@Robin: Thank you. Somehow I missed that post, and it was exactly what I was looking for.
@Vladimir Nesov: I agree with everything you said except for your statement that fiction is a valid argument, and your supporting analogy to mathematical proof.
Maybe the problem is the two different meanings of “valid argument”. First, the formal meaning where a valid argument is one in which premises are arranged correctly to prove a conclusion eg mathematical proofs and Aristotelian syllogisms. Well-crafted policy arguments, cost-benefit analyses, and statistical arguments linked to empirical studies probably also unpack into this category.
And then the colloquial meaning in which “valid argument” just means the same as “good point”, eg “Senator Brown was implicated in a scandal” is a “valid argument” against voting for Senator Brown. You can’t make a decision based on that fact alone, but you can include it in a broader decision-making process.
The problem with the second definition is that it makes “Slavery increases cotton production” a valid argument for slavery, which invites confusion. I’d rather say that the statement about cotton production is a “good point” (even better: “truthful point”) and then call the cost-benefit analysis where you eventually decide “increased cotton production isn’t worth the suffering, and therefore slavery is wrong” a “valid argument”.
I can’t really tell from the original post in which way Eliezer is using “valid argument”. I assumed the first way, because he uses the phrase “valid form of argument” a few times. But re-reading the post, maybe I was premature. But here’s my opinion:
Fiction isn’t the first type of valid argument because there are no stated premises, no stated conclusion, and no formal structure. Or, to put it another way, on what grounds could you claim that a work of fiction was an invalid argument?
Fiction can convincingly express the second type of valid argument (good point), and this is how I think of Uncle Tom’s Cabin. “Slavery is bad because slaves suffer” is a good point against slavery, and Uncle Tom’s Cabin is just a very emotionally intense way of making this point that is more useful than simple assertion would be for all the reasons previously mentioned.
My complaint in my original post is that fiction tends to focus the mind on a single good point with such emotional intensity that it can completely skew the rest of the cost-benefit analysis. For example, the hypothetical sweatshop book completely focuses the mind on the good point that people can suffer terribly while working in a sweatshop. Anyone who reads the sweatshop book is in danger of having this one point become so salient that it makes a “valid argument” of the first, more formal type much more difficult.
@Robin: Thank you. Somehow I missed that post, and it was exactly what I was looking for.
@Vladimir Nesov: I agree with everything you said except for your statement that fiction is a valid argument, and your supporting analogy to mathematical proof.
Maybe the problem is the two different meanings of “valid argument”. First, the formal meaning where a valid argument is one in which premises are arranged correctly to prove a conclusion eg mathematical proofs and Aristotelian syllogisms. Well-crafted policy arguments, cost-benefit analyses, and statistical arguments linked to empirical studies probably also unpack into this category.
And then the colloquial meaning in which “valid argument” just means the same as “good point”, eg “Senator Brown was implicated in a scandal” is a “valid argument” against voting for Senator Brown. You can’t make a decision based on that fact alone, but you can include it in a broader decision-making process.
The problem with the second definition is that it makes “Slavery increases cotton production” a valid argument for slavery, which invites confusion. I’d rather say that the statement about cotton production is a “good point” (even better: “truthful point”) and then call the cost-benefit analysis where you eventually decide “increased cotton production isn’t worth the suffering, and therefore slavery is wrong” a “valid argument”.
I can’t really tell from the original post in which way Eliezer is using “valid argument”. I assumed the first way, because he uses the phrase “valid form of argument” a few times. But re-reading the post, maybe I was premature. But here’s my opinion:
Fiction isn’t the first type of valid argument because there are no stated premises, no stated conclusion, and no formal structure. Or, to put it another way, on what grounds could you claim that a work of fiction was an invalid argument?
Fiction can convincingly express the second type of valid argument (good point), and this is how I think of Uncle Tom’s Cabin. “Slavery is bad because slaves suffer” is a good point against slavery, and Uncle Tom’s Cabin is just a very emotionally intense way of making this point that is more useful than simple assertion would be for all the reasons previously mentioned.
My complaint in my original post is that fiction tends to focus the mind on a single good point with such emotional intensity that it can completely skew the rest of the cost-benefit analysis. For example, the hypothetical sweatshop book completely focuses the mind on the good point that people can suffer terribly while working in a sweatshop. Anyone who reads the sweatshop book is in danger of having this one point become so salient that it makes a “valid argument” of the first, more formal type much more difficult.