become disinterested, self-conscious'. (Creative Evolution, p. 186.) This cleavage between Instinct and Intelligence has no parallel in Butler. Although not developed in Matter and Memory', the distinction cannot be without significance for the theory of memory.

When one asks how the memory which imagines is related to the memory which enacts the past, one is brought face to face with problems. In the first place it is not clear whether the records of the past belong to the realm of pure memory or to that of memory-images. In contrasting pure memory and memory-images M. Bergson states that it is pure memory which records the past, yet he often refers to the memory of the past as if the past survived in the form of memoryimages, e. g. :

"Though the whole series of our past images remains present within us, still the representation which is analogous to the present perception has to be chosen from among all possible representations.' (ibid., p. 114.) Consciousness. . . retains the image of the situations through which it has successively travelled, and lays them side by side in the order in which they took place.' (ibid., p. 96.)

Yet it is impossible to conceive how the past can survive in the form of memory-images. How can that which as an event in perception was in the things perceived rather than in the percipient, how can this survive as a memory-image? Again, what would become of the difference in kind between memory and perception, if the memory-image were in any literal sense a survival of the past perception. Such a suggestion seems foreign to the general tenor of M. Bergson's teaching.

Assuming then, the language of certain passages notwithstanding, that the record of the past survives only in pure memory, and that this is the work of the soul, we must conceive of a soul as present at the events of pure perception. And this soul must be the soul of the recording individual. In explanation of a memory which records the details of individual

history, it is not enough to claim the presence of that psychic principle in virtue of which there is life and action. When M. Bergson tells us that 'in perception we grasp, at one and the same time, a state of our consciousness and a reality independent of ourselves' (ibid., p. 270), is he recognizing the presence of an individual recording soul? If so, despite the stress laid upon perception as action, we are bound to regard it as being also, or rather as being in the first place, knowing. If not, how can we conceive of the events of pure perception as recorded by the individual?

This aspect of perception cannot be accounted for by memory-images, since it must itself be the only possible source for such images. What then are memory-images? M. Bergson tells us they are the result of an act sui generis, an act which changes the psychic tension.

Whenever we are trying to recover a recollection, to call up some period of our history, we become conscious of an act sui generis by which we detach ourselves from the present in order to replace ourselves, first in the past in general, then in a certain region of the past-a work of adjustment, something like the focussing of a camera. But our recollection still remains virtual; we simply prepare ourselves to receive it by adopting the appropriate attitude. Little by little it comes into view like a condensing cloud; from the virtual state it passes into the actual; and as its outlines become more distinct and its surface takes on colour, it tends to imitate perception.' (ibid., p. 171.)

Are, then, memory-images occasioned by this act sui generis? Are they akin to movement? Much of what M. Bergson says implies this, e. g. the passages describing the movements necessary to initiate the image (quoted p. 125); also the following:

The hearer places himself at once in the midst of the corresponding ideas, and then develops them into acoustic memories which go out to overlie the crude sounds perceived while fitting themselves into the motor diagram.' (ibid., p. 145.)

We are told that the memory-image is already partly sensation. It is difficult to know whether these memoryimages are by M. Bergson thought of as permanent entities or as entities which come into momentary being in consequence of the act. The figure of the cone suggests their continuous existence. The act sui generis would then occasion our consciousness of them, but not their existence. But, on the other hand, how can they be nascent sensations or in any sense partake of the nature of sensation if they have continuous existence? Where are we to draw the line which is to make memory different in kind from perception? Does it lie between memory and the memory-image or between the memory-image and sensation ? Apparently between memory and the memory-image, since we are told that the memory of a sensation is not a nascent sensation, while as we have seen the image is. But if we draw such a line at all, how can we hold that it is impossible to say where one begins and the other ends? (See ibid., p. 171.)

We saw that for connecting the various levels of his cone M. Bergson relied on the doctrine of psychic tension. Memory, memory-images, and perception represented different degrees of psychic tension. The problem is at bottom the problem of bridging the gulf between movement and knowledge, the one the affair of the body, the other the affair of the mind. M. Bergson would claim a fundamental identity between quantity and quality. Movement he maintains cannot be adequately expressed in terms of space, quantitatively. Movement is at bottom qualitative, for every movement has its specific character as a rhythm of life. In perceiving, the conscious individual catches up the rhythm of the perceived, transposing it into his own tempo. In place of the old opposition of matter and mind M. Bergson gives us a new one-life with its manifold rhythms and the individual soul with its degrees of tension. Will the fact that life is supra-consciousness make the gulf between the affairs of the body and the affairs of the soul narrower ? M. Bergson would fain show how essential

each is to each, but does this do more than fringe the brink?

Sometimes we seem to find in M. Bergson an opposition that threatens the very doctrine of perception itself; viz., an opposition between life and matter. For Butler the living and the dead, the organic and the inorganic world, formed an interdependent chain. For M. Bergson they sometimes appear as different, and even as opposing, forms of reality.

'Life is a movement, materiality is the inverse movement, and each of these two movements is simple, the matter which forms a world being an undivided flux, and undivided also the life that runs through it, cutting out in it living things all along its track. Of these two currents the second runs counter to the first, but the first obtains, all the same, something from the second.' (Creative Evolution, p. 263.)

If we take this passage at its surface value we can well understand M. Bergson's declaration,' The double form of consciousness' (Instinct and Intelligence) is then due to the double form of the real, and the theory of knowledge must be dependent on metaphysics.' (ibid., p. 188.) But if we are to recognize this duality in the real, we must reconsider the doctrine of pure perception. The activity of the world of change will include that of matter. It may be to this activity that we respond in perception. But if so, how does the activity of our living body consolidate itself with such action? We shall have a gulf to bridge between the life of a living body and the activity of dead matter. It may be argued that M. Bergson intends the opposition between life and matter to be one of direction only. In evidence of this is the description of matter as the extinguished fragments of the rocket of supraconsciousness which lights up organisms (cf. Creative Evolution, p. 275). But to accept this is to return to the previous position that all activity is the manifestation of some degree of life, and that all life is psychic in character.

The memory problem then remains. How is an advance to

be made from the memory which is continuity of life to the memory of an individual recording soul. Does M. Bergson's view of life furnish anything to explain the individual soul which in contracting the dance of life to its own span records that dance, the soul which in virtue of its tensions can function as pure memory or as memory-image? The present writer has failed to find any link between M. Bergson's pampsychism and his individualism.

We may grant that Butler wrestled and failed to make memory-images intelligible by a theory of ensouled vibrations, but where Butler gave us one mystery, M. Bergson presents us with three the impersonal memory of an ensouled body, the individual memory of a spirit mind, and the relation between them.

VI

RETENTIVENESS AS A PSYCHOLOGICAL
CONCEPTION

In the first chapter we asked whether the presentation of memory as a biological problem left the old memory problem of epistemological psychology untouched, whether it replaced it by substituting a wider for a narrower issue, or whether it put the old problem of the philosophers in a new light.

If the criticism of the first two chapters is well founded, it will enable us to rule out the second suggestion. The problem of memory as knowledge of the past cannot be resolved into problems of the lineage of organic processes. Is it then untouched by the teachings of biology? The discussion of New Realism and of the philosophy of M. Bergson will have shown that writers whose primary interest is in philosophy are nevertheless anxious to reconcile the teachings of philosophy with those of biology. Untouched