in colour? It is, perhaps, even a harder problem which is raised when we ask whether there may not be a certain generic element of identity in all sense-contents.

It is a fair hypothesis, though not one which requires to be introduced into our psychological analysis, that such qualitative distinctions as we now find, and which appear irreducible, may be regarded as themselves products; and that therefore primitive consciousness, when that term is extended beyond the limits of the individual human mind, may present many fewer qualitative differences than we are now bound to enumerate as elementary components of the human mind. Some qualitative differences we must always include; but everything points in the direction of the hypothesis that the extremely marked differences we now discover in even the first stage of human consciousness are results.

Accordingly, an enumeration of kinds of qualitatively distinct contents, which we call sensations - not from any feature which they present, but because we connect their origin with stimulation of some part of the body-constitutes the first part of a description of primary consciousness. I say we call them sensations because we connect their origin with stimulation of some part of the organism. Evidently this criterion is wholly insufficient. There is not the smallest ground for supposing that other contents of mind, which we do not enumerate among sensations, are not connected with stimulation of the organism. In fact it is not this criterion which in practice we employ. We enumerate on the basis of a much less definite principle-that of the function which the contents have in the after-development of mind. We call those contents sensations which, as we discover from consideration of their later development, discharge the function of informing us of the qualities of what we call the objective world. No doubt, at first, the two principles were regarded

as having at least the same scope: the outer world meant the extra-organic world; and sensations were therefore described as those changes of consciousness which came about through stimulations that were extra-organic in their origin.

When we include among sensations those which arise from intra-organic stimulations, we are compelled to drop the one principle-the affection of the organism-and employ the other that of giving data for the apprehension of the objective. The first principle would not allow us to distinguish between what is called an internal sensation and an idea or emotion. We make the distinction on the ground that the one kind of content serves to inform us of the existence and qualities of that which is objective—the body -while the other does not.

It follows, therefore, that our enumeration of the sensecontents must not be regarded as resting upon any clear well-defined feature of their own content-a consideration which will be found of some importance when we proceed to deal with another content of consciousness which is apparently primitive, namely, feeling.

II. Feeling. From the point of view of the inner observer the components of primary consciousness can only be characterised by differences among the features or combination of features which they possess as directly given there. For this reason it seems impossible to regard as primitive and fundamental the distinction between what are ordinarily called sense-presentations and feelings, as though the first involved from the outset the mark of being apprehensions of objects, while the second were from the outset marked as states of the subject. Such a difference we must regard as derivative-dependent, therefore, on the features, or some combination of the features, which all the contents of consciousness offer in their primary appearance.

We should undoubtedly be entitled to accept as among these features any relation of dependence, if such be discoverable-any such relation, for example, as is implied in Herbart's view of feeling as a state arising out of and having reference to some conflict or harmony among given presentations. Now there does appear to be, in our developed experience, some kind of relation of this sort between feelings and the other components of consciousness. Extending this relation from the developed state to the primary consciousness, it has been supposed that feeling may be regarded as a secondary fact, conditioned by and (one might conjecture) dependent on the presence of other components in consciousness.

Against this undoubtedly there stands the fact of experience that bodily pain, if not bodily pleasure, seems quite primary in character, that its occurrence may indeed depend upon a physiological change, but does not seem to depend on the previous occurrence in consciousness of a definite sensepresentation. Whatever be the relation between bodily pain and pleasure and the more ideal forms of feeling, we are bound, from the psychological point of view, to regard them as varieties of the same kind; and if, therefore, at any one point of the series of feeling-experience, we can detect independence, we must, in opposition to the other theory, accept such independence as the fundamental mark, and consider the dependence, which undoubtedly is observable, as affecting the intensity and direction of the feeling-experience, which nevertheless possesses its own roots.

There can indeed be no doubt that even the dependence which we do observe is very far from being a simple relation. Wundt, whose theory of feeling is most obscure, seems at times to include, as one of the integral features of sense, the feeling-tone of the presentation, assigning to it, therefore, a place similar to that of quality, intensity, and

duration. But it is obvious that there is no simple relation between the content of a sense-presentation and the accompanying feeling. Normal or average relations there may be, and such relations are fairly intelligible; but it seems evident that the feeling-tone is not simply determined by any one feature, or by any special combination of the features, of the sense-presentation. Under different conditions the same presentation will yield the most diverse feeling-tones. The feelings, then, and by those at present we mean the pleasure-pain experiences, seem to be of independent nature; and, however intimate their connexion with the other components of consciousness, they seem capable of explanation only by reference to some independent process of an organic kind.

The question next arises, do the characteristics of pleasurepain exhaust the qualitative distinctions we discover in feeling? To this question Wundt for two reasons seems to offer a negative answer. On the one hand he seems to think that the feeling-tone accompanying any content of consciousness that has itself distinctness must also be regarded as qualitatively distinct. The feeling-tone of a simple note, for example, he would insist, is qualitatively distinct from the feeling-tone of a harmony. But he has to admit in respect to this that we have no means of describing this qualitative difference; and a qualitative difference which is devoid of all definable character seems hardly worth retaining. The truth is probably that in describing these experiences we underestimate their complexity, and that more distinct factors are involved than are satisfactorily named in our generalised terms-the sense-presentation and the accompanying pleasure-pain. It is exceedingly improbable that any sense-presentation occurs without giving rise to a general alteration in the organic processes, which, yielding in its turn elements of sense and feeling, colours the

total result. In the realm of sense what occurs is probably very similar to what we find in the more developed region of ideas, where the total effect of any idea in its passage through consciousness is dependent largely on the vague illdiscriminated suggestions to which it gives occasion.

A second ground which Wundt advances for recognising more than pleasure-pain, concerns, I think, not so much the immediate experience ordinarily called feeling, as certain total effects due to the manner in which sensations and

feelings pass through consciousness. "Every feeling," says Wundt, "in this passage through consciousness has a threefold significance. First, it indicates a definite modification. of the immediately present state; this, on the whole, coincides with the fundamental difference between pleasurable and painful. Secondly, it exercises a definite influence on the immediately subsequent condition; and this may be distinguished, according to its main directions, as stimulating or repressing. In the third place, it is in its own character determined by the immediately preceding condition; and this effect makes itself manifest, in the given feeling, in the forms of tension and relaxation." 1

I cannot admit that these second and third points indicate simple primary experiences which we are entitled to place on the same level with pleasure and pain. The descriptive terms applied to them are very general, and indicate not necessarily simple direct experiences, but the results of mediate comparison relative to, and conditioned by, direct sense-experiences. For example, there is no reason to doubt that, owing to the intimate correlation of the organic processes, an experience which is either pleasurable or painful may indirectly exercise an effect of the kind which we express by the generalised terms stimulative or repressive on

1 Wundt, Grundriss der Psychologie, § 7, par. 9; [tr. Judd, Outlines of Psychology (1897), pp. 84-5.]