Expertise as Attribution – Towards a Solution to the Post-Fact World

The following is the text of a talk I gave at a symposium entitled The Politics of Expertise in Media and Cultural Research at Birmingham City University on the 30th November 2016. The references I used in constructing this text are available on request – reach me on Twitter @timsquirrell.

Expertise as Attribution

Tim Squirrell

PhD Candidate in Department of Science, Technology and Innovation Studies

University of Edinburgh

“I think people in this country have had enough of experts” – Michael Gove MP

“Experts, soothsayers, astrologers, are all in much the same category” – Jacob Rees Mogg MP

“Well, you can prove anything with facts, can’t you?” – Stewart Lee

A truly tremendous quantity of ink has been spilled over the past year over the problem of expertise. “Post-truth” is the Oxford Dictionary’s Word of the Year. The consensus seems to be that publics in Western liberal democracies have lost their trust in experts and expertise. The questions of when, why, and how this happened, and how the expert class can possibly hope to redeem itself are perpetually mooted in hand-wringing think-pieces in The Guardian and The Spectator alike.

In this talk, I’m going to do three things. First, I’ll problematise the existing, hidden consensus that experts just exist, and that it is our choice whether to listen to them or not. Second, I’ll propose an alternative view that conceptualises expertise as something which doesn’t exist a priori, but is attributed by particular audiences to particular individuals, groups and institutions in order to solve particular problems. Third, I’ll show why, far from being a simple redefinition which makes no practical difference, there are some specific advantages of taking this viewpoint, centring around (i) the way we ask those who would claim expert status to present themselves, and (ii) our ability to dismiss specific actors, rather than the concept of expertise as a whole.

The question that nobody seems to be asking is “What do we mean by ‘experts’?”. It’s taken as a given that when we say the word ‘expert’, we know of whom we speak: late middle-aged white men in suits and glasses, staring out at us from a television screen, speaking on behalf of a university, or an organisation with some acronym nobody can remember. IFS, IEA, ECB, CPS: they might as well all work for the same company, for all the scrutiny we give their credentials. We can identify who counts as an expert a priori: they speak for established institutions, they have qualifications and credentials and letters after their names, and they have experience to back up their authoritative-sounding words.

There are a couple of major issues with this. First, it allows individuals (and whole communities) to dismiss the entirety of what has come to be known as the “expert class”, without having to engage with their statements or arguments. When these experts make predictions about the economy, or elections, or the climate, they inherently stake (to some degree) their reputation on the veracity of their predictions. The problem is, if we link all experts together, when someone (or a group of people – say, psephologists) gets something wrong, then they harm not just their own reputation but the reputation of everyone linked to them through the label ‘expert’. When we bind people together with a particular label, we allow other people to use that label to have blanket beliefs about that whole group (experts) rather than making decisions about smaller, more appropriate, sub-divisions (psephologists, or Nate Silver, or the Huffington Post pollsters).

Second, it facilitates the complacency of the aforementioned expert class. If they speak, and nobody listens, they can throw up their hands and say, “Well, we gave them the facts and they refused to accept them. We can’t help it if the public are stupid and mistrustful.” There’s nothing they could have done to foresee this, and now that the epistemic gates are open and the horse of trust has bolted, there’s very little they can do to steer that horse back into the stable and regain the confidence of the public. Clearly, the solution is simply to end democracy and delegate all authority to the expert class, because the people can’t be trusted to make the right decisions when presented with the objective facts.

These problems alone should probably give us serious pause: is this view productive? Is it something we want to keep with us in this brave new post-fact world? When combined with some of the more conceptual issues I’m going to outline as we go on, I’m convinced we should wholeheartedly reject the view of expertise which says “these are the experts, believe them or not”.

“But what, pray, is the alternative?” I hear you cry, “We can’t allow just anyone to call themselves an expert. That way charlatanism lies!” Well, little strawman that I just made up, you’re halfway there. The alternative is as follows.

