The limits of expertise

I’m very much in favour of listening to experts. If I want to understand something, I want to know what the best informed people have to say on the subject. There are, however, a number of issues around how we collectively handle the idea of experts and sometimes the experts themselves are complicit in these problems.

We’ve seen it a lot around covid – experts in another area who are non-experts for covid being invited to share their opinions. It can be interesting to get insights from people in related areas, but it can also be incredibly problematic. 

I remember many years ago there was an issue in the UK with an expert witness who didn’t think cot death would happen more than once in the same family. He said once was a tragedy, twice suspicious, three would be murder. Only it turns out that isn’t true. Roy Meadow was certainly an expert on child health, but he was operating outside his own expertise with assertions about statistics that led to unjust imprisonment. He was eventually struck off.

Journalists are increasingly unreliable around working out who can give us informed, and informative insight. This is an example of people we should be able to trust as experts failing to do their job. Ideally, what the expert journalist does is identifies the people who are qualified to inform us about things. Instead we’re seeing far too many people being invited to share their opinion because they are famous, not because they know anything. It distorts debate, and is profoundly unhelpful.

We need people who really are experts in some things to be quick to say when they’re being asked to comment on things they don’t personally have a handle on. 

We need to be clear about what expertise a person is bringing to the table and how that relates to the topic in hand.

We should be wary about the opinions of people who are discredited in their field. Yes, sometimes it’s the lone voice that turns out to be right all along – but not often. Being considered useless by your peers is not proof that you’re right. Especially in science, where peer review and being able to test each other’s results is rightfully key. 

We need to be alert to the difference between opinions and facts. An informed opinion is actually worth more than the opinion of someone who has no idea what they’re talking about. Speculation from people who know what they’re talking about is more likely to be relevant than speculation from people who don’t have that basis to work from.

That experts can get things wrong does not invalidate the idea of expertise. It’s also the case that if an apparent expert is wrong, the person most likely to spot that will also be an expert. The problems with Roy Meadows were identified by The Royal Statistical Society, as a case in point.