thou shalt not commit logical fallacies

[Warren Dew]

Emphasis mine. Sorry for re-citing the same source I posted on a previous page, but maybe you skipped over it.

We didn't skip over it. We just recognized it as a fallacious appeal to authority.

You see, the problem is that some people don't understand what the meaning of the word fallacy is.

The topic is about logical fallacies. Haven't you noticed? It's a logical fallacy to argue that something is true on the grounds that an authoritative source says that it is, for the simple reason that past authoritative sources have turned out to be wrong. Appeals to subsequent authoritative sources encounter the same problem: How do we know they are right on account of their authority? It's turtles all the way down once more.

Exactly.

If you find that a legitimate authority agrees with you, does it make it more likely that you are correct? Yes, it may. Does it prove that you are correct? No, it does not; the authority may still be in error.

Formal logic is about proofs, not probabilities. Therefore appealing to an authority, which may affect probabilities but does not provide proof, is a logical fallacy.

Exactly the same situation pertains to many of the other logical fallacies mentioned in the original link. For example, take "false cause", also known as "a correlation does not prove a causation."

If two things are closely correlated, does it make it more likely that one causes the other? Yes, it does. Does it prove that one causes the other? No, it does not.

Just like appeal to authority, then, citing of correlations is a logical fallacy, because it does not provide a proof, only a probability.

If you don't know enough to reproduce the authority's data and reasoning, you probably don't know enough to be arguing the point in the first place.

[Hermit]

When you appeal correctly to a recognized authority, the burden of proof falls on the other party to refute.

I gave examples of people appealing to recognised authorities that turned out to be wrong. Who decides you have correctly appealed to an authority? Another authority? Oh wait. Now there's a genuine logical fallacy.

You're still drawing the same incorrect conclusion, that because some authorities have turned out to be wrong, therefore all authorities are wrong. There's a major logical fallacy right there. Your argument is going as follows:

THEOREM: That all appeals to authority are wrong.
- At least one authority has been wrong when appealed to.
- Therefore, all appeals to authority are wrong. QED.

Re the bolded bits: That is not my argument. To begin with, I have said that some experts turned out to be wrong. My argument is that it is fallacious to claim X to be true because an authority said it is. No more nor less than that.

[trdsf]

When you appeal correctly to a recognized authority, the burden of proof falls on the other party to refute.

I gave examples of people appealing to recognised authorities that turned out to be wrong. Who decides you have correctly appealed to an authority? Another authority? Oh wait. Now there's a genuine logical fallacy.

You're still drawing the same incorrect conclusion, that because some authorities have turned out to be wrong, therefore all authorities are wrong. There's a major logical fallacy right there. Your argument is going as follows:

THEOREM: That all appeals to authority are wrong.
- At least one authority has been wrong when appealed to.
- Therefore, all appeals to authority are wrong. QED.

Re the bolded bits: That is not my argument. To begin with, I have said that some experts turned out to be wrong. My argument is that it is fallacious to claim X to be true because an authority said it is. No more nor less than that.

If that isn't your argument, then why are you presenting that argument? You're dismissing all appeals to experts. That's incorrect. An appropriate reference to an appropriate authority is not fallacious. Or are you actually rejecting all knowledge that you personally can't reproduce for yourself?

No one person is the repository of all knowledge. That's the "magic" of the scientific method: I don't need to reproduce for myself most scientific findings -- given the equipment necessary to test the edges of modern science, I can't verify most of them for myself. So one team presents their findings. Another team (or multiple teams) verify them, or find where they went wrong.

This was so very perfectly demonstrated with the report of superluminal neutrinos. The team reported their findings. Several teams went out to verify them, and the original team went back and went over their experiment again, and eventually found the wiring problem that gave the false result.

They didn't come out and say, "Neutrinos move faster than light" and the entire scientific community just took their word for it.

This is how and why we can accept scientific reports at face value -- with the proviso "pending further observations", if you like -- and how and why an appropriate appeal to an appropriate authority is not prima facie fallacious. This is hugely different from, for example, some MD stating that the earth is only 6000 years old and expecting to take his word in the face of the consensus opinion of geologists, archeologists, biologists, and anthropologists.