We could, if we so chose, decide to carve up the word conceptually and acquire knowledge purely through the mediation of our own senses. But our senses are unreliable, and there’s not enough time or money to allow us to imbibe the knowledge of every discipline under the sun. So for most of us, most of the time, we delegate our epistemic authority to others: we allow them to tell us what’s true, and we decide how to act upon it. As children we listen to parents and teachers; as adults we read news publications and books, listen to particular individuals we find compelling or inspiring, heed the advice of our doctors regarding our health, and broadly take for granted that we aren’t being lied to or misled.

There are certain ways in which we try to tell the difference between those experts we ought to trust and those who are incompetent, misleading or just plain wrong. We can examine how they argue: how well do they present their arguments, are they quick to produce defeaters or counter-defeaters for the arguments of others? We can look at their track record of making good predictions, we can try to examine any potential biases or vested interests they might have, we can scrutinise their credentials, or we can look at how many other putative experts agree with them.

There are numerous problems with each of these metrics which render it very difficult to tell between ‘good’ experts and ‘bad’ experts. Often those who know the most aren’t necessarily the best at arguing their side of a debate (a problem I will attempt to solve with my view later on). Moreover, it is almost impossible for us to assess the claims of experts on a technical basis, because the very reason we are listening to them is that the knowledge they are articulating is esoteric and epistemically inaccessible. Similarly, what makes a “good prediction” is up for debate, so track records are difficult to assess. Most individuals in a given industry or field likely have some biasing factors behind the things they are saying, because nobody is objective and anyone who says otherwise is lying or deluded. Finally, if we’re relying on the agreement of other potential experts, then we just defer the problem of identifying “good experts” one step down the line.

Most of our actions, then, rely upon trust in others. When we decide to believe what someone says, we don’t usually do so based on pure logical reasoning. Instead, we listen to them based on a function of whether we trust them (qua friend, or parent, or expert, or politician) and whether what they are saying sounds intuitively plausible. Crucially, intuitive plausibility itself is contingent on our prior beliefs. If what someone is telling us conflicts with a deeply-held belief – they’re telling me the Earth is flat when I’ve been informed my entire life that it is round, for instance – then it’s unlikely I’m going to abandon my prior belief in favour of what they’re now telling me. That’s important, because it provides further fuel to the pyre of the realist view of expertise: if I have a prior disposition not to believe people we call ‘experts’, because I perceive them to have been mistaken before, then it’s unlikely that their telling me something is going to have a large positive impact upon my propensity to believe it.

Delegating our epistemic authority – our ability to carve up the world conceptually – is incredibly common. When we give that authority to a particular person, group, or institution, and we do so for the purpose of solving a particular problem or class of problems, I call those actors ‘experts’. We delegate our authority on matters astrophysical to astrophysicists; we listen to oncologists about cancer; we heed the words of the weather forecaster on meteorological matters. They are our experts on those things. Crucially, this means that they do not already have expert status, putting the burden upon individuals and communities to decide whether or not to validate that status. Rather, we grant them that status when they are able to provide information that is useful for the resolution of particular kinds of problems. Expertise doesn’t exist independent of an audience to grant the status of expert.

What are the implications of this? The little straw-man from earlier might say that now anyone can be an expert, and that this will only speed our civilisation’s inevitable decline into chaos. On the contrary, tiny straw-man. The realist view means that those we designate as experts are never required to learn how to communicate effectively: they’re told that all they have to do is say the facts, and the public will listen – and if they don’t, more fool them. When we switch to an attributionalist view, we are able to place the onus upon those who wish to be considered experts to step up their dialectical game. It’s no longer enough just to sit back on a throne made of credentials and qualifications. Charlatans, hustlers and liars will always be able to peddle nonsense smoothly. We have to trust that those who know what they are talking about will be able to engage with them and show why they are incorrect. We have to trust that, when presented with equal rhetorical skill, truth will be vivified by its collision with error.

