I shall assume, then, for the present that the conditions of some pleasure and some pain are not psychical,' and, leaving mere sensible feelings, shall examine those which attend psychical movements or dispositions. The two main conditions appear here to be harmony and expansion, and we are at once led to ask whether, as was the case with the intellect, these two characters will fall under a single head (MIND Xii. 376). I will begin the inquiry from the side of pain. There it seems to me that discord is the one constant feature. Mere loss, mere contraction of psychical existence, never pains us by itself. It does so only when some element feels itself thwarted or diminished, and for that we must have positive reaction and tension. If from the world which is dear to me you could isolate one fraction and extirpate it wholly, with all its memories and connexions, then I should never feel the loss of it. It is where the element with its connexions is left in part, and so reacts, that it becomes the seat of pain. Wherever we have pains whose origin does not seem physical, there we find a collision and a struggle of elements; and wherever we make a collision, which is not rapidly arranged or subordinated, there we can always find pain. It is true that pains and pleasures, not of psychical origin, may enter into and even occasion the tension, as when the idea or the remaining smart of a wound makes the tension of fear, or the removal of some dainty the struggle of disappointment. It is quite true that a collision often goes with pleasure on the whole, because the state, taken on the whole, is not a state of struggle, but contains the discord as an overpowered element. But it remains true that, so far as pain is not in its origin physical, it arises with tension, and that, wherever you have collision, you so far must have pain. And I cannot think that Prof. Bain is right in setting down surprise as a neutral state, even in the special sense which he assigns to the word "neutral" (MIND Xii. 577). It may of course be even pleasant, but if you take it as bare surprise, that is, apart from any supervening apperception and expansion, it seems certainly painful. If we then accept the result that the psychical origin of pain is tension, can we extend this view to cases where the origin seemed physical? It seems possible, first, that in such pain there is an unconscious psychical conflict, a collision of psychical states, an inroad and a resistance. But the objection is that, though possible, such a view lacks evidence.

1 The assertion of the opposite would in my opinion rest upon mere dogmatic preconception.

The existence of the unconscious struggling element would be a serious assumption and one not called for. It is far otherwise if we say that all pain comes from tension, either physical or psychical, and in the former case from the alteration and the resistance of a physical condition. So far as I know, this is a view which physiology can sanction, and, if so, pain in all cases may be set down to discord.

Passing now to the conditions of pleasure, we may expect to find the opposite. I do not say that we must find the exact counterpart of discord; but, if we did not, we should be discouraged. Pleasure, we said, seems to accompany both harmony and expansion, and there is a question whether both will fall under one head. Let us try first with expansion. There is no doubt that in general a mere increase of the psychical area seems pleasant, and no doubt again that as a recovery the increase is usually more pleasant. Can we say, then, that pleasure always comes from an expansion, or from a maintenance which against an opposite is really an increase? If so, harmony, as the removal or the overpowering of discord, would fall under expansion. To go so far as to call pleasure pain's mere negative would be an obvious absurdity. But, for all that, a precedent or a suppressed element might always be essential, and pleasure be dependent because a counter- if not a re-assertion. Or, if not that, yet perhaps the conditions, which would have gone on to pain, must be in part there for pleasure. Considered psychically, we may urge that every incoming sensation is at least to a certain extent an attack which necessitates reaction, and physiologically the stimulus required for the pleasant discharge may be taken as an invasion. Hence in both cases the positive will be really an expansion. In harmony too the variety is still negative against the unity. And in the mere "relativity" of pains and pleasures an unanswerable proof seems found, for there a pain, because a recovery, becomes actually a pleasure. Briefly then, if pain is felt hindrance, a pleasure is felt furtherance against defect or opposition, and in either case is expansion; or, if you prefer a modification, it is always a counter- or a re-assertion. Is such a view the correct one? I am not prepared to deny this, but, as before with activity, it is necessary to make assumptions quite beyond our real knowledge, or else to put an indefensible strain on the facts. It is not possible to find always a sense of defect as the condition of pleasant expansion. Again in some pleasures, e.g., of smell, it is often impossible to verify a tension. Further pleasures and pains are not wholly relative.

And lastly, if expansion per se were pleasant, then mere contraction should be painful, which assuredly it is not. I shall return to these objections, but will first attempt to state a view which, though not free from difficulty, seems to interpret the facts with the use of less force. If pain is discord, pleasure may be taken as the opposite. But, if the opposite is harmony, then harmony is ambiguous, since it may imply either the overpowering of collision or its simple absence. And if the latter is an improper use of the term, let us by all means drop the word harmony. Let us say, pleasure is the feeling which goes with presentation when that has not got the conditions of pain. A sensation is pleasant when not psychically or physiologically discordant. Pleasure thus will be the result of such positive conditions as imply the absence of pain. It will be the attendant either of all normal sensations, or of merely those where its (unknown) conditions of quantity or quality are present. The absence of hindrance does not constitute the pleasure, pleasure is essentially positive; but there would be no inconsistency in adding that its conditions must contain a variety-which is not painful but might become so if altered in character through quantity or quality.

