refers to how a pathogen that is too deadly cannot be spread as well. The related “theory of virulence recognizes that many germs will evolve less virulence as they circulate and adapt to the human population” (link). This example perhaps points to how cultural frames direct our attention towards certain facts (and away from others).
Returning to my behaviour and related possible lessons, I wonder to what extent my own thinking became influenced by interaction in groups in which ‘living with the virus’ is the central frame along with related agonistic interaction patterns that can give rise to polarisation. I failed to both consider this and reflect on how this may have focussed my attention on certain facts and led me to ignore or downplay other facts about COVID-19 I should have paid more attention to. We also need to be aware of the joint construction of meaning in the social groups we participate in. A further lesson is suggested by related sociological perspectives which argue that “facts have no intrinsic meaning”: we need to better consider what subjective meanings are being given to facts and the extent to which those meanings are culturally structured by dominant frame(s).
3) Group polarisation and attitudinal polarisation processes: this aspect concerns potential social dynamics shaping views and behaviour in the pandemic. Social psychological research has looked at how group discussion can result in members’ attitudes and opinions becoming more extreme in the direction already favoured by the group. Where this occurs it is referred to as ‘group polarisation’. For example, discussion amongst a group of people who are inclined to think lockdowns are a bad idea could result in members developing stronger convictions, potentially adopting a position more extreme than their prior privately held beliefs. Where such processes occur in different social groups with contrasting attitudes groups can move towards contrasting extremes. Some social scientists have argued that social media can plays influential roles in driving polarisation. Additionally, wider interaction patterns between groups could also play roles in attitudinal polarisation.
A key possible lesson here is the need to be aware that such social processes may influence our own convictions in ways we are unaware of. We may begin with weakly-held convictions or uncertainty but as we interact more within a group our attitudes may harden unexpectedly or become more ‘extreme’. Social media appears to have the potential to increase the influence of such social mechanisms due to echo chamber effects. Moreover, we need to consider how interaction patterns with like-minded people and with others holding different views could push us unexpectedly to extremes.
Additionally, reflecting on my own behaviour, I began with weakly-held convictions and uncertainties but social media provided a means of participating in group discussions which perhaps promoted far stronger convictions. Moreover, social media algorithms feed you content that can reinforce biases (as per echo chamber dynamics). Though my normal modus operandi is to consume a “balanced diet’ of content I think I failed to do this during much of the pandemic.
Beyond the above themes, I also intend to critically reflect on the extent to which I have given credence to contrarians in the context of an opposing expert consensus. Though I do believe contrarians can play important roles (e.g. where there may be premature convergence on an explanatory hypothesis or related belief), I also think I need to reflect of what might be leading me to uncritically give excessive credence to contrarians.
In addition to informing my own self-reflection, wider theoretical perspectives considered here – which can underpin cognitively-oriented social scientific analysis – may also offer ways of better understanding the public health policies and outcomes seen in different countries during the pandemic along with the related behaviour of key actors.