[Hermit]

If that isn't your argument, then why are you presenting that argument? You're dismissing all appeals to experts.

You are reading stuff into my posts that I did not say, trdsf. I have said that some experts turned out to be wrong. My argument is that it is fallacious to claim X to be true because an authority said it is. No more nor less than that. That falls far short of dismissing all appeals to authorities. It just means one cannot in principle say that X is true because an authority says it's true. People who claim the opposite are committing the logical fallacy known as "appeal to authority." Or, to appropriate the words used in the Wikipedia article on the subject, "the conclusion does not follow unconditionally in the sense of being logically necessary."

[trdsf]

If that isn't your argument, then why are you presenting that argument? You're dismissing all appeals to experts.

You are reading stuff into my posts that I did not say, trdsf. I have said that some experts turned out to be wrong. My argument is that it is fallacious to claim X to be true because an authority said it is. No more nor less than that. That falls far short of dismissing all appeals to authorities. It just means one cannot in principle say that X is true because an authority says it's true. People who claim the opposite are committing the logical fallacy known as "appeal to authority." Or, to appropriate the words used in the Wikipedia article on the subject, "the conclusion does not follow unconditionally in the sense of being logically necessary."

I'm not reading anything in that you're not putting there. There are circumstances under which it is appropriate to accept the word of an authority. It's not absolute, but they do exist. It isn't prima facie fallacious.

Do you or do you not accept, for example, that the top quark exists? Unless you have access to a Tevatron-scale particle accelerator, I'm sure you haven't generated one on your own. Which means if you accept its existence, you are necessarily accepting the word of the researchers who postulated it, discovered it, and verified it. And if you say to someone "The top quark was discovered in 1995", you are just as necessarily appealing to those same authorities.

[Tyrannical]

When citing a proper expert, you can assume their view true unless evidence is given to the contrary. And in a typical debate, experts with opposing views are often cited. When dealing with recognized experts, a level of integrity is earned and expected. While occasionally bad apples just make up data, we assume their work is honest unless evidence suggests otherwise.
When dealing with actors that play Doctors on TV, there is no expectation of accuracy.

[Hermit]

There are circumstances under which it is appropriate to accept the word of an authority. It's not absolute, but they do exist. It isn't prima facie fallacious.

Please stop conflating empirical truths with logical truths. I am much more likely to take an authority's word for saying that X is true than a random person's post on a forum. This in no way implies that, speaking in terms of logic, X is true because authority Y says it's true.

When citing a proper expert, you can assume their view true unless evidence is given to the contrary.

Yes. Agreed. You're not telling me anything I don't know already. We are talking about logical fallacies in this thread, though. It is a logical fallacy to say that X is true because an authority says it's true.


To adapt the relevant words from the Wikipedia once more: "the conclusion [that X is true because authority Y says it's true] does not follow unconditionally in the sense of being logically necessary."

[FBM]

appeal to authority -- Known also as the argumentum ad verecundiam fallacy. An appeal to authority is ordinarily one good way to buttress a line of thought. The practice becomes fallacious when one of the following happens: the authority is not an expert in the field in which one is speaking; the allusion to authority masks the fact that experts may be divided down the middle on the subject; no explicit reference is made to the authority.

http://www.philosophicalsociety.com/Log ... htm#appeal

A few thoughts:

To assert that the words of an authority prove something to be undeniably correct or accurate is sloppy argumentation. As Seraph points out, authorities are wrong often enough to make such an absolute claim untenable. OTOH, to use an authority's words to support one's reason for holding a certain position isn't. Since no one can be an expert in everything, abiding by the best available current scholarship is the logical way to go and doing so isn't fallacious.

If any appeal to any relevant authority were always fallacious, there would be no need for the sort of qualifications included in the above-quoted definition of the fallacy. IOW, one can appeal to a relevant authority correctly (as support of one's position and an at least tacit recognition that the authority may turn out to be wrong) and one can also appeal to a relevant authority incorrectly (as proof that one's position is undeniably true).

