Logical fallacies are a dangerous epistemological tool. I'm not as dismissive of them as you appear to be here, because I do think it's legitimately useful to give names to common patterns of bad reasoning so you can dismiss them in the future without having to go through the effort of explaining in detail why they're wrong every time you encounter them. That would be a waste of mental labor and force you to spend less time evaluating novel or compelling arguments. Basically, accusing someone of a logical fallacy is analogous to invoking a previously proven theorem in a mathematical proof. Invoking them is a shorthand for, "This argument is wrong for the same reason other arguments that match this pattern are wrong." Of course, the problem comes when people don't actually know what the fallacious pattern of reasoning is, don't know why it's wrong, or don't realize that the argument they're calling fallacious doesn't have the characteristics that make the pattern of reasoning wrong (this last case is basically someone using an epistemological version of the non-central fallacy). For almost all logical fallacies, the flaw in the argument isn't, "This argument provides literally no evidence for its conclusion," but, "This argument provides very weak evidence in most contexts where it's used, but is often presented as if it's strong evidence." Or occasionally, "This would provide evidence in some circumstances, but it's screened off in this context."
Unfortunately, many people who use logical fallacies don't understand this and use them as thought-terminating clichés, the exact opposite of how they're supposed to be used. The ad hominem example you give is a great one. I'm pretty sure the majority of people online who know the term think that it just means, "insult." And even those who do understand the difference between an ad hominem and an insult likely haven't thought about the fact that undermining the credibility of the source is a perfectly non-fallacious form of ad hominem argument.
//'m not as dismissive of them as you appear to be here, because I do think it's legitimately useful to give names to common patterns of bad reasoning so you can dismiss them in the future without having to go through the effort of explaining in detail why they're wrong every time you encounter them. //
I think some are fine. But most of them are such that most of the time they're misused and are not helpful in identifying bad arguments as they preuppose the argument is bad.
I laughed 5 times. So it might have been the funniest thing you've written, as the Note suggested.
This is another useful lens through which to see how many cognitive shortcuts we use to avoid the hassle of thinking. Having been on the receiving and dishing out end of this I can confirm that it's all extremely tiresome. So (to quote the favourite thing people would ask when I was biggish on Twitter) what do we do about it?
Nothing. Just feel smug and insightful is the answer I ended up with, before deleting my account without sharing it.
My favorite proof that the ad populum argument isn't always fallacious:
If popularity is, or can be, evidence for a claim, then it's not fallacious to point this out. So someone who thinks ad populum is always a fallacy is committed to the claim that popularity never provides any evidence at all for a claim. But this would imply that people's beliefs have no correlation (or a negative correlation) with the truth, since, if people's beliefs are correlated with truth, it's more likely that P will be a popular belief if P is actually true, and thus P's popularity is Bayesian evidence for P. But if our beliefs are uncorrected with the truth, we don't really know anything and should not trust any of our beliefs or the faculties that formed them.
In short, believing that ad populum is always a fallacy commits one to global skepticism.
I think we ought to be more charitable to the concept. When good faith, truth-seeking people use the phrase "ad populum", they're very often talking about a situation where incredibly easy empirical tests are available and the arguer is ignoring or refusing them in favor of group memes. A clear example right now would be people claiming the WNBA players are about as good at basketball as the NBA players.
"Pattern-matching" is an overly general way of characterizing some commonly copied rhetorical maneuvers that are mostly driven by rationally impertinent factors. The bulk of human cognition is pattern matching--it doesn't corrode thinking, it is thinking, give or take a few. Deduction, induction, and analogical reasoning are all instances of pattern-matching (of logical structures, observations, and relationships among parts, respectively). I grok what you mean by "cheap intellectual shortcut[s], whereby people don’t seriously think through ideas but instead just pattern match," but "ideological bingo" was a better label for this behavior I'm afraid. It's reasoning via the shortest path to the thing you want to say for unrelated reasons, and the shortest path is often a cued response that was learned because it was observed to be effective in practice. My intuition is that you can't truly eliminate the pull of the conclusion's import on the reasoning process, but in striving to minimize it, it's best to characterize it as sharply as you can. Reduce it to the most accurate and transferable pattern, if you like, to capture the instances you mean to. Non-philosophers might struggle with discerning the logic beneath more salient particulars of a situation, but epistemologists would all do well to study the particulars of the meat machine we use to make logic happen.
Bulldog, what is the *use* of philosophical theism? I have read Edward Feser and Thomas Aquinas and they managed to address maybe 5% of the Bible and contradict much of the rest. When they are talking about something that does not consist of parts and does not change (so does not get angry), they are not talking about Yahweh.
Or maybe I am missing something? Have there ever been people who ignored traditional religious practices and started a new religion based on philosophical theism? But what is the point? We cannot philosophically prove that worship is useful (though it could be proven by social psychology that it might simply remind people of their moral views, because our biggest problem is not bad morals but just forgetting about our moral beliefs without constant reminders), we cannot say much about the afterlife except that we can assume a sufficiently powerful and good being does not accept the utility loss of permanent death and so on.
