The very first time we’re asked to write a paper, we’re told to appeal to authority figures. Supposedly, we guarantee the information’s truth by doing so.
Of course, it’s not always guaranteed that the information is true. In light of that, we’re told to cite relevant authority figures — people who have spent a lot of time studying the subject we’re writing about. Statements from relevant sources are taken to be more likely true.
This makes a certain amount of sense. Experts have mastered a particular subject’s reasoning, techniques, background information, etc. It, therefore, seems likely that the expert’s statements are true. Hence, our claims appear to be on solid grounds when we appeal to an expert.
I’m not entirely convinced that appealing to relevant authority is good reasoning.
I’m of the opinion that any appeal to authority, insofar as we’re trying to make truth claims about the world, is fallacious. However, that’s an argument for another time. I’m taking a more modest aim: establishing that we have good reason to doubt that an appeal to relevant authority is always a good bet.
Here are three problems with that way of reasoning:
- Assuming that authorities are reliable sources of information requires us to assume that their subject’s reasoning, techniques, background information, etc. is true, accurate, reputable, etc. That might not be the case. For one thing, the foundations of the subject might be inherently fallacious. For another, what counts as valid reasoning changes over time. This is how it is for science — what counts as valid scientific reasoning has changed at various points throughout history, often quite drastically.In fact, to test the accuracy of any subject matter’s claims, we often have to appeal to the criteria of either that subject matter or some other subject matter. If we appeal to the criteria of the relevant subject, we’ve committed circular reasoning. If we appeal to the criteria of another subject matter, we will quickly run into the same problem. In short, we enter into a series of dilemma after dilemma, each of which can only be resolved by another dilemma, each of which contains at least one circular solution.
Okay, so what’s the upshot? We’re appealing to someone who cannot justify their methodology non-circularly, which renders the connection between their subject mastery and veracity of their claims dubious at best. In plain English: we’re justified in being skeptical about truth claims, even from authorities.
- Here’s a much simpler problem (although not as powerful as the first one): the expert’s reasoning might be crap. Because, let’s be honest, experts make mistakes too. It happens. Unfortunately, no one is sufficiently equipped to check the veracity of every expert’s claims. We simply don’t have the subject-matter training necessary to do so. The answer to this problem, then, is not to blindly trust their word. For all we know, their reasoning is faulty. The proper attitude to adopt towards faulty reasoning is one of skepticism — we shouldn’t trust the conclusion, at least not entirely.
- We have know way of knowing whether or not the expert is being honest. For all we know, there’s a shadow conspiracy going on to keep the public uninformed. Now, there’s a reason why I’m putting this one last — the crazy-conspiracy-theory version of this problem seems inherently dubious.There’s a not-so-crazy version of this one, though, and it’s the problem of the status quo: an authority figure might make a truth-claim simply because they know that claims to the contrary are seen as crazy or blatantly false, even though they turn out to be true. This, obviously, is far more plausible and really does happen (e.g. pharmaceutical research).
The last two problems are resolvable — on the assumption that the first problem can be solved. I’m not sure that it can be solved, but that means we would have to be skeptical of the very possibility of knowledge.
What do you think? Is it reasonable to appeal to relevant authority in order to establish that a claim about the world is true? Do you think the first problem can be solved?