Whatever, then, may be the origin of the belief--'intuition' or 'association'--it is clear that it can give us no knowledge except such as is derived from sensations. Moreover, Brown is thus led, as in the doctrine of causation, to accept a really sceptical position. He declares that he is in this respect at one with both Reid and Hume. They both accept two propositions: first, that we cannot 'by mere reasoning' prove the existence of an external world; secondly, that it is 'absolutely impossible for us not to believe' in its existence. Hume, he says, pronounces the first proposition in a 'loud tone of voice' and 'whispers' the second. Reid, conversely, passes over the first rapidly and 'dwells on the second with a tone of confidence.'[488] Brown accepts both statements. He has already said that there is no argument against Berkeley's denial of matter any more than against the 'infinite divisibility of matter.' But he adds, it is 'physically impossible' for us to admit the conclusion, at least without 'an instant dissent from a momentary logical admission.'[489] This, indeed, is but a version of Hume's familiar statement that Berkeley's arguments admit of no reply and produce no conviction. Another essential doctrine of the Mills, the 'association' theory, is treated differently by Brown. Brown, as we have seen, both in his theory of causation and in his theory of our belief in an external world, speaks of principles in the mind which somehow override 'ratiocination.' In the first case, he speaks of 'intuition,' but in the other, as I have said, he seems to prefer association. The difference is remarkable because the belief in an external world is upon his showing simply a case of causation. It means essentially the reference of our sensations as to an external cause. Now, in the argument upon causation, he has insisted upon the insufficiency of association to generate the belief; and he would have found it difficult to meet his own arguments if applied to the belief in an external world. Yet it does not seem to occur to him that there is any difficulty in explaining this belief in an external world as a case of what Mill called 'indissoluble association.' Brown, as Mill thought, was not sufficiently aware of the power of this principle, and the difference between them is marked by this divergence. Brown had a great deal to say about association, though he chose generally to substitute the word 'suggestion,' previously familiar to Reid and Berkeley.[490] He considers it, however, mainly in another relation. He proposes to trace the order in which 'trains' of ideas succeed each other in our minds. He does not dwell upon the influence of association in producing belief. His question is not primarily as to the logic, but as to the actual succession of our thoughts. He explains that he uses the word 'suggestion' in order to avoid the hypothesis that the sequence of two ideas necessarily implies a