Further, and I think potentially even more importantly, when we refuse to engage in a system which identifies and protects a particular “expert class”, we afford ourselves the ability to avoid future situations like the one in which we find ourselves today. By recognising that expertise is a status rather than a trait, we head off at the pass any attempts to tar all ‘experts’ with the same brush. Instead of fetishizing credentials as the sole means by which people can enter the expert class, we should allow anyone to call themselves an expert, and then to have that claim tested through argument. Instead of dismissing an entirely disparate group of people who happen to have been lumped together we enable ourselves to dismiss individuals who make bad judgements, or institutions which have a reputation for making bad calls.

This is incredibly freeing. An economist makes a bad forecast? Be hesitant about trusting them in future. All economists fail to predict and prevent some financial catastrophe? Be incredibly wary of them, unless particular individuals or institutions show themselves to be worthy of a second chance. The failures of pollsters shouldn’t be used to render untrustworthy the predictions of political pundits; likewise, the success of one technocratic elite shouldn’t be seen as validation for other, unrelated people who happen to have a few degrees under their belts.

Expertise as a concept has to endure, and for this it has to be flexible enough to allow audiences to attribute it where they see fit. If they choose some charlatan with the “best words” and clever one-liners, we have to be prepared to fight them (discursively, of course) with equal levels of rhetorical polish. But crucially, those who actually do have knowledge, or skill, or predictive or explanatory powers, should be able to win out against those who do not, provided that they too are good at arguing and presenting their ideas clearly and simply to those who are not familiar with their fields.

If we do all of this, then maybe those who know what they are talking about have a chance of coming back from this anti-intellectual, anti-expert moment we find ourselves in. But if we don’t, then we’re doomed to repeat the technocratic mistakes of the past. We shouldn’t ever again have to hear that “people in this country have had enough of experts”. I hope that we don’t.

An Introduction to the Strong Programme in the Sociology of Scientific Knowledge

I mentioned in a recent blog post that the work I do takes place in the theoretical framework of the Strong Programme in the Sociology of Scientific Knowledge, also known as the Edinburgh School.

“Great,” you might have been left thinking, “but what the shit does that mean?”

Excellent question. Like 99.99% of the planet, I had little to no idea about the Strong Programme before I came to Edinburgh this year. My sum total knowledge of it came from one 4-lecture series about the Sociology of Scientific Knowledge, delivered by Simon Schaffer (who is by all accounts a fantastic lecturer), who told us that many people in the department (History and Philosophy of Science, at Cambridge) didn’t think he should be teaching us this. It was an excellent series, but the material was so new and so different to any of the core material we’d studied in Philosophy of Science that taking it on as an object of serious study at a time when exams were looming would have been grade suicide. I think the sole reference to it after that came in a supervision I had about expertise, where my supervisor told me about Martin Kusch’s Knowledge by Agreement, which promotes ‘communitarian epistemology’, the idea that knowledge exists only in groups. At the time I thought it sounded like nonsense. It’s certainly a deeply unintuitive concept for anyone to accept, let alone someone who’d spent the last two years immersed in the standard philosophical literature on knowledge – ‘justified true belief’, ‘the Gettier paper’, etc.

But since coming to Edinburgh and throwing myself at some of the literature (not to mention being taught by people who’ve studied and like this stuff), I’ve fallen in love. It’s a rare thing to plough through highly conceptual literature and not end up feeling mystified. It’s rarer still to have it make total sense in your mind, to have it change the way you view the world, the people in it, and relations between us. For me, the Strong Programme has done that. It’s pretty uncommon to find people working within this framework: it’s often dismissed because it’s a relativist theory, or because it’s social constructivism. What’s strange, though, is that there’s rarely any actual weight behind these criticisms. They’re used as heuristics to dismiss it out of hand without ever engaging with the actual substance of what the theory claims. Alternatively, people overlook the SP in favour of Actor-Network Theory, either because ANT seems more readily applicable to other fields, because it already has more work already done using it, or because Bruno Latour is a fabulous self-publicist (he thinks he’s Foucault. He’s not Foucault.).

