essence consists in banishing mystery from the origin of our moral instincts. If it too easily degenerated into an assertion of the absolute selfishness of human nature, the assertion that the moral sense is derivative was a necessary preliminary to all fruitful investigation of the phenomena.

87. The doctrine, scientific in spirit if crude in form, is supported by the scientific method of an appeal to experience. Locke insists upon the variability of the moral standard in different races and ages. The Tououpinambos,' for example, thought that they would merit paradise by revenge, and by eating their enemies. They have not so much as a name for God, and have no religion and no worship,' and these peculiarities of the Tououpinambos may be paralleled by equally strange aberrations of the moral instinct in other races. Now, though a special breach of the law may be no proof that it is unknown, a general permission to break it is a proof that it is not innate. The very recognition of any duty implies the presence in the mind of ideas of God, law, obligation, punishment, and a future life; and these ideas, so far from being universal, are not always clear and distinct, even in 'thinking and studious men.'2 The vast diversity of opinion which exists would be impossible if threats of Almighty vengeance were stamped in indelible characters upon the minds of all men ; nor can anyone, in fact, tell us what are these innate practical principles' which are yet asserted to be so palpably evident. Lord Herbert's five principles, for example, are illusory. To say that repentance for sin is a duty is idle, unless you are agreed as to the particular actions which are sinful. And the attempt to evade the appeal to experience by arguing that the innate principles are dulled by education and custom is really a mode of begging the question. The argument comes simply to this; 'the principles which all men allow for true are innate; those that men of right reason admit are the principles allowed by all mankind; we, and those of our mind are men of reason; wherefore, we agreeing, our principles are innate-which is a very pretty way of arguing and a short cut to infallibility.' The real fact is, that men, having taken up many principles on trust, and 1 Locke's Essay, book i. ch. iii. sec. 9. 2 Ib. sec. 12.

8 Ib. sec. 20.

having entirely forgotten whence they came, assume them to be divinely implanted axioms; and thus doctrines that have been derived from no better original than the superstition of a nurse and the authority of an old woman may, by length of time and consent of neighbours, grow up to the dignity of principles in morality and religion.'1

88. Locke brings down his logical sledgehammer on the principles of his antagonist with masculine vigour. If his objections are crudely stated, the dogmas which he smashes. were at least equally crude. But it must be granted that he has left little behind him but ruins. We ask, in some alarm, what then is morality? The conscience as a mysterious and independent guide is annihilated; the only motive left is selfinterest; and it almost seems as if the Tououpinambos had almost as much to say for themselves as the English or the Jews. Mandeville, indeed, in his denial of the real existence of virtue has simply carried Locke's method one step further. Assume that the standard of virtue is so variable that no single duty can be singled out as universally binding and recognised, and it is easy to infer that virtue is a mere sham.

2

89. No conclusion, of course, could be more repulsive to Locke himself, and it is curious that he did not perceive the application which might be made of his doctrines. Bending his whole energy to destroy the belief in the autocratic and irresponsible character of conscience, he never thinks of supplying its place. Apparently the need of reconstruction scarcely occurred to him. He speaks of the 'eternal and unalterable nature of right and wrong,' and he asserts emphatically that 'morality is capable of demonstration as well as mathematics.'3 If ethics are mathematically demonstrable, it must be possible to form a code applicable alike to Tououpinambos and Englishmen, or, at least, to assign some fixed principles from which the varying codes might be constructed. Locke would partly answer by referring to the will of God. The discussion is given in the twenty-eighth chapter of his second book,

1 Locke, book i. ch. iii. sec. 22.

2 Book ii. ch. xxviii., note to sec. II.

Ib. book iii. ch. xi. sec. 16; book iv. ch. iv. sec. 7; book iv. ch. xii. sec. 8; and see Reasonableness of Christianity,' Works, vi. 146.

under the head of 'Moral Relations.' He there defines moral good and evil to be the conformity to 'some law,' whereby good and evil are drawn upon us by the will and power of the lawmaker. We are subject to three kinds of laws, the law of God, the civil law, and the law of opinion or reputation.' 2 The law of God is enforced by the pains and penalties of the next world. Nobody can take us out of his hands. His will is the only true touchstone of moral rectitude; and by comparing them to his law it is that men judge of the most considerable moral good or evil of their actions; that is, whether as duties or sins they are likely to procure them happiness or misery from the hands of the Almighty.'3 The civil law determines men's criminality or innocence, and the 'philosophical law,' or law of opinion, varying widely in different countries, determines their virtue or vice. This, though wanting in precision, is the law by which men most frequently govern themselves; for its sanctions, vaguer than those of other laws, are more continually present to the imagination than those of the divine law, and less easily evaded than those of the civil law. These various laws may, of course, conflict as in the case of duelling, which is a sin tried by the law of God, a virtuous action by the 'law of fashion-another synonym for the law of opinion-and a capital crime according to the civil law of some countries. "

