I agree with you that accuracy should be the goal, but Seligman doesn't. I read most of the book, he thinks that delusional optimism is good, because it will make you happier and more productive.

He's used his research to teach optimism to soldiers so that they are more effective. But perhaps it is appropriate for soldiers in the middle of a war to not feel super optimistic?

>If the premise is that the effects upon your past success are equally likely to be general or specific regardless of their sign

I don't know if that's a true premise. It's an empirical question. There's no fundamental reason why the universe should be even-handed.

I read a good portion of Learned Optimism a while back. I think it's a dangerous book to read for certain types of depressive personalities. Ironically, there's a reading of the book that reinforces pessimism.