I want more people to love the Strong Programme as much as I love it. I think it’s an absolute travesty that more people aren’t aware of its existence. Not only could people working in STS benefit from it, but I think it has much broader applicability in other disciplines. The social theory which underpins SP, the Performative Theory of Social Institutions, is extremely flexible and makes a great deal of sense. The underlying ontology, finitism, is not just a useful tool for understanding how knowledge is created in disseminated – it is also, to my mind, correct. I recognise that saying so may undermine my credibility not only with people who despise relativism but also with those who support it wholeheartedly, but frankly it’s a hit I’m prepared to take. I really like the Strong Programme.

So let’s get stuck in.

The Strong Programme grew through the ’80s and ’90s as a reaction to ‘weak’ sociologies of knowledge. Previous attempts at understanding knowledge through a social lens restricted themselves to understanding failed knowledge claims: they were a sociology of error, rather than of knowledge. This meant that phrenology, homeopathy and spontaneous generation theory would all be suitable candidates for sociological analysis; but relativity, evolution by natural selection and the Big Bang theory would not be. Belief in the latter could only be understood as a ‘rational’ response to the evidence of our senses.

There are a number of reasons this claim is total bullshit. Here are two. First, there’s no reason to believe that our current theories are correct. Every single theory we’ve subscribed to, ever, has over time been shown to be incomplete or flawed in some way, and there is precisely no reason to believe that our current theories are going to be the ones to buck this trend. As such, current successful theories are just failed theories waiting to happen, and should be susceptible to sociological analysis on these grounds. Second, it’s just untrue that we believe in, say, evolution because it’s the rational thing to do. I believe in evolution because a large number of people more intelligent than I have spent their entire lives studying evolutionary biology (so that I don’t have to), and have managed to entrench this belief in all of the institutions of our society, so that I was just taught that evolution is the best theory we have to understand how life came to exist in the form we see it today. It would be preposterously arrogant of me to say that the only reason I believe in evolution is because of the evidence of my own senses: I have literally never witnessed evolution take place, and if I had witnessed something which seems to validate the theory of evolution by natural selection, there are a thousand and one other theories which could explain that phenomenon equally well. There is nothing rational which makes me, or you, or anyone, believe in a particular way of understanding the world.

What’s really bizarre is that the same people who are rabidly pro-science (or at least, pro the idea of science as a transhistorical arbiter of objective truth) are also really, really bad at understanding epistemology. Faced with claims from the Bible or another holy text that god exists, they will posit the existence of a Flying Spaghetti Monster, and say that there is just as much evidence for its existence, and so it is just as rational to believe in it. I don’t care to weigh in on that particular shitshow of a debate. What I do want to say is that if you formualte that kind of argument, you should also naturally be a big fan of the Strong Programme, and epistemic relativism more generally. Why? Because you recognise that any set of evidence can be used to push you towards a potentially limitless number of conclusions. If you look at the world around us and decide that there is a god, the form that god will take is underdetermined by the evidence at hand. There could be any number of potential gods or deity-like entities which could explain the way the world is, and belief in any particular one is likely to be best explained by social factors – where you were born, what religion your parents were, what kind of school you went to – rather than by reference to the natural world itself.

You should probably admit that you believe in most scientific theories because someone told you to, rather than because you observed the evidence for them first-hand. That means that there’s a place for sociological analysis in understanding how we, as people, come to believe in some theories and not others based on the say-so of particular individuals.

But it goes deeper than that. Your response to all the above might be, “so what? I know that the scientists who work in gravitational wave physics have good reasons to believe in gravitational waves, based on the evidence before them, and that’s why I believe them over some over schmuck”. Fair enough. But there are a couple of things you might want to consider. First, how do you know they have good reasons to believe in gravitational waves? Sure, you’ve read a really interesting Guardian article about them which explains them super well, but people get tricked into believing convincing-sounding explanations all the time, like when I was 7 and Daniel from down the street told me that he would look after my Pokémon cards for me and then he never gave them back and told me that I’d never given them to him. Fucking Daniel. He can get in the sea.

