A Discussion About Epistemic Authority

I’ve been browsing some blogs using the tag “epistemology” for the first time yesterday and immediately found an entry that struck me as interesting and well written. In his post “Trust the Experts? Maybe Not”, Ryan Cook discusses why “we have good reason to doubt that an appeal to relevant authority is always a good bet”.

I felt inclined to make some minor remarks and ended up writing this huge comment, which is substantial enough to make for an own blog post. It can be read on its own, but if you’d like to read the whole discussion you can do so here.


The topic of epistemic authority is an interesting, relevant, and also difficult one. If I understand you correctly, you are imagining the situation of an individual that has just started to learn about a certain scientific topic. Because of this, I assume that the person only has a very basic stock of information and is not yet fully trained in the method of the relevant science either (if one presupposes a more advanced student the situation changes, of course).

You mention one general problem when it comes to making use of another person’s premises. We have to assume quite a lot of things:

a) honesty on their part
b) correctness of reasoning and
c) use of adequate methods and technology

I wish to add two more issues:

d) epistemic security – and, therewith, I describe my knowledge of the track record (the beliefs that are (implicitly) presupposed) and the justification relations that have been used by that person.
e) existing power structures and biases (I realize that you included that point in your text already. I just want to separate it from the second one, because I think it is of greater importance).

I assume that only the fourth (and the last) aspect(s) are of importance in the situation described above and I’ll try to show you why I think that this is the case.

Discussion of b): The expert’s reasoning may be faulty.
The first thing I want to assume is that an authority in a certain scientific field usually makes his argument transparent to the reader, which means that we can analyze it and decide for ourselves. I understand that some of the commonly used premises might be problematic, but considering the situation of the new student, I would still assume that the expert is in a distinctively better position to evaluate the existing premises and theses. This is the case, because the person is simply much more informed and can access a greater amount of data to make their decision.
Additionally, although it is certainly not impossible for anyone to be wrong, the papers that an expert has written are peer reviewed and discussed by his critics. The person would not have the status of an expert for a very long time, if their reasoning had a tendency to be faulty.
On the other hand, there is a student, who is still in training and has not fully developed his argumentative skills. He has a smaller stack of known argumentative structures and common fallacies and is therewith much more likely to make mistakes. Additionally, all problems that you describe for the expert in this section can also be applied to the student.

An analogous argument can be made for problem c).

Problem a): Honesty
My assumption is that dishonesty usually does not occur without any reason (which is mainly an empirical question, but depends on ones conception of a “lie”). Unless there is some benefit for the scientist or some underlying issues of power (point e)), it is unlikely that the expert is lying. Therewith, the problem should be attributed to the last problem mentioned above.
Although people do lie, it is not reasonable to assume that they do so most of the time, because it would render any scientific discourse useless and leave us with the possibility of absolute epistemic egoism, only (this position is problematic, because it only enables us to gain a very limited amount of knowledge about the world).

Problem d): Epistemic security
This reason can even be used to argue for epistemic egoism. Everything else equal, we will always have a better track record of our own beliefs and justification relations, because we can try to access them introspectively. We also have the possibility to try and be sensitive to our own biases and include this self-knowledge in the evaluation of our argument (although this is not successful in many cases).
On the other hand, we can only use the information that another person offers us when we rely on their assumptions. There is also always the difficulty of miscommunication.

You already gave reasons for e) in your post.

Most of my thoughts about this are evidently based on the study of philosophy, because I’m most knowledgeable in that field. There may be differences when focusing on other sciences.


A last note: Although I was arguing for the trust in an expert that I tried to describe as a reasonable epistemic authority, I do not in the least want to encourage people to not read and think critically! Please do so and please keep doing so when you advance in your studies. If you study a lot, there will be the day where you come to be one of the experts and contribute to your field (and maybe even significantly change it).
I hope to see that day soon as well😛.