41

36

what seemed to be good reasoning wete goog reasoning, I reply that according to my description of the phenomena od reasoning, the only fact which the soundness of all reasoning and the truth of all human thought really depends is that a man's conjectures are somewhat better than purely random propositions.
The idea that the criticism of the criticism of reasoning involves some new reasoning overlooks the fact that the criticism is sustained by the original inference.
"Reasoning, says Hobbes, is computation", and although the criticism of the criticism of reasoning simply repeats the process, like adding up a column of figures a second time.
It is conceivable that a blunder should be repeated, but after the column has been added, say ten times, and always with the same result, the aeithmetician has no longer any discernible doubt to be quieted; and