A Crisis of Expertise? Revisiting a panel from the CSPC 2021 after the trucker convoy

Research
Institute for Science Society and Policy

By Alana Lajoie O'Malley

PhD Candidate, Sociology (Science & Technology Studies), uOttawa

Alana Lajoie O'Malley
Tabaret
Our U.S. neighbours have been lamenting the dissolution of their shared sense of reality for years. The trucker convoys that overtook Ottawa and several other Canadian cities this winter appear to have ushered in similar anxieties on this side of the border.

With the onset of the “truckers convoy,” media commenters wrote about a crisis of expertise in Canada resulting in the disintegration of a shared sense of reality. Michael Den Tandt called the growing lack of a shared sense of reality in Canada “poison in the sinews of society.” Charlie Angus lamented the fact that people increasingly rely on their “own Facebook feeds, Reddit channels, and Slack chat information buttressing an utterly alternate reality of science, medicine and politics” instead of relying on “anchors of commonality” like national media and the experts reflected in it. Andrew Coyne worried that the convoy was  “a movement in opposition not merely to vaccines, but to science, authority, expertise of all kinds: in a word, knowledge.”

While these concerns may have taken on a new sense of urgency today, they are not new. Questions about how societies arrive at a shared understanding of truth and reality, about the role of experts in arriving at this shared sense of reality, and about how to take decisive policy action based on it, have preoccupied people who think about the role of expertise in political institutions for decades (123), if not centuries (1).

In fact, just a few short months before these concerns began circulating in the mainstream national conversation, the ISSP sponsored a panel at the annual Canadian Science Policy Conference (CSPC). Moderated by Dr. Kelly Bronson (UOttawa) and Dr. Gwendolyn Blue (UCalgary),  panelists Dr. Melanie Rock (UCalgary), Dr. David Guston (Arizona State), and Dr. C. Scott Findlay (UOttawa/Office o the Chief Science Advisor) shared perspectives on questions like “whose expertise counts?”, “what is expertise?”, and “what kinds of expertise are often left out of policy processes?” Revisiting these perspectives now offers a different angle into understanding this “crisis of expertise” in Canada; instead of seeing recent events as the product of a deficit within the Canadian public we might see them as an invitation to think carefully about how we as a society grant experts authority and, in so doing, construct our sense of reality, together.

During the panel discussion, no one disagreed with four main criteria, proposed by Dr. Findlay, that define what makes “experts”:

  1. they should have knowledge that is relevant to the policy area in question,
  2. this knowledge should be validated by some kind of external and legitimate community,
  3. they should be willing to have their knowledge challenged by others,
  4. they should explicitly disclose any competing interests.

This vision of expertise is similar to the one circulating in the media, and for good reason. After all, it does reflect, in general terms, how our governments and courts are currently designed to adjudicate expertise.

One of the problems the trucker convoy seems to have raised for Canadians, though, is that we can’t all agree on: what counts as relevant knowledge in every context; what counts as a legitimate community for external validation of that knowledge; who should have the right to challenge that knowledge and how they should be able to do it; or how to handle the fact that all experts have competing interests of some kind. Arguably, it is disagreements at this level that lead to the divergent versions of reality which are worrisome for so many.

Many of us have seen versions of a meme depicting someone “doing their own research” about COVID and commenting that based on this research they’ve discovered something that government public health experts missed. Many of us may even have had a good chuckle at the idea that “doing one’s own research” in this way is better than listening to the person with credentials and the entire machinery of a government resourcing data collecting, analysis, and dissemination. Far from being a sign of intelligent critical thought, we are meant to conclude that “doing one’s own research” (on the internet, via tools like Google and on platforms like Reddit, Twitter and Facebook) highlights lack of scientific knowledge and literacy as well as lack of capacity for truly critical thought.

But is it really so cut and dry? Many highly knowledgeable (and credentialled!) people turned to Twitter and other social media platforms to share robust data and analysis throughout the pandemic, especially at times when they felt that governments weren’t doing enough to protect the health of their citizens (123). We would be naïve to think that these rumblings on social media platforms didn’t have an impact on how governments pivoted their pandemic responses over the course of time. Would we be as ready to laugh at someone “doing their own research” if they were listening to these kinds of people on Twitter? Why or why not? What, specifically, is different between this role for social media in driving evidence-informed decision-making and the ways of interacting with these online platforms at which many are so eager to laugh?

If our answers to these questions are easy and simple, then we’re probably answering them poorly. As Dr. Guston put it, societies decide on who counts as an expert (and therefore who gets to help define collective reality) based on to whom they are willing to delegate authority. This can be as true online as in formal judicial or policy development proceedings. Normally, societies do this based on the presumption of a knowledge asymmetry: experts know more than others about the topic at hand, and societies delegate advice-giving and some forms of decision-making to them on this basis. This delegation is a signal of trust not only in the experts themselves, but also in the institutions into which they are meant to be included. This relational definition of expertise doesn’t at all preclude the four criteria for expertise we already mentioned; however, by putting the act of delegation front and centre, rather than the question of who should count as a legitimate expert, it draws attention to how the process of delegating to experts is not only social, it is also part of democratic process.

If there’s one thing we know about democracy, it is that it is messy and contentious. If delegating to experts is part of democratic struggle, making absolute pronouncements about who should and should not count as an expert across all settings and all time, who should and should not get to help define reality, is not too different from making absolute pronouncements about which politicians will be better at governing the country. While this kind of rhetoric may be par for the course during election campaigns, we all understand that at the end of the day the answer to this question comes down to matters of values, ethics, ideology, and taste.

This idea that delegating to experts is a democratic exercise, rather than an exercise based on clearly established ‘best practice’ criteria, might be unsettling to some. Doesn’t it, after all, open the floodgates for the kind of disinformation and social division we seem to be growing more and more concerned about? Indeed, this is a question that CSPC panelists and panel attendees didn’t necessarily agree on. For some, too much democratization muddies the waters and hinders good decision-making by bringing into the domain of experts those who possess neither the credentials nor the qualities to be there. For others, this democratization, even if inconvenient, is a simple fact of how recourse to experts unfolds in democracies today.

From this perspective, pining for a return to the “good old days” when more of us trusted the same experts to define reality for us is like trying to put the horse back into the stable. The alternative is to acknowledge that our institutions have failed to adapt and adjust to the fact that this approach to authorizing expertise no longer works (if it ever did 1, 2).

So, what would it look like to enter this “crisis of expertise” from a different angle? We can’t answer that fully here, but we can offer some questions to get things started. What if governments and their appointed experts were to step back from looking outwards and laying the blame for the current situation on fake news and on technically ignorant people “doing their own research”? What if instead they were to look inwards to consider the possibility that holding on to a vision of expertise that fails to be convincing to a large swath of society may be at least as big a problem as the disinformation that seems to be keeping so many up at night? Governments and institutions of knowledge-making have greater ability to change themselves than to change others, and a little humble self-reflection rarely goes amiss.