Page:Mind (New Series) Volume 8.djvu/205

 THE NATUEE OF JUDGMENT. 191 nise that the propositions from which he is deducing are themselves necessary, and that there may therefore be other necessary propositions, with a like claim to certainty, not to be deduced from them. He therefore imagines himself to have exhausted the field of knowledge ; whereas in fact he has only shown certain logical connexions within that field. But it is not here proposed to dispute the truth of particular existential propositions ; and though, unlike Kant, we admit them to be merely assumed, we may be thankful that he has shown us what can be inferred from them. Moreover, Kant's distinction between space and time on the one hand, and the categories on the other, also retains its value, though we can no longer describe their general difference as he did. It seems rather to be this : That time alone is sufficient for some sort of experience, since it alone seems to be involved in the simplest kind of existential pro- position, e.g., " Pleasure exists " ; and that again time and space together will suffice to account for the possibility of other pieces of knowledge, without the use of the categories. It is necessary to make a fresh assumption of propositions such as even Hume recognised, and such as are universal in physical science, in order to find the principles of substance and accident and causality implied. In all such propositions time and space are presupposed as well, but these categories are not implied in every proposition involving time and space. The simplest existential propositions are then to be re- garded as necessary propositions of a peculiar sort. In one kind the necessary properties of time are involved ; in another those of space also. But though this fact, which Kant points out, is very important against empiricists, we cannot regard it with him as establishing the truth of geometry and of the corresponding propositions about time. For existential propositions which are false, as well as those which are true, involve the same propositions about space and time. No existential proposition of any sort seems dis- coverable, which might not thus be false ; not even the famous " cogito " is indubitable. We cannot, therefore, take the " possibility of experience," in any possible sense, as sufficient warrant for our knowledge of space and time ; and we must regard the truths of geometry as independently known for true, just in the same way as some existential propositions are so known. Similarly, those propositions which involve substance and attribute are not sufficient to establish the truth of the pro- positions thereby involved. The permanence of substance is