The gravitational wave scientists tell us they observed the same phenomenon which is supposed to suggest the existence of the waves at two different stations, though. Surely that’s enough for them to believe in the waves based on pure rationality, right? Well, not really. First, we have to explain why each team would believe the other team. Then we have to explain why particular runs of an experiment get to count as ‘good’ runs – what about all the times when they didn’t detect gravitational waves? Then we need to account for why this time is the big one, when various members of the gravitational wave community have been claiming to have detected the waves for the last forty years. What made all the scientists in the community agree with each other that they had, indeed, detected a gravitational wave?

The point I’m making is that it’s not as simple as ‘scientists observe material world, get evidence, evidence leads rationally to theory’, and then ‘scientists tell us correct theory, we believe them’. It’s far, far more complicated than that, and we do ourselves a disservice by refusing to acknowledge that.

There’s an awful lot more to say about the Strong Programme, but this post has just edged past the point where people are likely to stop paying attention, so I’ll leave it here for now and resume in another screed.

Freakonomics and Expertise: What’s Missing?

This week’s episode of Freakonomics Radio is about the topic of expertise.1 As someone working towards a PhD studying expertise, I listened with interest: it’s reasonably rare to hear an in-depth discussion about this field, and trying to comprehend other people’s takes upon the subject and reconcile them with the work I’ve been doing. The academic niche that I work in (Science and Technology Studies, specifically the Strong Programme in the Sociology of Scientific Knowledge, which is heavily social constructivist in its approach) draws on a slightly different set of approaches to the topic of expertise to those cited in Freakonomics, which come more from the Psychology side of the spectrum of social sciences. As a result, I have a few thoughts about the framework of expertise articulated in the programme. Hopefully some of them might be considered constructive.

The academics consulted in this episode, chief amongst them Anders Ericsson, Professor of Psychology at Florida State University, advocate for the idea that expertise can be achieved through ‘deliberate practice’.2 This effectively cashes out as activity which constantly pushes its practitioner out of their comfort zone, often focussing upon ameliorating specific problems or stumbling blocks which they may currently be facing. They note that peak human performance in many areas has improved dramatically over recent centuries: the record time for the fastest marathon has decreased by nearly an hour since the first modern Olympics in 1896;3 Mozart’s ability to perform music at various stages of childhood would likely be considered ‘average’ amongst children at a musical academy today; and so on. The idea is that we stand on the pedagogic shoulders of giants: over time, we have learned the ability to learn better.

The ‘10,000 hour rule’ popularised by Malcolm Gladwell in his book Outliers also makes an appearance,4 with some discussion over whether this amount of practice is indeed the ‘magic number’ for becoming an expert in something, whether it is how one spends those hours that matters more than the sheer volume, and if the proper formulation might in fact be ‘10,000 hours + basic talent’.

I think the formulation of expertise in this programme might benefit from a couple of interlinked observations.

First, I would posit that what they are describing is not the process of acquisition of expertise; rather, they are describing what it means to become adept at a skill. Much of the discussion is focussed around musical or sporting ability, with a cursory mention of writing at one point during the episode.

It is unclear to me that these are forms of expertise. When we talk of experts in society, we tend to talk about them as people who are unusually knowledgeable or skilled in something, but that isn’t a sufficient condition: there’s something extra. Usually that extra thing is the way we relate to them as experts. We ask them, as representatives of their field (whatever the nature of that field), to solve particular problems for us.5 We ask physicists to tell us about gravitational waves. We ask psychologists to help us understand how we learn. We ask cardiologists to tell us how we can minimise our chances of suffering a heart attack in the future. The crucial aspect of expertise which is missing is its sociality: without it, there is no expertise, there is only skill.

