determine the elements, factors or constituents of a concrete state of consciousness or psychosis, as it has been termed: (2) Is there any definite cycle or order of succession among these, and how are they causally related? Having determined these points -more or less in course of so doing-it may become possible to attain to a more exact terminology.

Feeling.

§3. Keeping as much as may be to the first question, we are at once confronted by the doctrine that feeling alone is primordial and invariably present wherever there is consciousness at all. Every living creature, it is said, feels, though it may never do anything more: only the higher animals, and these only after a time, learn to discriminate and identify and to act with a purpose. This doctrine, as might be expected, derives its plausibility partly from the vagueness of the word 'feeling,' and partly from the intimate connexion that undoubtedly exists between feeling and cognition on the one hand and feeling and volition on the other. As to the meaning of the term, it is plain that further definition is requisite for a word that may denote (a) a touch, as feeling of roughness; (b) an organic sensation, as feeling of hunger; (c) an emotion, as feeling of anger; (d) any purely subjective state, as feeling of certainty or of activity; (e) the one subjective state that is purely 'affective,' as feeling of pleasure or pain. Since we find precisely the same variety of usage in the case of the equivalent German Gefühl and more or less of it in the case of the French sentiment, it may well be asked if there are no common traits connecting these various significations together. There seem to be three. Feeling in the last sense accompanies organic sensations and is present in emotions. Passivity, which renders passion almost a synonym for emotion, is but another aspect of feeling as affective and of sensation as given. Immediacy, the common mark of all subjective states, is applicable to sensations also and the more applicable the more their so-called 'feeling-tone' predominates and the less they have of any specific quale. In this respect the sensations of touch have, after organic sensations, the best title to the name feeling, and they are probably the first of all our specific sensations to be clearly differentiated from the general sensibility

or general feeling, as it is indifferently called'. But all three characteristics apply to, and exhaust the meaning of, feeling only in the last (e), which we may therefore call its strictest sense. In all the remaining meanings some of these characteristics are lacking while others beside are present. And feeling is taken in this sense, by those who maintain-with any show of plausibility-that all the more complex forms of experience are resolvable into, or at least have been developed from, feeling2.

The only proof of such position, since we cannot observe the beginnings of conscious life, consists of considerations such as the following. So far as we can judge, we find feeling everywhere; but, as we work downwards from higher to lower forms of life, the possible variety and the definiteness of sense-impressions

1 Cf. below, ch. v, § 3.

2 This doctrine was a natural reaction from the one-sided 'intellectualism' which culminated in the teaching of the Leibniz-Wolffians. A full and careful history of this movement is still a desideratum. It seems to have been fostered by-if it did not originate in-the 'sentimentalism' of Rousseau and the Romanticists. From the 'faith and feeling philosophy' of Herder and Jacobi it passed over to the psychology of Beneke and Fortlage, to be finally worked out with great ingenuity and thoroughness by A. Horwicz in his Psychologische Analysen auf physiologischer Grundlage (1872-8). And here the reaction is complete: a position is reached which is perhaps as indefensible as the opposite extreme that it was meant to supersede. But, in truth, Horwicz, who had to recognise sensation and movement as distinct in his 'physiological basis,' is nevertheless driven to admit that feeling and conation are inseparable on the psychological side. So likewise with his immediate predecessor, Fortlage. The main difference between them was that Fortlage, following Schopenhauer, began with conation (Trieb) and Horwicz, influenced rather by Wundt, began with feeling (Gefühl).

There is another doctrine to be mentioned here that can hardly be called even 'plausible' and which had a very different source: the doctrine already referred to as presentationism or sensationalism (ch. i, § 5: cf. also ch. iii, § 2). Where sensations are called feelings-as they sometimes are even now, and still oftener were in the past -there is a verbal resemblance between sensationalism and the doctrine just discussed. And, thanks to the ambiguity in their leading term, the two doctrines tend to merge, as, for example in the following:-"In the beginning there is...nothing beyond presentation which has two sides, sensation and pleasure and pain.... All is feeling in the sense, not of pleasure and pain, but of a whole given without relations, and given therefore as one with its own pleasure and pain" (F. H. Bradley, Mind, O.S. 1887, xii. p. 367). What Mr Bradley has said en passant of Horwicz's position (Mind, N.S. 1893, ii. p. 212) will doubtless be regarded by many as applicable to this-it does not seem worth discussing,' and it is questionable how far Mr Bradley would still uphold it or indeed ever meant what it seems to mean (cf. his article "On our Knowledge of Immediate Experience," Mind, 1909, pp. 40 ff.; Truth and Reality, 1914, ch. vi.).

Views more or less akin to the above were advocated by Spencer, Maudsley, Ribot, Münsterberg and Titchener. Cf. Villa, Contemporary Psychology, 1903.

both steadily diminish.