This is funny, and yes, most "fallacies" are actually some sort of informal argument which do provide evidence of the conclusion. (The quality of that evidence can vary widely.)
On the other hand, are you not "pattern matching" the "pattern matchers" right here in this article? Is the problem you are critiquing something you are in fact engaging in?
People forget the “logical” part of “logical fallacies”. They’re only actually fallacious in a formal logical argument. “If A then B. B, therefore A” is fallacious. “If A then B. B, therefore A is likelier” isn’t.
I have been misled by purportedly good arguments so much in my life, that I find pattern-matching safer. I loved the deductive logic of mises.org and did not notice for years this geometrical thinking has a huge weakness: you do not use empiricism to prove the Pythagora theorem, but you do use empiricism to check whether that triangle is on an Euclydian plane. So basically their ideology completely failed at showing how to actually apply it to the real world. I have also been misled by heredity-and-IQ by twin studies, which looked like excellent empirical evidence.
Pattern matching works better. I notice famous people do not have famous parents and kids, which shows IQ is not hereditarian much. I also notice literally every politician who wasn't downright crazy or evil eventually ended up with going for some level of taxed-and-regulated capitalism and the rest is really just details.
It seems to me that fallacies are articulated to emphasise the importance of a specific conclusion Y not necessarily following from X, pointing out that more work needs to be done to establish it.
That's not the same as saying that it's a completely invalid inference, but a warning sign for "be careful about what you conclude on that basis". As you point out, it can go the direction of completely paralyzing honest thought when it drifts in the direction of implying a thought pattern is completely invalid.
The model involving the husband, wife, and McDonalds seems open to interpretation regarding what sort of game is being played.
Examples:
1 An adversarial game where Wife wants to test the hypothesis of whether there's a McDonalds nearby, and Husband's goal is to debunk each of her proposed tests
2 Same set up as 1, but Husband's goal is to cooperatively lead Wife to the best possible test by pointing out potential problems with each proposed test
3 Same set up as 1, but Husband's goal is to strengthen via practice Wife's ability to respond to sophistry
In sum, it seems Husband's responses are geared towards some goal. Moreover, it seems that "patterning" could alternatively be defined as how tailored his responses are to his goal. That is to say, there could be a hypothetical "ideal pattern" for each type of game, including those where the goal was "mutual and respectful cooperative pursuit of truth" (as far-fetched as that may sound for American married couples)
wow, so just because my dismissal comes from pattern-matching, it's not legitimate? genetic fallacy much?
idk why i felt the need to be the first to make the stupid joke
Logical fallacies are a dangerous epistemological tool. I'm not as dismissive of them as you appear to be here, because I do think it's legitimately useful to give names to common patterns of bad reasoning so you can dismiss them in the future without having to go through the effort of explaining in detail why they're wrong every time you encounter them. That would be a waste of mental labor and force you to spend less time evaluating novel or compelling arguments. Basically, accusing someone of a logical fallacy is analogous to invoking a previously proven theorem in a mathematical proof. Invoking them is a shorthand for, "This argument is wrong for the same reason other arguments that match this pattern are wrong." Of course, the problem comes when people don't actually know what the fallacious pattern of reasoning is, don't know why it's wrong, or don't realize that the argument they're calling fallacious doesn't have the characteristics that make the pattern of reasoning wrong (this last case is basically someone using an epistemological version of the non-central fallacy). For almost all logical fallacies, the flaw in the argument isn't, "This argument provides literally no evidence for its conclusion," but, "This argument provides very weak evidence in most contexts where it's used, but is often presented as if it's strong evidence." Or occasionally, "This would provide evidence in some circumstances, but it's screened off in this context."
Unfortunately, many people who use logical fallacies don't understand this and use them as thought-terminating clichés, the exact opposite of how they're supposed to be used. The ad hominem example you give is a great one. I'm pretty sure the majority of people online who know the term think that it just means, "insult." And even those who do understand the difference between an ad hominem and an insult likely haven't thought about the fact that undermining the credibility of the source is a perfectly non-fallacious form of ad hominem argument.
//'m not as dismissive of them as you appear to be here, because I do think it's legitimately useful to give names to common patterns of bad reasoning so you can dismiss them in the future without having to go through the effort of explaining in detail why they're wrong every time you encounter them. //
I think some are fine. But most of them are such that most of the time they're misused and are not helpful in identifying bad arguments as they preuppose the argument is bad.
I laughed 5 times. So it might have been the funniest thing you've written, as the Note suggested.
This is another useful lens through which to see how many cognitive shortcuts we use to avoid the hassle of thinking. Having been on the receiving and dishing out end of this I can confirm that it's all extremely tiresome. So (to quote the favourite thing people would ask when I was biggish on Twitter) what do we do about it?
Nothing. Just feel smug and insightful is the answer I ended up with, before deleting my account without sharing it.
Thank you!
