Follow the reluctant adventures in the life of a Welsh astrophysicist sent around the world for some reason, wherein I photograph potatoes and destroy galaxies in the name of science. And don't forget about my website,

Wednesday 2 December 2015

Because I Said So

No-one is immune to being irrational. From time to time, everyone - without exception - commits logical errors. Some of these are trickier to understand than others, and one of them made little real sense to me until yesterday. In the hope that this will help anyone else who's confused about it, here's my short interpretation of the "appeal to authority" fallacy.

Asking for an expert opinion is not a fallacy in and of itself, but there are several ways in which it can become a fallacy :

1) They're not really an expert
Asking an expert for their opinion about something well outside their specialist field (sciolism). Stephen Hawking isn't especially qualified to talk about global warming, Richard Dawkins doesn't know jack about theology, and generally speaking engineers know nothing about flower arranging. You can certainly feel free to ask their opinion anyway, but there's not really any good reason to give Hawking's opinion on genetic engineering any more weight than that of Boris Johnson. It's an easy mistake to make, and possibly the most common form of the fallacy.

2) They're not an authority
If you want to build an aircraft, you hire an engineer, not a florist. When someone has a proven track record of successfully building aircraft, it makes sense to consider them to be both an expert and an authority on their subject. Engineers might not be correct 100% of the time, but they are vastly more reliable in their knowledge of what works and what doesn't than a team of a thousand florists.

But there's a difference between expertise and authority, especially when it comes to active research. Aircraft design is an incredibly well-tested and proven field. Meteorologists, on the other hand, make mistakes about tomorrow's weather all the time - there are no authority figures in meteorology. Of course, there are experts, and their opinions still count for more than those of pigeon fanciers. It's just that you can't use the opinion of a meteorologist to be certain of tomorrow's weather in the same way you can use an engineer's opinion as to whether a plane will be able to fly. An engineer can give you something close enough to certainty, a meteorologist deals in probabilities.

3) Because I said so
This seems to be the way the fallacy is most often stated but also explained the most badly. If you ask an expert for an opinion and they say, "well, I'm an expert, so I must be right", then they have committed a serious error. Their expertise doesn't by itself guarantee that they are correct. What their expertise should be able to do is allow them to rigorously justify their assertion - it does not entitle them to make sweeping claims based solely on their expert status and have everyone else kowtow to their infinite wisdom.

EDIT : Long after I originally posted this, I encountered a curious phenomena which may explain why people make the above error. It seems they sometimes confuse being asked (in good faith) for a demonstration of their expertise with having their expertise questioned. Of course no-one is obligated to participate in all debates, but when they choose to do so they should be careful to try and understand if someone is just trying to learn something or is using this as a veil for an attack.

Sometimes the experts themselves do this, worst of all when it's from one expert to another. Sometimes it's laymen looking to justify their own argument. "Well, this astronomer told me that the Moon isn't made of cheese, so it can't be." If the explanation stops there, then technically this is a fallacy. It amounts to saying, "I know this isn't true, so it can't be true" (or even more succinctly, "I must be right") which is obviously circular and doesn't even admit the possibility of error.

If, on the other hand, the explanation continues : "They said they measured the spectrum of the Moon and found it was inconsistent with that of any known cheese and actually is pretty similar to rocks, and I think they mentioned something about some sample return mission thingy..." then no fallacy has been committed. The astronomer has used their expert knowledge to rule out the Lunar Cheese Hypothesis, not relied on their status as an expert to quash the discussion.

The Appeal To Stupidity

On the other extreme, I also see people claiming the exact opposite : "they're an expert, so they must be wrong." Well, almost. The "I'm not a scientist, but..." defence is the idea that non-expert opinion can trump that of an expert in their own specialist field. Reality check time : unless that expert is exceptionally biased or has made a terrible mistake, it can't. True, experts might be wrong, and make mistakes just like anyone else. But though not infallible, they are far more likely to be right than any non-expert. That is, after all, their job. It's why we have them. There are cases where it is absurd to give equal weight to non-expert opinions.

So I rather disagree with the above meme, provisionally. There are some areas in which not being an expert does automatically make your point invalid. So you think you can build a rocket, and your qualification is, what... a degree specialising in Mesopotamian art ? You literally can't do basic arithmetic ? Then nope, sorry, your opinion is invalid. Of course there are areas in which intelligent non-experts can make valuable contributions and sometimes outdo the specialists, but sometimes it really is appropriate to dismiss non-expert opinions. It depends heavily on the details of the situation.


The "appeal to authority" argument, then, is somewhat subtle. When a subject is so well-established that there are figures who can be considered authorities, it's not a fallacy... as long as they're speaking about their specialist area. If the subject is controversial, then you should still give more weight to expert opinion, but you can't rely on one expert to be correct. This is especially true when the expert refuses to explain their reasoning. Being an expert is supposed to enable you to reason, not entitle you to pronounce judgement.

The "appeal to the consensus" argument is only a fallacy when experts do it in their own discipline. This creates a false consensus : it must be right because everyone else agrees with it, whereas a true consensus should be everyone coming to the same conclusion independently. Assuming the consensus has been arrived at properly, it's not a fallacy for non-experts to place more weight in the expert consensus than the opinions of non-experts - but it is usually wrong to assume it's 100% certain.

The "appeal to stupidity" argument is however very much worse than appeals to authority or the consensus. That experts are fallible in no way means that non-experts are somehow magically able to skip years of hard work. This doesn't mean that experts can use the "because I said so" fallacy to avoid discussing why they've come to their conclusions. Nor does it mean that when they take the time to explain the details to an intelligent non-expert, they can dismiss their opinion out of hand. It only means that if it's a choice between the opinion of an expert and an uninformed non-expert, the latter really is an appeal to stupidity.

No comments:

Post a Comment

Due to a small but consistent influx of spam, comments will now be checked before publishing. Only egregious spam/illegal/racist crap will be disapproved, everything else will be published.