327. But when we reach this point we have begun (as I say) to transcend the special province to which, in Chapter I (Introduction. Human Personality And Its Survival Of Bodily Death)., I assigned the title of genius. I there pointed out that the influence of the subliminal on the supraliminal might conveniently be divided under three main heads. When the subliminal mentation co-operates with and supplements the supraliminal, without changing the apparent phase of personality, we have genius. When subliminal operations change the apparent phase of personality from the state of waking in the direction of trance, we have hypnotism. When the subliminal mentation forces itself up through the supraliminal, without amalgamation, as in crystal-vision, automatic writing, etc, we have sensory or motor automatism. In accordance with this definition, the content of the inspirations of genius is supposed to be of the same general type as the content of ordinary thought. We have regarded genius as crystallising fluid ideas; or, if you will, as concentrating and throwing upwards in its clear fountain a maze of subterranean streams.

But we have not regarded it as modifying, in such operation, the ordinary alert wakefulness of the thinker, nor as providing him with any fresh knowledge, obtainable by supernormal methods alone.

1 More definite than most of the descriptions of this type in musical literature are the following words of Schumann (Robert Schumann's Early Letters, p. 268): " The piano is getting too limited for me. It is most extraordinary how I write almost everything in canon, and then only detect the imitation afterwards, - and often find inversions, rhythms in contrary motion, etc." And again, p. 271: "I do not realise all this while I am composing; it only comes to me afterwards; you who are at the top of the tree will understand what I mean".

It is plain, however, that such distinctions as those which I have drawn between genius, trance, automatism, cannot possibly be rigid or absolute. They are distinctions made for convenience between different phases of what must really be a continuous process - namely, the influence of the Self below the threshold upon the Self above it. Between each of these definite phases all kinds of connections and intermediate stages must surely exist.

Connections between trance and automatism, indeed, are obvious enough. The difficulty has rather lain in their clear separation. Trance, when habitual, is pretty sure to lead to automatic speech or writing. Automatism, when prolonged, is similarly apt to induce a state of trance.

The links between Genius and these cognate states are of a less conspicuous kind. They do, however, exist in such variety as to confirm in marked fashion the analogies suggested above.