My favorite proof that the ad populum argument isn't always fallacious:
If popularity is, or can be, evidence for a claim, then it's not fallacious to point this out. So someone who thinks ad populum is always a fallacy is committed to the claim that popularity never provides any evidence at all for a claim. But this would imply that people's beliefs have no correlation (or a negative correlation) with the truth, since, if people's beliefs are correlated with truth, it's more likely that P will be a popular belief if P is actually true, and thus P's popularity is Bayesian evidence for P. But if our beliefs are uncorrected with the truth, we don't really know anything and should not trust any of our beliefs or the faculties that formed them.
In short, believing that ad populum is always a fallacy commits one to global skepticism.
Nice!
I think we ought to be more charitable to the concept. When good faith, truth-seeking people use the phrase "ad populum", they're very often talking about a situation where incredibly easy empirical tests are available and the arguer is ignoring or refusing them in favor of group memes. A clear example right now would be people claiming the WNBA players are about as good at basketball as the NBA players.
"Pattern-matching" is an overly general way of characterizing some commonly copied rhetorical maneuvers that are mostly driven by rationally impertinent factors. The bulk of human cognition is pattern matching--it doesn't corrode thinking, it is thinking, give or take a few. Deduction, induction, and analogical reasoning are all instances of pattern-matching (of logical structures, observations, and relationships among parts, respectively). I grok what you mean by "cheap intellectual shortcut[s], whereby people don’t seriously think through ideas but instead just pattern match," but "ideological bingo" was a better label for this behavior I'm afraid. It's reasoning via the shortest path to the thing you want to say for unrelated reasons, and the shortest path is often a cued response that was learned because it was observed to be effective in practice. My intuition is that you can't truly eliminate the pull of the conclusion's import on the reasoning process, but in striving to minimize it, it's best to characterize it as sharply as you can. Reduce it to the most accurate and transferable pattern, if you like, to capture the instances you mean to. Non-philosophers might struggle with discerning the logic beneath more salient particulars of a situation, but epistemologists would all do well to study the particulars of the meat machine we use to make logic happen.
Bulldog, what is the *use* of philosophical theism? I have read Edward Feser and Thomas Aquinas and they managed to address maybe 5% of the Bible and contradict much of the rest. When they are talking about something that does not consist of parts and does not change (so does not get angry), they are not talking about Yahweh.
Or maybe I am missing something? Have there ever been people who ignored traditional religious practices and started a new religion based on philosophical theism? But what is the point? We cannot philosophically prove that worship is useful (though it could be proven by social psychology that it might simply remind people of their moral views, because our biggest problem is not bad morals but just forgetting about our moral beliefs without constant reminders), we cannot say much about the afterlife except that we can assume a sufficiently powerful and good being does not accept the utility loss of permanent death and so on.
This is funny, and yes, most "fallacies" are actually some sort of informal argument which do provide evidence of the conclusion. (The quality of that evidence can vary widely.)
On the other hand, are you not "pattern matching" the "pattern matchers" right here in this article? Is the problem you are critiquing something you are in fact engaging in?
People forget the “logical” part of “logical fallacies”. They’re only actually fallacious in a formal logical argument. “If A then B. B, therefore A” is fallacious. “If A then B. B, therefore A is likelier” isn’t.
I have been misled by purportedly good arguments so much in my life, that I find pattern-matching safer. I loved the deductive logic of mises.org and did not notice for years this geometrical thinking has a huge weakness: you do not use empiricism to prove the Pythagora theorem, but you do use empiricism to check whether that triangle is on an Euclydian plane. So basically their ideology completely failed at showing how to actually apply it to the real world. I have also been misled by heredity-and-IQ by twin studies, which looked like excellent empirical evidence.
Pattern matching works better. I notice famous people do not have famous parents and kids, which shows IQ is not hereditarian much. I also notice literally every politician who wasn't downright crazy or evil eventually ended up with going for some level of taxed-and-regulated capitalism and the rest is really just details.
It seems to me that fallacies are articulated to emphasise the importance of a specific conclusion Y not necessarily following from X, pointing out that more work needs to be done to establish it.
That's not the same as saying that it's a completely invalid inference, but a warning sign for "be careful about what you conclude on that basis". As you point out, it can go the direction of completely paralyzing honest thought when it drifts in the direction of implying a thought pattern is completely invalid.
The model involving the husband, wife, and McDonalds seems open to interpretation regarding what sort of game is being played.
Examples:
1 An adversarial game where Wife wants to test the hypothesis of whether there's a McDonalds nearby, and Husband's goal is to debunk each of her proposed tests
2 Same set up as 1, but Husband's goal is to cooperatively lead Wife to the best possible test by pointing out potential problems with each proposed test
3 Same set up as 1, but Husband's goal is to strengthen via practice Wife's ability to respond to sophistry
In sum, it seems Husband's responses are geared towards some goal. Moreover, it seems that "patterning" could alternatively be defined as how tailored his responses are to his goal. That is to say, there could be a hypothetical "ideal pattern" for each type of game, including those where the goal was "mutual and respectful cooperative pursuit of truth" (as far-fetched as that may sound for American married couples)