When a researcher submits work to a peer-reviewed journal, the work is reviewed by relevant authorities in that field, not people from unrelated fields. Peer review itself is a non-fallacious appeal to relevant authorities. But no careful researcher (or editor) would go so far as to say that since someone's work has passed peer review(s) that it is undeniably true for all eternity. That's just sloppy, at best. But judging from the definitions of argumentum ad verecundiam that I've seen to date, it's not included in the definition of the fallacy. Utilizing the works of predecessors is indispensible in making progress in practically any field. Otherwise, everyone would have to re-invent the wheel for him/herself, and collective human knowledge could never progress beyond a single life-span. A wise researcher uses and represents the work of his/her predecessors as tentatively true, not absolutely so (except for such trivial things as definitions, tautologies, etc).

[Coito ergo sum]

Fast and loose with the definition. There is an overlap between rhetorical tricks and subterfuges on one hand and logical fallacies on the other, but a list of each is not identical. An ad hominem, for example, is not a logical fallacy, and the gambler's fallacy is not usually employed as a rhetorical trick or subterfuge.

Agreed. And "slippery slope" is not even a fallacy; if you can show that A will actually lead to bad thing B, that actually is a reason to avoid A.

It's not a fallacy when you can show that B happening is a good possibility. But when you're talking about A eventually leading to Z, it becomes a logical fallacy. I've always found slippery slopes to be feeble arguments; they tend to rely on paranoia.

Whether a slipper slope is considered valid or not generally depends on whose ox is being gored.

[Coito ergo sum]

Or people forget that an appeal to authority is not a fallacy, when the authority actually is an expert.

Experts are still frequently mistaken, so it's still a fallacy.

No it's not.

Authority is not fallacious when the issue is credibility.

When authority is proposed as proof of an assertion, however, then it is a fallacy. It's not proof that there are black holes to point to Stephen Hawking and say "Stephen Hawking says there are black holes, therefore, there are black holes." However, the arguments and proofs Stephen Hawking has provided for black holes are good proof that there are black holes. And, his authority on the subject lends to the credibility that he is presenting fair and honest proofs.

However, at bottom, it's not his authority, but the verifiability of his proofs that support the theory.

[Hermit]

no careful researcher (or editor) would go so far as to say that since someone's work has passed peer review(s) that it is undeniably true for all eternity. That's just sloppy, at best. But judging from the definitions of argumentum ad verecundiam that I've seen to date, it's not included in the definition of the fallacy.

You too seem to conflate the empirical aspect of appeals to authorities with its meaning of logical fallacies. The poster linked to at the start of the thread was about the latter. I don't know how many times I have to reiterate that I have no problem with people, including myself, deferring to properly qualified authorities as far as empirical matters are concerned. Is anyone suggesting that I might not prefer the conclusions made by people aware of the law of gravity, theory of general relativity, the concepts of quantum mechanics, competent with calculus and familiar with observations made via radio-telescopes over those of bronze age goat herders? I don't know how many more times I need to reiterate that this thread is about a list of logical fallacies? I argue that from this angle the logical fallacy of appealing to an authority essentially goes like this: Authority Y says X is true. Therefore X is true. As you noted, argumentum ad verecundiam is another label for appeal to authority. Are you aware that the literal translation of argumentum ad verecundiam is "appeal to reverence"? Not very scientific, is it?

[FBM]

Then I don't see how we might disagree.

[Tyrannical]

When citing a proper expert, you can assume their view true unless evidence is given to the contrary.

Yes. Agreed. You're not telling me anything I don't know already. We are talking about logical fallacies in this thread, though. It is a logical fallacy to say that X is true because an authority says it's true.


To adapt the relevant words from the Wikipedia once more: "the conclusion [that X is true because authority Y says it's true] does not follow unconditionally in the sense of being logically necessary."

The authority fallacy only occurs when the "expert" has been falsely called an authority. Blindly believing a bona fide authority when they are wrong is something else entirely. The fallacy occurs when the premise "they are an expert" is false.

[Hermit]

double post

[Hermit]

Then I don't see how we might disagree.

We don't. Unfortunately two or three other participants keep reading stuff into my post I never said and ignore what I did say. From my point of view their persistence in doing so injects a fair bit of frustration to the proceedings. In retrospect I might have been better off to terminate my participation in this thread after my first contribution, which is the third post on the first page or encouraging those who kept disagreeing with me to bloody well read it.