Is it worthwhile to teach about “Logical Fallacies?”
When in high school, one of my English classes had a unit on logical fallacies. Everyone was given a list of “logical fallacies” like “appeal to authority” and “slippery slope.” We had to do things like match examples with the names of the fallacies (which would almost always have multiple reasonable answers), and come up with examples of various fallacies.
At the time, I thought that this was a huge waste of time. My reasoning was that there were many more ways to give arguments incorrectly than to give arguments correctly, and that we should instead be teaching people what valid arguments are, and not to trust anything else. I have not really questioned this initial judgement until just now.
Now I am forming a new opinion on this question, and would like collect some opinions from Less Wrong.
I both disagree and agree with your high-school self.
Learning to recognize common failure modes, and developing a common language for talking about them with each other, is a relatively cost-effective way to improve the average validity of my arguments, in much the same way that reducing infant mortality is a relatively cost-effective way to increase average lifespan.
It doesn’t do anything to improve the validity of my most valid arguments, though. Depending on how reliably I reach that maximum and how high that maximum is, that might be OK. Or it might not, and an entirely different approach (like teaching what valid arguments are) might work better.
And the relationship between learning-to-X and teaching-X-in-high-school is of course a whole different problem.
All of that said, I’m curious: how would you go about teaching what a valid argument is, to a degree that “don’t trust anything else” is actually good advice to follow?
I do not know. To be honest, my high school self had a strong tendency to overestimate the rationality and learning potential of the general population.
Do you have any sense of how you learned it? For my own part, I feel like I learned what a valid argument is, to the extent that I have learned it, almost entirely by a series of negative examples. For that matter, I’m not sure I can articulate what a valid argument is in non-question-begging terms… though to be fair, I haven’t sat down and tried for five minutes.
I do not know how I learned how to argue, but I do not think it has anything to do with negative examples.
For me, it seems similar to understanding what is a valid mathematical proof (one which in theory could be expanded to following the logical rules at each step) but where you are allowed to make observations and probabilistic reasoning, all of which came naturally to me. I do not feel like I ever had any inclination to use logical fallacies, and I feel like I am quick to recognize when arguments do not make sense.
This is in contrast with cognitive biases. I feel like I am very dependent on parts of our brain that have biases, I will not be able to get past them easily, and can learn to mitigate them by being aware of them.
On the other hand, some fallacies recur much, much more often.
Ad hominem and appeal to authority are de facto THE way humans argue with each other, with strawmanning as a very strong third. So at least learning how to spot and correct those can alone improve someone rationality.
There are though much funnier ways to learn about fallacies, Biased Pandemic being one of my favourite.
Cognitive biases are different from logical fallacies.
Yes, in the sense that cognitive bias are a subset of logical fallacies systematically applied by our brain. So I maybe can refine my answer: is it worth to teach about logical fallacies? Yes, especially when they become cognitive bias. A fun way to do this is Biased Pandemics.
Is it worthwhile to teach about “Logical Fallacies?”
It’s worth learning about logical fallacies and internalizing them. It might not be worthwhile to teach people about them in school because people often don’t remember & internalize what they’re taught there.
My reasoning was that there were many more ways to give arguments incorrectly than to give arguments correctly, and that we should instead be teaching people what valid arguments are, and not to trust anything else.
While it’s important to be able to recognize & build a valid argument, it’s still been useful for me to use knowledge of fallacies to set mental triggers which activate when I mentally reach for a fallacy. Instead of unreflectively using an appeal to authority (for example) as a cognitive short-cut without checking whether it actually works, the not-quite-conscious sensation of making the appeal gets flagged for conscious attention, and I realize, “Oh! I’m making an appeal to authority. Does that appeal actually have much evidential weight?”
Edit: this isn’t directly responsive to you, but I can also imagine LWers who’ve moved on to noticing newerfallacies finding it harder to understand why it’s worth studying more well-known & canonical fallacies, even if the latter are as important.
There are “valley of bad rationality” effects here. Most so-called fallacies are in fact valuable heuristics. For instance, appeal to authority and slippery slope are both generally correct.
Knowing about fallacies creates in people the illusion that they can tell good arguments from bad. It lets them refute, in a superficially intelligent way, whatever position they don’t want to believe.
Though I do think it’s worth learning them anyway, for some people, at least so you know what people mean when they say “ad hominem”.
I don’t think that there evidence that it’s very useful. Knowing the label “appeal to authority” doesn’t tell you anything about when you want to believe in an authority when when you don’t.
The whole idea of teaching logical fallacies without providing the student any evidence that it’s useful to teach logical fallacies also very ironic.
In the real world scholarship is important for understanding a topic on a higher level.
Any professional educators able to comment on this?
I had the opposite experience, I never learnt about logical fallacies in school and was shocked it was never taught when I found the formal definitions later in life.