Such a view seems to me to include all the facts, and it explains at once the pleasure of expansion. For, if mere positive sensation is per se pleasant, then more of the same will be naturally more pleasant. Expansion will make more of the same where the state was pleasure, or turn the scale where pleasure and pain were balanced. Just as increase of sugar is not sweet as mere increase but as presence of more sugar, so will it be with sensation; and expansion will be reduced to the head of mere position. And the same principle explains the pleasure of harmony proper. This enables us to have more of what already is pleasant, or to have without tension and with a balance of pleasure what would otherwise become painful. And we will now weigh our objection against the view that expansion per se is pleasant. We said that, if so, contraction should be painful per se, and per se it is not painful. It is painful only through repression and the tension of what is forced in. If we take the state where in full comfort a creature falls asleep, there its psychical area is progressively diminished, but it feels no beginning of pain. You may say that in repose other functions are set free, other sensations and also ideas come into being, while others are intensified, and that the change adds to the pleasure, so that in repose the soul might even be fuller than before. I admit all this, but if we keep to our crucial example of an

animal which normally falls asleep after a healthy meal, it is all too little for the purpose. With every possible admission, it remains still a monstrous paradox to say either that the psychical contents do not diminish or that the animal suffers pain. And, if so, contraction is not painful per se. It is painful where the removed survives still in idea, or is recalled and struggles for existence. And I admit further that this is usual. But where the idea fails, there contraction is not noticed. It becomes bare contraction and ceases at once to give pain. And I would press this strongly against the doctrine that expansion per se can be pleasant.

But our view that pleasure arises from unhindered position has to meet greater difficulties. We shall be told that pleasure is never pure, but that pain is of its essence, either as precedent or ingredient. I reply that, even were there no pure pleasure, yet the impurity might be external, like insoluble dirt in water. And secondly I deny the fact. I repeat the classical example of pleasures without want, where, if we keep to what is verifiable, we cannot find pain. And the fact again that in infancy (according to Preyer) pains come before pleasures seems to be without relevance, even if it were shown (as it is not shown) that in each case the special pain comes before the special pleasure. If we next go on to compare different levels of psychical life, we cannot find that the different balances of pleasure are merely in proportion to the contrasts which those lives contain. That a life monotonous, but without pain or care, could not at a low level be pleasant, seems to me a mere paradox, the offspring and again the parent of error. And I think such mistakes bring no real danger to our view. Its main difficulty arises from the pains of mere contrast and the pleasures of relief. It will be urged that one and the same state may be pleasant or painful because of its relations; that in pain, if I remember my yesterday's torment, my state becomes pleasant, and that my pleasure, if I think of all that I once hoped, may be turned to pain. How, if pleasure comes from what is positive, can these facts be explained?

The explanation is not easy, but still I think it is practicable, and there are three points to be considered. In the first place, the physical conditions may be so altered as to give an opposite result. In the second place, in the result we may have new positive sensations. In the third place, we must allow for the influence of ideas. I will apply these considerations to the case where the commonly painful is now pleasant. (1) In the first place, our physical state may be so changed that the conditions of pain are in consequence

not present. Where there is no discord there will be no pain. And this is true again psychically; for what was painful, because it jarred, may by the suppression of its antagonist have lost its painfulness. (2) In the second place, though the pain remains actually present, it may be overbalanced by new pleasure. In partial relief we may still have pain there: but its diminution has set free those normal sensations and ideas, both from a physical and psychical source, the conditions of which were suppressed before by the greater pain. The state may on the whole therefore be pleasant, and the fact that, if the pain were being increased rather than diminished, it would overbalance the pleasure though the amounts were the same, is on our view quite normal. For it is the newness of the pleasant sensations, as they rise, which directs attention upon them, so that they preponderate and depress the rest which is painful. And, if the pain were increasing, its novelty would for the same reason overbalance the pleasure.2

These principles will explain a large part of the facts, but they do not explain everything. For, if we take a case where the state of relief continues, it may cease to be pleasant. Habituation to pain has limits, and on the other hand our healthy sensations lose freshness and get feeble. We become depressed, and the balance of our state is pain. Yet even here, if we remember a worse pain behind, our state once more may be pleasure; which, if pleasure is really positive, seems inexplicable. (3) Here we must fall back on our third principle-the influence of ideas. The idea of pleasure is itself a positively pleasant fact, and can so turn the scale. I will explain this briefly. It should be a commonplace that ideas are psychical realities, and we cannot represent without using a psychical fact. Further, what represents a pain must be a pain, and so again with pleasure. It is not true that the idea of the greater pleasure or greater pain must itself be a stronger pleasure or pain (MIND ix. 289), but to think of a pain or a pleasure without in some degree feeling them is quite impossible. That is the first point. The second point is that, though pleasures and pains are not "relative," our ideas of them are largely so. We think of feeling as a series, a scale which rises or falls from agony to delight; and in this scale more or less of pain or pleasure stands for less or more of the

1

1 Volkmann, Lehrbuch der Psychologie, §§ 70, 71.

2 Why novelty attracts attention, whether from a psychical or only a physical cause, I cannot discuss.