(Recommended Pre-reading: https://magehat.wordpress.com/2016/07/07/experts/)
This post is going to try and build on my previous post on experts, linked above. Specifically, I’m going to be looking at the idea of expert opinion and why it is useless as a persuasive tool, especially in the context of Coronavirus.
My previous post made the case that people tend to believe things that benefit them. For example, I don’t think a lot of people would believe in atoms and the rest of Chemistry if the theory was all we had. People believe in it because we use the results of that theory every day. As you drift away from theories grounded in repeatable experiments, especially towards ones that run on modelling, you drift away from people’s ability to trust them.
This causes splits to occur. People who want the theory to be true will believe the theory. People who want it to be false will not. A common idea at that point is to defer to experts, but what does it mean to be an expert?
At the risk of being a little contentious, I think an accurate definition of an expert is “a person smarter than you, who agrees with you.” If you think this is harsh, try a thought experiment. Think of a contentious issue where you know an expert that you trust, then imagine that someone with exactly the same credentials thought the opposite. Would that significantly decrease your confidence in your position? Or would you assume that the opposite person was just a crank or sheep, depending on their relationship to the consensus opinion?
I suspect most people would not do an about-face if they found out that someone with lots of relevant credentials disagreed with their opinion, and I doubt they would even consider that person an expert. Some try and solve this by using expert consensus as a proxy for which experts to trust, but consensus is not infallible. This is especially true closer to issues with no clear right answer and lots of experts on both sides, like whether to raise the minimum wage.
This is why I think that while we dodged a bullet with Coronavirus in a physical sense, we took one in a societal sense.
As a virus, SARS-COV-2 could have been much worse. It could have had a much higher fatality rate, or hit with significant side-effects in a larger percentage of cases. It could attack children and young people with more vigour, which would make things like keeping schools or factories open much harder and the supply chain could fall apart. In a sense, we avoided catastrophe.
But because of all these factors, it is able to evade black and white decision making. This turns it into less of a chemistry problem and more of a minimum wage style problem. The consequences of the virus are small enough that the average person doesn’t notice them unless things have gotten way out of control. Even then, the chances of lasting consequences to an individual are low enough that they are likely to end up with the same outcome regardless of their action. Especially if you aren’t in a packed major city, you could lock down and mask-up all the time, or go out and party with your friends consistently, and odds are in both cases you will end up fine at the end of it even if your actions may have inadvertently hurt others.
The fact the virus is comparatively benign in terms of consequences makes it much harder to figure out the right way to fight it. Use Sweden and lockdowns as an example – Sweden elected not to lock down at the beginning of the virus and focused on protecting vulnerable populations while allowing the rest of the population far more degrees of freedom. As a result, it is held up by both sides as a shining example of either what to do, or what not to do.
The reason it can be both is because the statistics aren’t lopsided enough to be inarguable in either direction. What we want as humans is a large identifiable signal. For example, if nobody in Sweden died of Coronavirus, we would say their strategy worked. If they all died, we would say it failed.
Instead, anti-lockdown activists can say the strategy worked because out of 30 European countries analyzed, Sweden only suffered a 7.7% increase in excess deaths in 2020, beating 21 countries. This is compared to highs of 16-18% from the worst hit, and the worst hit enacted severe lockdowns. On the other hand, pro-lockdown activists point to the fact that when compared to its direct neighbours (who did lock down), Norway, Finland and Denmark, Sweden saw significantly higher excess deaths. Norway didn’t even register any excess deaths at all.
The signal is small enough that you can likely find some data to support whatever view you want, even if it is wrong. This means that if you are looking for an expert consensus, you will still find lots of qualified people that disagree with you. When you combine that with antics like the experts of the U.S. changing their stance on masks partway through the pandemic, you can see how trust erodes rapidly. Regardless of your opinion on masks, the experts were wrong on it at some point.
Because of this, I think discussions of Coronavirus problems can’t function based on appeals to experts. Even if you tend towards trusting them, this isn’t the kind of issue where it makes sense. Trusting experts turns a complicated problem of uncertainty into a black and white decision, but it does so at the cost of finding any kind of middle ground or eventually converging on a good answer. Instead you end up with polarization, which gets everyone nowhere.
I think a better solution is transparency. Let people know as many of the actual details as possible and try not to stifle the debates with appeals to expertise. Build trust and admit that your side doesn’t know 100% that this is the best path forward, but that you think it is better than all the alternatives, instead of claiming you’re right because Jim said so.
You can trust me on this – I’m an expert.
Peace out.
Magehat's Interesting Things 