§1. Let, then, pleasure be defined as feeling which the sentient individual at the time of feeling it implicitly or explicitly apprehends to be desirable;---desirable, that is, when considered merely as feeling, and not in respect of its objective conditions or consequences, or of any facts that come directly within the cognisance and judgment of others besides the sentient individual. And let it be provisionally assumed that feelings generally can be compared from this point of view, with sufficient definiteness for practical purposes, and empirically known to be more or less pleasant in some definite degree. Then the empirical-reflective method of Egoistic Hedonism will be, to represent beforehand the different series of feelings that our knowledge of physical and psychical causes leads us to expect from the different lines of conduct that lie open to us; judge, which series, as thus represented, appears on the whole preferable, taking all probabilities into account; and adopt the corresponding line of conduct. It may be objected that the calculation is too complex for practice; since any complete forecast of the future would involve a vast number of contingencies of varying degrees of probability, and to calculate the Hedonistic value of each of these chances of feeling would be interminable. Still we may perhaps reduce the calculation within manageable limits, without serious loss of accuracy, by discarding all manifestly imprudent conduct, and neglecting the less probable and less important contingencies; as we do in some of the arts that have more definite ends, such as strategy and medicine. For if the general in ordering a march, or the physician in recommending a change of abode, took into consideration all the circumstances that were at all relevant to the end sought, their calculations would become impracticable; accordingly they confine themselves to the most important; and we may deal similarly with the Hedonistic art of life.
There are, however, objections urged against the Hedonistic method which go much deeper; and by some writers are pressed to the extreme of rejecting the method altogether. A careful examination of these objections seems to be the most convenient way of obtaining a clear view, both of the method itself and of the results that may reasonably be expected from it.
I should, however, point out that we are now only concerned with what may be called intrinsic objections to Egoistic Hedonism; arguments, that is, against the possibility of obtaining by it the results at which it aims. We are not now to consider whether it is reasonable for an individual to take his own happiness as his ultimate end; or how far the rules of action deduced from the adoption of this end, and from the actual conditions of the individual's existence, will coincide with current opinions as to what is right. These questions, according to the plan of my work, are postponed for future consideration: our sole concern at present is with objections tending to show the intrinsic impracticability of Hedonism as a rational method.
We are met, in the first place, by an objection which, if valid at all, must be admitted to be decisive. It has been affirmed by Green that ``pleasure as feeling, in distinction from its conditions that are not feelings, cannot be conceived''. If so, Rational Hedonism would certainly be impossible: but the proposition seems equally opposed to common sense, and to the universal assumption of empirical psychologists; who, in investigating elaborately and systematically the conditions, mental and physical, of pleasure and pain, necessarily assume that these feelings can be distinguished in thought from their ``conditions which are not feelings''. I also find that the writer himself from whom I have quoted, in a later treatise, conducts long arguments respecting pleasure which are only intelligible if the distinction between pleasure and its conditions is thoroughly grasped and steadily contemplated. Indeed he carries a distinction of this kind to an extreme point of subtlety; as be requires us to distinguish the ``self-satisfaction sought in all desire that amounts to will'' from the ``pleasure'' that ``there is in all self-satisfaction if attained'': whereas other moralists regard self-satisfaction as a species of pleasure. To maintain that we can distinguish pleasure from self-satisfaction and cannot distinguish it from its conditions, seems to me too violent a paradox to need refutation. It is possible that Green may only mean that pleasure cannot be thought to exist apart from conditions which are not feelings, and that it necessarily varies with any variation in its conditions. The statement thus interpreted I do not deny: but it is quite irrelevant to the question whether pleasure can be estimated separately from its conditions, or whether pleasures received under different conditions can be quantitatively compared. I cannot have the pleasure of witnessing a tragedy or the pleasure of witnessing a farce, without having along with either a complex of innumerable thoughts and images, very diverse in quality in the two cases: but this does not prevent me from deciding confidently whether the tragedy or the farce will afford me most pleasure on the whole.
I pass to another objection made by the same writer to the Hedonistic conception of the supreme end of action as ``the greatest possible sum of pleasures''. (It should be ``the greatest possible surplus of pleasure over pain'': but the difference is unimportant for the present argument.) The phrase, he says, is ``intrinsically unmeaning'': but his justification for this statement appears to be different in different treatises. At first he boldly affirmed that ``pleasant feelings are not quantities that can be added'', apparently because ``each is over before the other begins''. The latter statement, however, is equally true of the parts of time: but it would be obviously absurd to say that hours, days, years are ``not quantities that can be added''. Possibly this consideration occurred to Green before writing the Prolegomena to Ethics: at any rate in the latter treatise he admits that states ``of pleasant feeling'' can be added together in ``thought'', only denying that they can be added ``in enjoyment or imagination of enjoyment''. But this concedes all that is required for the Hedonistic valuation of future feelings; no Hedonist ever supposed that the happiness he aims at making as great as possible was something to be enjoyed all at once, or ever wanted to imagine it as so enjoyed. And unless the transiency of pleasure diminishes its pleasantness---a point which I will presently consider---I cannot see that the possibility of realising the Hedonistic end is at all affected by the necessity of realising it in successive parts. Green, in another passage, appears to lay down that ``an end'' which is ``to serve the purpose of a criterion'' must ``enable us to distinguish actions that bring men nearer to it from those which do not''. This, however, would only be the case if by an ``end'' is necessarily meant a goal or consummation, which, after gradually drawing nearer to it, we reach all at once: but this is not, I conceive, the sense in which the word is ordinarily understood by ethical writers: and certainly all that I mean by it is an object of rational aim---whether attained in successive parts or not---which is not sought as a means to the attainment of any ulterior object, but for itself. And so long as any one's prospective balance of pleasure over pain admits of being made greater or less by immediate action in one way or another, there seems no reason why `Maximum Happiness' should not provide as serviceable a criterion of conduct as any `chief good' capable of being possessed all at once, or in some way independent of the condition of time.