You might ask, so what? This is just linguistic nit-picking. It doesn’t matter if we call it skill or expertise, we just want to know how people get really good at things. I should just get back in my ivory tower and complain to the approximately half a dozen people in the world who care. Fair enough. But I think there’s a deeper insight to be gained from the distinction here. A skill can often be learned in some kind of isolation: I can get really good at playing scales on my piano just sitting in my bedroom with some sheet music.

But there’s something missing. We know that people tend to learn much better when they are taught by others. The best tennis players have other extremely skilled players coach them; the same goes for pianists or even academics. This isn’t just because those people know more stuff. It’s because they know how to apply that stuff. What’s missing from the deliberate practice model is the recognition of the power of tacit knowledge: the things we can’t articulate, but which can only be learned from being immersed in the community which surrounds our interest.6 You can practice your scales as long as you like, but you’re never going to understand what it means to give an emotional performance which makes a crowd love you if you don’t mix with (and learn from) people who know how to do just that.

In fact, it’s impossible to even know what constitutes an emotional or moving performance without socialisation. Why? Because our standards for what is good or bad, overdramatic or underplayed, technically accomplished or pretentious nonsense, all vary between times and communities. The extent of this variation is different in different fields, but it always exists. There was a time when Isaac Newton would have been recruited as a virtuoso physicist; if his reanimated corpse were to be dug up today, he would no longer be considered an expert, because he has spent the last three hundred years not immersed within the culture of physics and mathematics. Zombie Isaac Newton could probably, with time and adequate socialisation, be a great contributor to modern physics. But critically, he would be totally incapable of doing so without becoming part of the physics community: he would not only need to read modern textbooks and academic papers to know what physics consists in nowadays, he would also need to know which journals and authors to take seriously and which to ignore, as well as how to converse in the language of modern day physics. He wouldn’t have a snowball’s chance in hell of becoming a virtuoso without extensive social contact with other physicists.

The missing ingredient in this otherwise highly interesting treatment of the problem of expertise acquisition is the social. Malcolm Gladwell recognises in the podcast that ‘you can’t do [10,000 hours] by yourself’, but what he means is that 10,000 hours is a hell of a long time and you’re likely to need people to help you with perform your basic needs whilst you’re playing fifty games of chess every day. Even if that weren’t true, it would still be the case that nobody can become an expert, or adept, or a virtuoso, on their own. Social immersion and tacit knowledge is at the very core of what it means to be truly great at something, and to be recognised as such.

If you’re interested in STS or the Strong Programme and its approaches to expertise and knowledge, there are a few books and papers I’d highly recommend:

Barnes, Barry, The Nature of Power (Urbana: University of Illinois Press, 1988)
Barnes, Barry, David Bloor, and John Henry, Scientific Knowledge: A Sociological Analysis (The Athlone Press, 1996)
Bloor, David, ‘Idealism and the Sociology of Knowledge’, Social Studies of Science, 26 (1996), 839–56
Collins, H. M., and Robert Evans, Rethinking Expertise (University of Chicago Press, 2007)


1Greg Rosalsky, ‘How to Become Great at Just About Anything’, Freakonomics, 2016 <http://freakonomics.com/podcast/peak/&gt; [accessed 28 April 2016].

2K. Anders Ericsson, Ralf T. Krampe and Clemens Tesch-Römer, ‘The Role of Deliberate Practice in the Acquisition of Expert Performance.’, Psychological Review, 100.3 (1993), 363.

3Wikipedia, ‘Marathon World Record Progression’, Wikipedia, the Free Encyclopedia, 2016 <https://en.wikipedia.org/w/index.php?title=Marathon_world_record_progression&oldid=716163592&gt; [accessed 28 April 2016].

4Malcolm Gladwell, Outliers: The Story of Success (London; New York: Penguin, 2009).

5Zoltan P. Majdik and William M. Keith, ‘Expertise as Argument: Authority, Democracy, and Problem-Solving’, Argumentation, 25.3 (2011), 371–84 <http://dx.doi.org/10.1007/s10503-011-9221-z&gt;.

6H. M. Collins and Robert Evans, Rethinking Expertise (University of Chicago Press, 2007).