Is it worthwhile to teach about “Logical Fallacies?”
When in high school, one of my English classes had a unit on logical fallacies. Everyone was given a list of “logical fallacies” like “appeal to authority” and “slippery slope.” We had to do things like match examples with the names of the fallacies (which would almost always have multiple reasonable answers), and come up with examples of various fallacies.
At the time, I thought that this was a huge waste of time. My reasoning was that there were many more ways to give arguments incorrectly than to give arguments correctly, and that we should instead be teaching people what valid arguments are, and not to trust anything else. I have not really questioned this initial judgement until just now.
Now I am forming a new opinion on this question, and would like collect some opinions from Less Wrong.
I both disagree and agree with your high-school self.
Learning to recognize common failure modes, and developing a common language for talking about them with each other, is a relatively cost-effective way to improve the average validity of my arguments, in much the same way that reducing infant mortality is a relatively cost-effective way to increase average lifespan.
It doesn’t do anything to improve the validity of my most valid arguments, though. Depending on how reliably I reach that maximum and how high that maximum is, that might be OK. Or it might not, and an entirely different approach (like teaching what valid arguments are) might work better.
And the relationship between learning-to-X and teaching-X-in-high-school is of course a whole different problem.
All of that said, I’m curious: how would you go about teaching what a valid argument is, to a degree that “don’t trust anything else” is actually good advice to follow?
I do not know. To be honest, my high school self had a strong tendency to overestimate the rationality and learning potential of the general population.
That’s a very fair answer.
Do you have any sense of how you learned it? For my own part, I feel like I learned what a valid argument is, to the extent that I have learned it, almost entirely by a series of negative examples. For that matter, I’m not sure I can articulate what a valid argument is in non-question-begging terms… though to be fair, I haven’t sat down and tried for five minutes.
I do not know how I learned how to argue, but I do not think it has anything to do with negative examples.
For me, it seems similar to understanding what is a valid mathematical proof (one which in theory could be expanded to following the logical rules at each step) but where you are allowed to make observations and probabilistic reasoning, all of which came naturally to me. I do not feel like I ever had any inclination to use logical fallacies, and I feel like I am quick to recognize when arguments do not make sense.
This is in contrast with cognitive biases. I feel like I am very dependent on parts of our brain that have biases, I will not be able to get past them easily, and can learn to mitigate them by being aware of them.
On the other hand, some fallacies recur much, much more often.
Ad hominem and appeal to authority are de facto THE way humans argue with each other, with strawmanning as a very strong third. So at least learning how to spot and correct those can alone improve someone rationality.
There are though much funnier ways to learn about fallacies, Biased Pandemic being one of my favourite.
Biased Pandemic is about learning about cognitive biases. Cognitive biases are different from logical fallacies.
Yes, in the sense that cognitive bias are a subset of logical fallacies systematically applied by our brain. So I maybe can refine my answer: is it worth to teach about logical fallacies? Yes, especially when they become cognitive bias. A fun way to do this is Biased Pandemics.
Many cognitive biases don’t have much to do with logic as humans generally don’t make decisions via logic.
It’s worth learning about logical fallacies and internalizing them. It might not be worthwhile to teach people about them in school because people often don’t remember & internalize what they’re taught there.
While it’s important to be able to recognize & build a valid argument, it’s still been useful for me to use knowledge of fallacies to set mental triggers which activate when I mentally reach for a fallacy. Instead of unreflectively using an appeal to authority (for example) as a cognitive short-cut without checking whether it actually works, the not-quite-conscious sensation of making the appeal gets flagged for conscious attention, and I realize, “Oh! I’m making an appeal to authority. Does that appeal actually have much evidential weight?”
Edit: this isn’t directly responsive to you, but I can also imagine LWers who’ve moved on to noticing newer fallacies finding it harder to understand why it’s worth studying more well-known & canonical fallacies, even if the latter are as important.
There are “valley of bad rationality” effects here. Most so-called fallacies are in fact valuable heuristics. For instance, appeal to authority and slippery slope are both generally correct.
Knowing about fallacies creates in people the illusion that they can tell good arguments from bad. It lets them refute, in a superficially intelligent way, whatever position they don’t want to believe.
Though I do think it’s worth learning them anyway, for some people, at least so you know what people mean when they say “ad hominem”.
I don’t think that there evidence that it’s very useful. Knowing the label “appeal to authority” doesn’t tell you anything about when you want to believe in an authority when when you don’t.
The whole idea of teaching logical fallacies without providing the student any evidence that it’s useful to teach logical fallacies also very ironic. In the real world scholarship is important for understanding a topic on a higher level.
Any professional educators able to comment on this?
I had the opposite experience, I never learnt about logical fallacies in school and was shocked it was never taught when I found the formal definitions later in life.