The social proposition ‘other men exist’ and the pragmatist proposition ‘it is expedient to believe that other men exist’ come from different universes of discourse.  One can believe the second without being logically compelled to believe the first; one can believe the first without having ever heard of the second; or one can believe them both.  The first expresses the object of a belief, the second tells of one condition of the belief’s power to maintain itself.  There is no identity of any kind, save the term ‘other men’ which they contain in common, in the two propositions; and to treat them as mutually substitutable, or to insist that we shall do so, is to give up dealing with realities altogether.

Mr. Ralph Hawtrey, who seems also to serve under the banner of abstractionist logic, convicts us pragmatists of absurdity by arguments similar to Mr. Russell’s. [Footnote:  See The New Quarterly, for March, 1908.]

As a favor to us and for the sake of the argument, he abandons the word ‘true’ to our fury, allowing it to mean nothing but the fact that certain beliefs are expedient; and he uses the word ‘correctness’ (as Mr. Pratt uses the word ‘trueness’) to designate a fact, not about the belief, but about the belief’s object, namely that it is as the belief declares it.  ‘When therefore,’ he writes, ’I say it is correct to say that Caesar is dead, I mean “Caesar is dead.”  This must be regarded as the definition of correctness.’  And Mr. Hawtrey then goes on to demolish me by the conflict of the definitions.  What is ‘true’ for the pragmatist cannot be what is ‘correct,’ he says, ’for the definitions are not logically interchangeable; or if we interchange them, we reach the tautology: 

“Caesar is dead” means “it is expedient to believe that Caesar is dead.”  But what is it expedient to believe?  Why, “that Caesar is dead.”  A precious definition indeed of ‘Caesar is dead.’

Mr. Hawtrey’s conclusion would seem to be that the pragmatic definition of the truth of a belief in no way implies—­what?—­that the believer shall believe in his own belief’s deliverance?—­or that the pragmatist who is talking about him shall believe in that deliverance?  The two cases are quite different.  For the believer, Caesar must of course really exist; for the pragmatist critic he need not, for the pragmatic deliverance belongs, as I have just said, to another universe of discourse altogether.  When one argues by substituting definition for definition, one needs to stay in the same universe.

The great shifting of universes in this discussion occurs when we carry the word ‘truth’ from the subjective into the objective realm, applying it sometimes to a property of opinions, sometimes to the facts which the opinions assert.  A number of writers, as Mr. Russell himself, Mr. G. E. Moore, and others, favor the unlucky word ‘proposition,’ which seems expressly invented to foster this confusion, for they speak of truth as a property of ‘propositions.’  But in naming propositions it is almost impossible not to use the word ‘that.’