He

sense theory reinforces his theory of benevolence, it embarrasses his theory of virtue, and it does so all the more because he does not avail himself of the 'will' as the seat of virtue. seems once on the point of doing so (§ 442, note), but he was probably unwilling to involve his theory in the free-will controversy, and we for our part may be thankful that he did not. Since Kant the will has been freely referred to as the ultimate residence of virtue, but not always with profit

26. Is a moral sense necessary ?

In the moral sense theory the questions of the nature and subject-matter and motives of virtue are so mixed up that it is almost impossible to separate them, as Price would have us do (§ 586). It is therefore difficult, and would after all be rather artificial, to develop one's criticism of the theory in any very logical or consecutive way. But before coming to the discussion of desire, which is in some ways the most interesting part of the writings of this period, we may mention some miscellaneous criticisms of the moral sense theory.

The intellectualists of course denounce the moral sense theory not only as offensive but as gratuitous (§§ 538, 607). Butler does not commit himself (§ 244), but Adam Smith denounces it as contrary to the economy of nature (§ 347), and Gay says that it is at the best based on an argument ad ignorantiam, by which we should be as justified in asserting a 'pecuniary sense' as a moral sense (§§ 855, 883). As a matter of fact, Hutcheson displays a most alarming readiness to multiply senses (§§ 441-3), which finds its proper caricature in Kames' 'sense of property' (§ 948 f.). The real sting however of these criticisms lies in their counter-assertions of sympathy or association of ideas as explanations of the admitted phemomenon of 'immediate approbation.' To these we shall recur.

There is also certainly some ground for J. Clarke's assertion that what the theory gives with one hand it takes away with the other that it invents a sense to make virtue pleasant, and then says we must not pursue that pleasure (§ 806). Hutcheson, who had criticized Shaftesbury for allowing the virtuous man to have regard to the concomitant pleasure of benevolence (§ 470), is most careful to impress on us that our benevolence must be entirely disinterested if it is to be virtuous: the concomitant pleasure of benevolence must not and indeed cannot be the motive to benevolence. But he is not so clear about the pleasure of the moral sense. He of course asserts that approbation is itself disinterested and is not excited by desire to obtain the concomitant pleasure of approbation, but he does admit (§ 460) that 'the prospect of the pleasure of selfapprobation is often a motive to choose one action rather than another,' and he would presumably regard it as a proper motive in choosing to continue in the agreeable state' of benevolence (§ 131). In general, however, he runs a risk with his theory of disinterested desire of proving too much-viz. that all desire is disinterested, in which case disinterestedness is no longer the mark of virtuous desire; or that no thought of the pleasure of moral sense must enter into the mind of the virtuous person, in which case the moral sense is not very useful to virtue, but on the contrary frequently imperils its existence. It may also be noted that Hutcheson's limitation of the function of moral sense to the production of a peculiar pleasure opens the way to such an assimilation of this pleasure to other pleasure as Hume carried out through the medium of sympathy. His theory comes perilously near to saying that virtue is that which pleases us after a particular manner' (Hume, Treatise, p. 470).

27. Virtue as benevolence.

As for benevolence itself, the sentimentalists are quite sure that disinterested benevolence is the foundation and summary of virtue. The rigour of their altruism is, however, qualified by the admission that in considering the good of 'the system of rationals' a man is allowed to regard himself as a member of that system, and if the good resulting to others from a given act is not so great as the evil resulting to himself he may properly abstain from it for that reason (§§ 117-118, cf. §§ 133, 180). Benevolence itself, or regard for the good of a system, requires a man to be solicitous about himself, and to have special regard to his relations and friends.

It is quite clear here that something else is considered than the amount of benevolence implied in an act. It may perhaps be said that regard to the good of a wide system requires more benevolence than regard to the good of a narrow system, but when we are instructed to prefer the good of the higher to that of the lower system the appeal is evidently to other considerations than those of benevolence: the difficulty is, in fact, the same as arises for the hedonist over 'higher' and 'lower' pleasures (§ 479, cf. § 476).

The theory of benevolence, moreover, was founded on the assumption of what Butler calls 'the natural principle of attraction between man and man' (§ 207), or a benevolence, as Hutcheson says, 'in some degree extended to all mankind' (108). Hume had attacked the 'benevolent' theory by declaring that there was no such passion in human minds as the love of mankind merely as such, independent of personal qualities, of genius, or of relation to oneself' (Treatise, pp. 481-2), or, as Kames puts it, there is no such principle of general fondness of man to man by nature as there is in dogs towards

man' (§ 937). Such general benevolence as is displayed is said to be due to 'sympathy,' on the theory of which a good deal of the controversy turns. The benevolent theory was also attacked by the intellectualists as basing virtue upon instincts the operation of which is necessary and so devoid of merit (§§ 532-5). As against the selfish school and their instinct of self-love, Hutcheson is prepared to defend a 'benevolent universal instinct' (§ 131), but as a rule he prefers to emphasize against both criticisms the distinction between 'calm universal benevolence,' the product of reflection, and the particular benevolent affections (§ 442). This reflection upon 'all mankind or the system of rationals' turns out, however, to be only the reflection that by regard to them 'we may gratify either our self-love or kind affections in the fullest manner.' The good of the species appears to be hardly a possible object of affection, and the reflective love seems hardly disinterested (§§ 452-3). Thus the idea of the 'universal natural good of mankind' or 'the system of rationals' which in his earlier writings is distinctly constitutive' (§ 112) becomes attenuated into a very regulative principle in his later writings. And if we appeal to the moral sense we find that it often approves and disapproves without any regard to the good of any system (§ 480), and it turns out (probably under the influence of Butler) that 'the righteousness or goodness of actions is not the same notion with their tendency to universal happiness or flowing from the desire of it.'

Butler's treatment of benevolence is indeed of great importance in the history of moral philosophy: benevolence is disinterested indeed, but it is no more disinterested than any of the particular affections, every one of which 'rests in its object as an end' (§ 207). The love of our neighbour is as interested or disinterested as the love of anything else; there

is no peculiar contrariety between benevolence and self-love (§§ 233-4); disinterestedness is not the distinguishing mark of virtue, and 'benevolence and the want of it, simply considered, are in no sort the whole of virtue and vice' (§ 249, cf. § 532), though most of the common virtues and vices may be traced up to benevolence or the want of it (§ 242). Benevolence is for some purposes placed by Butler on the same level as the particular affections, though it is not therefore a blind propension, but is to be regarded as naturally allied with calculative reason (§ 240), but on a lower level than the two great reflective principles, self-love and conscience. Both of these combine to encourage benevolence to the greatest extent, though conscience certainly is influenced by other considerations than the amount of happiness produced, and more particularly by that of 'desert' (§ 244). The way is thus opened for a more liberal view of human nature and its 'perfection,' a conception which had been almost stifled by the weight of benevolence, and for other aspects of morality besides its hedonistic, though he is not afraid to admit that 'nothing can be of consequence to mankind or any other creature but happiness' (§ 241). Butler's theory is by no means free from confusion, but he gets rid of the confusions which grew so thick round the 'calm universal benevolence' of the sentimentalists, and also of that narrowness which is so apt to make the 'disinterested' theory merely uninteresting.

28. Sympathy.

Before we consider the significance of Butler's theory of desire it may be convenient to notice the two great principles which have been used to explain the admitted immediacy of moral approbation and the alleged disinterestedness of both