Here, of course, we find ourselves face to face with a familiar objection to those philosophies of perception which deny that we have any access to external reality, except through ideas which are its copy. But they are in a better case than science. They need not explicitly admit a discrepancy between their premises and their conclusions. They arrive at the subjectivity of perception by methods of introspection. They interrogate consciousness, and are convinced that every experience can be analysed into sensations and ideas, some of which, no doubt, suggest externality, but none of which are external. If, then, the worst comes to the worst, they can, and often do, lighten their philosophic ship by pitching the whole material universe overboard as a bit of superfluous cargo. But physical science cannot (at least in my opinion) do anything of the kind. Its whole business is with the material universe. Its premises are experiences of external things, not of internal sensation and ideas. And if it has associated its fortunes with a theory of perception which treats experience as a natural effect of the thing experienced; if it has thereby wandered within sight of the perilous problems which haunt the frontier where mind and matter meet, it has not done so in a spirit of reckless adventure, but in the legitimate pursuit of its own affairs. This does not necessarily make things easier. We are not here concerned with questionings about the remoter provinces of knowledge—provinces unexplored except by specialists, negligible by ordinary men engaged on ordinary business. On the contrary, the difficulties to which I have called your attention threaten the unquestioned assumption of daily life, the presupposition of every scientific experiment, and the meaning of every scientific generalisation. They cannot be ignored. On the other hand, threaten as they may, these difficulties can never modify our attitude either towards practical action or scientific theory. Beliefs which were inevitable before remain inevitable still. The supreme act of instinctive faith involved in the perception of external objects stands quite unshaken. Whatever we may think of Berkeley, we cannot give up Dr. Johnson. “Seeing,” says the proverb, “is believing”; and it speaks better than it knows. VI Can we, then, adopt a middle course, and, imitating the serene acquiescence of Hume, accept the position of sceptics in the study and believers in the market-place? This seems eminently unsatisfactory; and, since believers on this subject we must perforce remain, it behoves us to consider how, and on what terms, we can best qualify our scepticism.