Slippery slope -- Asserting that if we allow A to happen, then Z will consequently happen too, therefore A should not happen.
Slippery slope is not necessarily a fallacy -- it is only a fallacy if the warrant is extreme. If we were arguing about dropping lit cigarette butts into the trash and we both agreed that fires were bad and I made the claim that we shouldn't drop cigarettes in the trash because they often cause fires (and cited data to that effect) there would be no fallacy in my argument.
Appeal to authority -- Using the opinion or position of an authority figure, or institution of authority, in place of an actual argument.
This one actually infuriates me because it may be the most wrongly used logical fallacy. First, because most people who use this argument are not making a logical deduction, they're making a statistical argument. To cite from Wikipedia:
The appeal to authority may take several forms. As a statistical syllogism, it will have the following basic structure:[1]
Most of what authority a has to say on subject matter S is correct.
a says p about S.
Therefore, p is correct.
The strength of this argument depends upon two factors:[1][2]
The authority is a legitimate expert on the subject.
A consensus exists among legitimate experts on the matter under discussion.
There is nothing wrong with this argument. The example given in the poster is very bad. As a PL student I have a bit of experience with formal methods and I'll remind everyone that the only things which can be proven using logic are those which follow directly from the definitions. In real arguments, these structures basically never exist. The truth never follows necessarily from the things people say, it's almost always a statistical argument. The fact that medical authorities used to think wrong things about the body didn't necessarily mean that it is or was wrong to believe in the body of scientific knowledge at the time. It only means that you have to be aware of the error margin in your statistics.
First, those links address my points improperly, if at all. Slippery slope once again assumes that the warrant is extreme but there is no such guarantee in the construction presented.
Second, even if the extra links on the site were correct I don't see why that would make the poster itself any less wrong.
From the description: "[...] shifts attention to extreme hypotheticals [...]" (also present in the A1 poster)
In any case, I see this poster as a cheat sheet: not something to learn from, but a quick reference for someone who has already done their homework and learned about those logical fallacies.
The problem is that I can't see how this is correct, even to the trained eye:
Argument from authority -- It is important to note with this fallacy that authorities in given fields may very well have valid arguments, and that one should not dismiss another's experience and expertise. To form an argument, however, one must defend it on its merits i.e. know why the person in authority holds the particular position that they do.
Only vaguely. You don't actually have to know what general relativity is or how it works to form an argument based on the statistical likelihood of physicists being correct. You don't really have to know "why" the person in authority holds their opinion, only that they are both an expert on the topic they are covering and that there is a general consensus among other experts on the same topic.
It is, of course, entirely possible that the opinion of a person or institution of authority is wrong; therefore the authority that such a person or institution holds does not have any intrinsic bearing upon whether their claims are true or not.
I'm not sure what this is trying to say, but there's certainly justifiable basis in believing that person's authority has an intrinsic bearing on the probability that their claim is true.
How is the quoted definition at all useful -- to either logicians or laymen?
Perhaps suggest an alternative to be put on the poster? Seems like this is a project about education and not about proving expertise in logical fallacies. I would expect that the authors would be welcoming of increased clarity and correctness in their examples.
> Slippery slope is not necessarily a fallacy -- it is only a fallacy if the warrant is extreme
It's only a fallacy to suggest that a slippery slope is logically entailed, i.e. to assert without further argument that it's logically necessary for Z to happen if A happens simply because Z is a more extreme form of A.
But slippery slopes certainly do exist empirically, especially with respect to politics, and given the actual nature of a particular political milieu, it can be entirely reasonable to suggest that implementing one policy makes it more likely for a more extreme version of that policy to come about in the future.
> Appeal to authority -- Using the opinion or position of an authority figure, or institution of authority, in place of an actual argument.
Same thing here. As you've said, it's entirely reasonable to apply a heuristic that gives greater credibility to a source whose statements have previously proven useful or valid than one that hasn't.
The fallacy of the appeal to authority is in assuming that a given statement is true because an 'expert' has made it, but it's entirely reasonable to expect that a person you regard as an expert is more likely to make statements that are valid in their own right.
The primary fallacy of those who excessively discuss logical fallacies is the assumption that most arguments actually are applications of deterministic logic intended to assert an unambiguously correct answer to an unambiguous question in the first place. Most arguments really involve the application of heuristics in complex and stochastic contexts, in order to establish an understanding that's marginally more useful than the status quo. Many arguments also involve attempts to reconcile competing conceptions of the desired end state, and are not simply disputes regarding the reasoning applied to obtain a presumptively agreed-upon end state. Formal deductive logic isn't the most relevant template in either of these scenarios.
Slippery slope -- Asserting that if we allow A to happen, then Z will consequently happen too, therefore A should not happen.
Slippery slope is not necessarily a fallacy -- it is only a fallacy if the warrant is extreme. If we were arguing about dropping lit cigarette butts into the trash and we both agreed that fires were bad and I made the claim that we shouldn't drop cigarettes in the trash because they often cause fires (and cited data to that effect) there would be no fallacy in my argument.
Appeal to authority -- Using the opinion or position of an authority figure, or institution of authority, in place of an actual argument.
This one actually infuriates me because it may be the most wrongly used logical fallacy. First, because most people who use this argument are not making a logical deduction, they're making a statistical argument. To cite from Wikipedia:
The appeal to authority may take several forms. As a statistical syllogism, it will have the following basic structure:[1] Most of what authority a has to say on subject matter S is correct. a says p about S. Therefore, p is correct. The strength of this argument depends upon two factors:[1][2] The authority is a legitimate expert on the subject. A consensus exists among legitimate experts on the matter under discussion.
There is nothing wrong with this argument. The example given in the poster is very bad. As a PL student I have a bit of experience with formal methods and I'll remind everyone that the only things which can be proven using logic are those which follow directly from the definitions. In real arguments, these structures basically never exist. The truth never follows necessarily from the things people say, it's almost always a statistical argument. The fact that medical authorities used to think wrong things about the body didn't necessarily mean that it is or was wrong to believe in the body of scientific knowledge at the time. It only means that you have to be aware of the error margin in your statistics.