Fallacies: Appeal to Authority

[Squatch347]

So it took me awhile to get around to writing the first thread in what will hopefully be a series of discussions.

I'll point out that I tried to provide a contextual intro here: http://www.onlinedebate.net/forums/s...n-on-fallacies


The first fallacy is one I've thought about a lot over the last five or so years, the appeal to authority fallacy.


For those reading, Nizkor has a pretty good summary:

Description of Appeal to Authority

An Appeal to Authority is a fallacy with the following form:

1) Person A is (claimed to be) an authority on subject S.
2) Person A makes claim C about subject S.
3) Therefore, C is true.

This fallacy is committed when the person in question is not a legitimate authority on the subject. More formally, if person A is not qualified to make reliable claims in subject S, then the argument will be fallacious.

More here: http://www.nizkor.org/features/falla...authority.html

To frame this discussion, I would like to pose a few questions to my fellow Peers:

Do we agree that this is fallacious?

Are there exceptions to it being a fallacy?

What, specifically makes it a fallacy?

I think we will generally all agree that there are times when it is a fallacy. Arguing "The sky is blue because that mentally deranged homeless man said so" is clearly a fallacious argument given the working definition in the intro thread (A fallacy is when the statement “Conclusion is true because of premises” is false).

Often we will make exceptions when the expert cited in the premise is in a relevant field however.

IE: "Eating a low carb diet will facilitate weightloss because PhD X, a nutritionist said so."


While I 100% agree that this gives us warrant to hold the conclusion (I'm all about specialization and trade after all), it would still seem to run afoul of the definition, and what's more would still lead to untrue conclusions.

Clearly the diet doesn't elicit weightloss because the nutritionist said so. Rather, the nutritionist, presumably, had some reasoning why that conclusion was true and we are using her title as short hand for her evidence.

That would seem to be problematic from a philosophic point of view (and absolutely essential from a practical point of view). After all, if we are going to compose a logically supported argument, it should be the underlying evidence that is the support, rather than title of the person putting forward the evidence.


I'll set aside all the practical implications of that. Obviously it isn't an effective technique for debate because not many will understand that kind of argument. Obviously there are fraud issues and more benign issues revolving around the evolution of knowledge in science to be dealt with. I would suggest setting those aside.

So my question here (aside from the structural questions above), is do we simply accept the authority as a generally useful fallacy? Or is there a point I'm missing that makes it not fallacious in those scenarios?

[GoldPhoenix]

Off the top of my head:

Clive and I were recently discussing something of this nature, and he pointed out something that I think is worth repeating in this context. The rigor and strength of an argument goes beyond preventing a syntactic error (e.g. violating a basic law of logic or having an incoherent form), instead the more essential features of a strong and rigorous arguments go into preventing the much more complicated realm of semantic errors (A formally valid argument, but contains other, more subtle flaws in reasoning).

A.) I think that several of the informal fallacies are actually trying to get at semantic errors in reasoning, rather than syntactic errors in reasoning. An appeal to authority is a very valid argument, but it's taking on board several important, unexplained (and, in some instances, incorrect --it depends on the "expert") premises.

1.) The expert actually has a knowledge exceeding that of the laypersons on the subject.

2.) If asked to a give a proof, the expert can, in fact, give the evidence and the reasoning supporting their opinions, and this proof will be both valid and warranted (i.e. a sound argument).

3.) I think the part that's missed here is that an expert opinion is worth much less than a consensus opinion (i.e. getting a bunch of experts together and getting an overwhelming majority opinion, e.g. like 95% of them agree about proposition X). Strictly speaking, you should take a consensus opinion over an expert opinion every time. (This means collections of well-cited scientific papers with many proponents in the scientific community.)

So appealing to an expert is more about having well-founded or warranted premises than making an entirely self-contained valid argument (Of which none, outside of math, really exist). However, if any one of the premises listed above aren't true, then it's the appeal to authority is basically, by definition, a false appeal to a false authority, which is a fallacy.


B.) As stated, semantic errors aren't about logical form. So, yeah, everyone can agree upon "If A, then B. A. Therefore, B.", unless they're extremely green at logic. The subtle question, however, is "Why do you think A is a warranted assumption?" and "What warrants you to think that A and B are correlated in such a manner?" Essentially, this is why making a valid argument is such a low, low bar at some level. If you can stop yourself from making the trivial syntactic errors, you're still leaving yourself wide open to the vast array of semantic errors. The premises "If A, then B" and "A" are usually subject to inductive reasoning (probabilistic arguments) and abductive reasoning (intuitive reasoning), and those are often non-trivial to evaluate and understand.

The appeal to an authority is placing some trust in (1.)-(3.) above, but if that person is actually an expert, then there's good reason to suspect that they will be garbage handling at least some of the semantic errors for you. In other words, they'll be part of a community that hopefully has been confronted with and come up with effective methods of dealing with probability theory, honing their intuitions, understanding how to reason about issues related to A and B, they'll have an advanced network of knowledge about facts related to A and B, and so on. Therefore, all things being equal, they'll give more reliable premises to start your arguments from than people who don't have experience on the subject matter. But here the criterion is having reasonabe, reliable arguments, not having verified, 100% true facts of the matter.

[CliveStaples]

1.) The expert actually has a knowledge exceeding that of the laypersons on the subject.

I think this principle can be generalized to anyone who has epistemic advantage. So e.g. someone who has seen a painting 3 times should be trusted to recount its features more than someone who has seen a painting only once, ceteris paribus (e.g. WRT memory, attentiveness, faculties of sight, soundness of mind, etc.).