This book from 1930 or so (PDF) seems chiefly concerned with describing ways thinking can go wrong. The author does not seem to know much of Bayes, but his book appears largely sound. It ends with practical suggestions for dealing with various "dishonest tricks" -- though the list assumes these tricks come from someone other than the reader -- and an imaginary conversation between advocates of different political views who all exhibit flawed thinking. People who sympathize with one side or another can get practice taking apart a bad argument for that side.

Found via recommendation from Answers in Genes.