like to take “given,” that content is not real. And any attempt, either to deny this, or to maintain that in the given there is never discrepancy, may be left to itself. But I will go on to consider the same view as it wears a more plausible form. “We do not,” I may be told, “add or take away predicates simply at our pleasure. We do not, so long as this arbitrary result does not visibly contradict itself, consider it true.” And I have not said that we should do this.
Outside known truth and error we may, of course, have simple ignorance.[1] An assertion, that is, must in every case be right or be wrong; but, for us and for the present, it may not yet be either. Still, on the other hand, we do know that, if the statement is an error, it will be so because its content collides internally. “But this” (an objector may reply) “is really not the case. Take the statement that at a certain time an event did, or did not, happen. This would be erroneous because of disagreement with fact, and not always because it is inconsistent with itself.” Still I must insist that we have some further reason for condemning this want of correspondence with fact. For why, apart from such a reason, should either we or the fact make an objection to this defect? Suppose that when William has been hung, I assert that it was John. My assertion will then be false, because the reality does not admit of both events, and because William is certain. And if so, then after all my error surely will consist in giving to the real a self-discrepant content. For otherwise, when John is suggested, I could not reject the idea. I could only say that certainly it was William, and might also, for all that I knew, be John too. But in our actual practice we proceed thus: since “both John and William” forms a discordant content, that statement is in
- ↑ For further explanation, see Chapter xxvii.
