I expect that if, in the first tell them phase, one states confident conclusions that are not justified until later in reading the writing, then, in expectation across humans, there will be documents whose justifications fail but which convince many people anyway.
Yes. This is one of the many reasons why it’s important to ensure that nothing stands in the way of astute readers writing comments like “actually this is wrong, dumb, and bullshit, and here’s why”. Effective criticism is the primary bulwark against convincing-but-wrong claims.
(EDIT: Indeed, one of the benefits of stating your thesis clearly, and more than once, is that it makes your ideas more easy to criticize, which is a good thing.)
I advise both my friends and AIs I talk to to build up claims without asserting until you’ve laid out the reasons to make the assertion. In my experience, this is difficult, but seems to me to be worth it for reduced rate of reasoning errors.
The major downside of that approach is not that it’s difficult, but that it makes for vastly less comprehensible writing. (Indeed, it seems to me that the effect of this is to increase the “rate of reasoning errors”, and that this more than outweighs any reduction thereof due to preventing “emotion-based updating”.)
As I’ve said before, the scarce resource is not ideas, but attention. If you write in a way that’s hard to follow, then people will simply not read what you write, or will read it but will not understand. That’s much worse than some people becoming convinced of your ideas despite them being wrong. The latter can be corrected via criticism. The former cannot.
Yes. This is one of the many reasons why it’s important to ensure that nothing stands in the way of astute readers writing comments like “actually this is wrong, dumb, and bullshit, and here’s why”. Effective criticism is the primary bulwark against convincing-but-wrong claims.
(EDIT: Indeed, one of the benefits of stating your thesis clearly, and more than once, is that it makes your ideas more easy to criticize, which is a good thing.)
The major downside of that approach is not that it’s difficult, but that it makes for vastly less comprehensible writing. (Indeed, it seems to me that the effect of this is to increase the “rate of reasoning errors”, and that this more than outweighs any reduction thereof due to preventing “emotion-based updating”.)
As I’ve said before, the scarce resource is not ideas, but attention. If you write in a way that’s hard to follow, then people will simply not read what you write, or will read it but will not understand. That’s much worse than some people becoming convinced of your ideas despite them being wrong. The latter can be corrected via criticism. The former cannot.