4

90. The law of God, then, is the only permanent and invariable standard; for the other laws vary-and, so far as Locke expounds his theory-vary indefinitely according to time, place, and circumstance. The law of God, too, must override the other laws in case of conflict; or, in his own language, be the only true touchstone of moral rectitude.' How, then, is the all-important question, can this law be discovered? If God's will be concealed in impenetrable mystery, virtue would apparently become a mere arbitrary fashion. That is Mandeville's solution. If the divine will be discoverable only by revelation, Locke's theory coincides with that of the theological utilitarians. The motive is with him, as with Paley, the dread of hell and the hope of heaven. himself that the Gospel gives an absolutely pure

1 Locke, book ii. ch. xxviii. sec. 5.

2 Ib. sec. 7.

He tells us code of mo

Ib. sec. 8.

4 Ib. sec. 15.

rality, and for that reason he excuses himself to Molyneux for not undertaking to write a treatise on the subject.' Locke, however, would not have admitted that our knowledge of morality was dependent on revelation. In fact, the whole argument of the treatise on the Reasonableness of Christianity' implies that the heathen philosophers could discover a system approximating very closely to that directly promulgated from heaven. How, then, could they arrive at a knowledge of the divine law? What was the criterion by which they were to distinguish between moral good and evil?

91. The curious vacillation which runs through Locke's reasoning upon morality, and which thus makes moral truth alternately quite uncertain and mathematically demonstrable, is but one instance of the general inconsistency in his theory.. of reality. According to Locke, as I have elsewhere observed, our knowledge of the external world cannot be scientifical,' We can only know phenomena, and know that they do not correspond (except in the case of the 'primary' qualities) to the objective facts beneath them. Certainty is attained only by comparison of ideas. We may know them adequately, for they exist entirely in our minds. Hence we may obtain certainty in mathematics; we have only to compare our ideas in order to discover geometrical relations, and we know (it matters not how) that those ideas are the counterparts of external realities. The same, according to Locke, may be said of moral relations. Though he expresses himself very indistinctly, his notion seems to be that in moral questions we are reasoning about certain things of which we know 'the precise real essence,' because they are entirely 'ideas in the mind.' 2 Thus, for example, we might compare our idea of justice with our idea of stealing, and observe that they did not correspond; whence the truth that stealing is unjust may be proved with the same certainty as the truth that three angles of a triangle are equal to two right angles. The obvious difficulty is, that this doctrine seems to make morality certain in the sense in which a verbal proposition is certain, and in that sense alone. We are merely unfolding our definition, or explaining that what we call just does not include what we call stealing. 1 See letter to Molyneux of March 30, 1696. 2 Locke's Essay, book iii. ch. xi. secs. 16 and 17.

This remark was made by Berkeley. To demonstrate morality,' he says, in his commonplace book, 'it seems one need only make a discovery of words and see which included which. . . Locke's instances of demonstration in morality are, according to his own rule, trifling propositions.' Locke, it is clear, never distinctly realised his own position, and, whatever escape he might have attempted, it is plain that no such process as he contemplates could be reconciled with his general utilitarianism. The certainty which he would attain is not a certainty as to the tendency of actions to produce happiness. Any such theory must involve an objective element, and, on Locke's general theory, cannot be part of scientifical knowledge. Here, as in the whole philosophy of which it forms a part, Locke's teaching is palpably inconsistent, and the attempt to deduce a coherent doctrine would be waste of labour.

92. Berkeley's moral theory is not sufficiently prominent to require investigation. The next great theorist of Locke's school was Hume, and Hume preferred his moral treatise to all his other writings. The reason for this preference, so far as one ever can discover an author's motives for selfjudgment, will be tolerably plain. Here, we may say, Hume has, at least, some excuse for saying that he has obtained a definite constructive result. When Hume gave a second version of his metaphysics and psychology in the Essays, he mangled the earlier Treatise of Human Nature' with singular want of parental affection. Part is rewritten, and much is altogether omitted. The later version of his ethics contained in the 'Enquiry' bears a different relation to the ethics of the treatise. All the essential principles reappear, though some points are more lightly touched; but they reappear in a substantially new exposition. The literary texture of the 'Enquiry' shows everywhere the magic touch of Hume's lucid intellect. Morality, perplexed or mysterious with most of his predecessors, becomes admirably simple. All the doctrines fall into their place spontaneously. One obvious principle solves all doubts. The very lucidity may appear suspicious to many thinkers; but all must admit that the essential doctrines of utilitarianism are stated by Hume with a clearness

1 Berkeley, Works, iv. 449.