hypothesis without affecting its essential truth is the extent of its application. The general tendency of the human mind is to exaggerate the extent to which a newly discovered principle is applicable to reality. Subsequent progress consists in a more careful adjustment of the limits within which it holds true. The history of science and philosophy offers numberless illustrations. The rise of mathematical science, and the triumphs of its principles in the fields of number and space, led the Pythagoreans to apply them to explain the nature of reality in general. At a later date the physical method produced the atomic philosophy. In our own time the discovery of the close connexion between physiological and psychological processes has been made the ground of a theory which reduces thought to a secretion of the brain. Subsequent reflection has in each case sought to limit the application of these principles, but the limitations introduced have left the validity of the principles within their own field untouched.

So with conceptions which are recognised as more strictly hypothetical in their origin. The Darwinian theory again offers an illustration. There are two

main respects in which attempts have been made to limit its application. From the side of science grave question has arisen as to how far the principle of inheritance can be carried. Does it extend to traits of character which primarily owe their origin to habit, as Darwin seems to have supposed? Or must we limit it, with Weismann, to organic variations? Whatever the conclusion of this controversy may be, the validity of the Darwinian hypothesis as a whole is unaffected. Inherited variation is still a vera causa of divergence of species, though its action may be shown to be

crossed and complicated to a greater extent, or, again, to be simpler and more direct in its operation than was at first supposed. Secondly, from the side of philosophy, the claim put forward by professed Darwinians to have offered a satisfactory explanation of the intellectual and moral life has been questioned, and rightly questioned, by more recent thinkers, on grounds that do not concern us here. This limitation, however, again refers only to attempts to apply it beyond its legitimate sphere. While limiting in one respect, philosophy may be said to have confirmed it in another by showing that it possesses the best guarantee of truth in having made the origin of natural species only another illustration of the principle of evolution previously recognised in the history of human civilisation and the human mind (e.g. by Hegel, Comte, and Spencer), and thus brought the theory of it into line with conclusions already established in other fields.

IV.

IS THE KNOWLEDGE OF SPACE

A PRIORI?

THE subject was suggested to me by the casual re

mark of a friend who was himself a Neo-Kantian, and an intelligent person to boot, to the effect that, in view of the advances in recent psychology, the Kantian doctrine of Space was no longer tenable. It came as a surprise to me, as it had never occurred to me that they had anything to do with that doctrine as now understood. But as it seemed to be possible for one even of the initiated to hold this view, I thought it might be worth while to try to re-state the Kantian doctrine in connexion with the most recent results of psychology.

What is meant by Knowledge of Space? There are three senses in which we may understand the phrase:—

1. The presentation of extensity, which, so soon as it becomes an element in anything that can be called "experience," is the presentation of a world whose parts are outside one another.

2. Knowledge of the definite Space relations: position, magnitude, form, distance within that world. 3. Knowledge of Space "as a whole "-the abstract idea of Space as a form of unity in our experience.

As these upon any view must be regarded as stages in the development of complete spatial knowledge, the

controversy on hand must clearly concern itself in the first instance with the first.

What is meant by a priori? The doctrine that the knowledge of Space is a priori may be stated in two forms: (1) There is the older form, of which Schopenhauer's statement as quoted by Professor James1 may be taken as a type. According to this, space is an "innate a priori, anti-experiential form," which somehow "lies already performed in the intellect or brains," and within which we construct the physical universe. (2) It hardly needs to be explained that this is not the sense in which the Kantian doctrine is now held. There is no question of a fully developed preexperiential form into which the mind fits an empirically given non-spatial matter. What is contended is that in our earliest experience of spatiality there is an element which is not given in the mere sensation (if we can speak of such a mere sensation at all), but is given with it. It is a matter of nomenclature whether we call this a "form," or an elementary act of judgment. The point is that it is there as the rudiment of our later spatial judgments. Before going on to oppose this doctrine to the most recent statement of the Sensationalist position, I shall notice shortly the older attempt to assign an a posteriori origin to our knowledge of Space.

The older doctrine started from the assumption that spatiality is not immediately given in sense-experience. It must therefore be shown to be "evolved," or chemically compounded," out of the immediate data of sense. The most distinguished representative of this view is, of course, Mr. Herbert Spencer. His view may be found in his Principles of Psychology, 1 Principles of Psychology, vol. ii. p. 273.

chapters xiii., xiv., xxii., and (re-stated in reply to criticism) in Mind, of July, 1890.

The data from which the knowledge of co-existing positions, which is the kernel of our knowledge of spatial relations and of Space in general, is evolved, are these (1) Disparate, tactual, or visual sensations, as at the two corners of a book upon the table. (2) A series of reversible muscular sensations as the organ moves from one to the other. (3) The simultaneous (or what amounts to the simultaneous) presentation of the disparate sensations, as when our hands touch both corners of the book at once. The process by which out of these data the knowledge of co-existence is generated is the process by which the simultaneous presentation, being a "quasi-single state of consciousness," comes in time to be taken as the equivalent of the series of states, and to be habitually thought of in place of that which it symbolises.

Now if this is all that Sensationalism has to offer by way of explaining the origin of our knowledge of Space, it is not surprising that J. S. Mill, the candid friend of British Empiricism, should feel himself constrained to admit that "the idea of space is at bottom one of time." Mr. Spencer is not so easily daunted. He has scientific analogies at command to show how by a gradual transition we may pass from one form of reality to another which is qualitatively different from it. It is no more difficult to conceive of the idea of space emerging from the idea of a time series than to conceive of a circle becoming a straight line by gradually prolonging one of the axes, the molar motion of the stroke of a hammer passing as molecular motion into heat, or the same uniform dermal tissue transforming itself in the process of evolution into hair and nerve, tooth and eye.