Moreover, we can directly observe in our own organic sensations—and these seem to come nearest to the whole content of primitive or infantile experience-an almost entire absence of any assignable quale. Finally, in our senseexperience generally, we find the element of feeling at a maximum in the lower senses and the cognitive element at a maximum in the higher. But the so-called intellectual senses are the most used, and use (we know) blunts feeling and favours intellection, as we see in chemists, who sort out the most filthy mixtures by smell and taste without discomfort. If, then, feeling predominates more and more as we approach the beginning of conscious life, may we not conclude that feeling is its only essential constituent? On the contrary, such a conclusion would be rash in the extreme. Two lines, eg., may get nearer and nearer and yet will never meet, if the rate of approach is simply proportional to the distance. A triangle may be diminished indefinitely, and yet we cannot infer that it becomes eventually all angles, though the angles get no less and the sides do. Before, then, we attempt to decide whether pleasure or pain alone can ever constitute a complete experience, it may be well to inquire into the connexion between feeling and cognition, on the one hand, and between feeling and conation on the other, so far as we can now observe them at the stage where all these are present—an inquiry which is tantamount to the second question raised above.

Broadly speaking, in many states of mind that we can now directly observe, what we find is (1) that we are aware of a certain change that has occurred either 'in things without or in our thoughts within,' (2) that we are pleased or pained by the change, and (3) that, being pleased or pained, we want and strive for the continuance of what pleases us, and still more urgently for the cessation of what pains us. But we never find that feeling directly alters-i.e. without the intervention of the action to which it prompts-either our sensations or our situation, but that regularly these latter with remarkable promptness and certainty alter it. We have not first a change of feeling, and then a change in our sensations, perceptions and ideas; but, these changing, change of feeling follows. In short, feeling appears to be an effect, which therefore cannot exist without its cause, though in different circumstances the same immediate cause may produce a different

amount or even a different state of feeling. Turning from what is often called the receptive phase of an experience to what is called the active or appetitive phase1, we find in like manner that feeling is certainly not-in such cases as we can clearly observe— the whole of what we experience at any moment. True, in common speech we talk of liking pleasure and disliking pain; but this is either tautology, equivalent to saying we are pleased when we are pleased and pained when we are pained; or else it is an allowable abbreviation, and means that we like pleasurable objects and dislike painful objects, as when we say we like feeling warm and dislike feeling hungry. But feeling warm or feeling hungry, we must remember, is not pure feeling in the stricter sense of the word. Within the limits of our observation, then, we find that feeling accompanies some more or less definite presentation which, on account of it, becomes the object of appetite or aversion; in other words, feeling implies a relation to a pleasurable or painful presentation or situation, that, as cause of feeling or as end of the action to which feeling prompts, is doubly distinguished from it. Thus the very facts that lead us to distinguish feeling from cognition and conation make against the hypothesis that consciousness can ever be all feeling.

But, as already said, the plausibility of this hypothesis is in good part due to a laxity in the use of terms. Most psychologists before Kant, and some even to the present day, speak of pleasure and pain as sensations. It is plain however that pleasure and pain are not ideas, as Locke called them, in the sense in which touches and tastes, colours and sounds, are—that is to say, they are never localised like the former or projected like the latter, nor are they elaborated in conjunction with other sensations and movements into percepts or intuitions of the external. This confusion of feeling with sensations is largely consequent on the use of one word pain both for certain organic sensations and for the purely subjective state of being displeased. Yet organic pains-which, of course, are subjectively displeasing -are not only always more or less definitely localised-and this of itself is so far cognition—but they may also be distinguished as shooting, burning, gnawing, &c., all which symptoms indicate

1 Though, strictly speaking, there is rarely or never in actual experience any such exclusive alternation. Cf. below, ch. v. fin.

a certain objective quality. Accordingly psychologists have been driven by one means or another to recognise two 'aspects' (Bain), or 'properties' (Wundt), in what they call a sensation, the one a 'sensible or intellectual' or 'qualitative,' the other an 'affective' or 'emotive,' aspect or property-the latter being also called the 'feeling-tone' (Gefühlston or Betonung) of the sensation. The term 'aspect' is figurative and obviously inaccurate; and to describe pleasure and pain, strictly understood, as 'properties' of sensation is a flagrant psychological barbarism.

The one point however which at present concerns us is simply that when feeling is said to be the primordial element in consciousness more is usually included under feeling than pure pleasure and pain, viz. some characteristic or quality by which one pleasurable or painful sensation is distinguishable from another. No doubt, as we go downwards in the chain of life the qualitative characteristics of the so-called sensations become steadily less and less definite; and at the same time organisms with welldeveloped sense-organs give place to others without any clearly differentiated organs at all. But we have no reason to suppose even the Amoeba itself to be affected in all respects the same whether by changes of temperature or of pressure or by changes in its internal fluids; albeit all of these changes will further or hinder its life and so presumably be in some sort pleasurable or painful. On the whole, therefore, there are grounds. for saying that the endeavour to represent all the various facts of consciousness as evolved out of feeling is due to a hasty striving after simplicity, and has been favoured by the ambiguity of the term feeling itself. If by feeling we mean a certain subjective state varying continuously in intensity and passing from time to time from its positive phase (pleasure) to its negative phase (pain) or vice versa, then this purely subjective state implies some agreeing or disagreeing object which psychologically determines it. If, on the other hand, we let feeling stand for both this state and that cause of it, then, perhaps, a succession of such 'feelings' may make up a consciousness; but in that case we are including two of our elementary facts under the name of one of them. The simplest form of psychical life, therefore, involves not only a subject feeling but a subject having qualitatively distinguishable objective presentations which are the causes of its feeling.