The fact of illusion is not the tenet of any particular philosophy, nor even of philosophy itself; it is a recognised factor in common life and in physical science, but in instancing the theory of Bergson's philosophy I am choosing an extreme case. Berkeley held that illusion is practically universal; Kant taught that the apparent objectivity of phenomena is the form that the understanding imposes on things; but Bergson teaches {74} not only that all material reality is illusion, but also that this very illusion is the work of the intellect, that the intellect is formed for this purpose, intellect and matter being correlative, evolving _pari passu_. To such a doctrine there is of necessity a positive side, for it is impossible that it can rest on universal scepticism--scepticism both of knowledge and of the instrument of knowledge. If the intellectual view of reality as solid matter in absolute space is illusion, it must be possible to apprehend the reality from which the judgment that it is illusion is derived. If the intellect distorts, there must be an intuition which is pure, and the relation between these will be the relation between reality and appearance. Neither, then, is reality truth, nor appearance error. There is a truth of appearance, a truth that is a value in itself, a truth that is more than the mere negation that appearance is not reality. The appearance is our hold upon the reality, our actual contact with it, the mode and direction of our action upon it. What, then, is error? It cannot consist in the fact that we know appearance only, not reality, for we can only know reality by its appearance. It cannot be an appearance behind which there is no reality, for non-being cannot appear. It cannot be nothing at all or pure non-being, for to think of absolute nothing is not to think. In error there is some object of thought which is denied real being. What this is is the problem of error. CHAPTER VIII THE PROBLEM OF ERROR In the _Theætetus_ of Plato, Socrates has been discussing with Theætetus what knowledge is, and when at last agreement seems to be reached in the definition that {75} knowledge is true opinion, a new difficulty occurs to Socrates: "There is a point which often troubles me and is a great perplexity to me both in regard to myself and to others. I cannot make out the nature or origin of the mental experience to which I refer. How there can be false opinion--that difficulty still troubles the eye of my mind. Do we not speak of false opinion, and say that one man holds a false and another a true opinion, as though there were some natural distinction between them? All things and everything are either known or not known. He who knows, cannot but know; and he who does not know, cannot know.... Where, then, is false opinion? For if all things are either known or unknown, there can be no opinion which is not comprehended under this alternative, and so false opinion is excluded."