The increase of knowledge, however, has not only tended to limit the influence of language upon thought, but it has given a new conception of experience. The value of experience in scientific enquiry depends on the amount of facts already collected, and on the progress that has been made in digesting them in the form of generalisations. Every new fact of observation, every impression on the senses, calls up a series of accepted and ascertained theories; and it is from this stock of theory that it derives, not indeed its truth as a fact, but its power of modifying or confirming opinion, its clearness to the understanding, and even its power of retaining a hold on the memory. Plato did not start at a point in the progress of science at which the observation of particulars is applicable, except in the most imperfect way, to discovery. He is like a man using his eyes for the first time, who fancies, because everything seems equally near, that sight cannot tell him the forms and distances of objects. Hence he could not systematically test opinions or notions by facts, but by comparing them with other opinions and notions, either consciously held or implied in language. His error was not in devoting himself to the analysis of abstractions; for, as Mr. Jowett says, summing up the whole matter in a line: Before men could observe the world they must be able to conceive the world.' His error lay in giving to abstractions, as such, an absolute value; in supposing that the clearness which general notions give to experience was a clearness which they had in themselves apart from experience. Yet the Platonic mode of thought, which concerns itself with the abstractions under which phenomena are conceived, has its place alongside of the study of these phenomena in detail. The clearness and just co-ordination of ideas which makes the philosophic habit of mind is not the same thing as the agreement of ideas with facts which constitute scientific accuracy; and positive science does not supersede metaphysics, except as it works out in their application the conceptions which metaphysics have supplied.

The value of Plato's scheme of Dialectic, as Mr. Jowett is careful to point out, lay in the high ideal which it held up as an aim to the science of the future. 'The correlation of the sciences, the consciousness of the unity of knowledge, the sense of the importance of classification, the unwillingness to stop short of certainty or to confound probability with truth, are important principles of the higher education' (vol. ii. p. 157). On the other hand the weakness of the theory was soon felt in the difficulty of ex

plaining consistently the very various degrees of value which Plato would not but recognise in the impressions and beliefs included by him under the term 'opinion' or the seeming.' He is far from treating everything which falls short of his concep tion of knowledge as equally worthless; but he is much at a loss for a satisfactory account of the true or valuable element contained in particular instincts, conjectures, habits, and feelings. The modes in which he approaches the different sides of this problem form, perhaps, the most generally interesting part of his philosophy; for (as may be readily supposed) it is in connec tion with these attempts, rather than with more abstruse enquiries, that positive and fruitful results are chiefly to be found. Three or four points of view may be distinguished, from which the solution is more or less consciously attempted: (1.) Mythology; (2.) Supernatural influence or madness; (3.) Morality based upon habits only; and (4.) Systems of positive law.

1. Plato's view of the office of mythology is expressed in the 'Republic,' where he recognises it as the earliest instrument of edu cation, to be used in order to accommodate truth to the tender mind; but partly also on account of our own uncertainty. In the tales of mythology, of which we were just now speaking, because we do not know the truth about ancient tradition, we make falsehood as much like truth as we can.' So in the Phædrus,' the famous allegory (as we should term it) is called by Socrates himself a tolerably creditable and possibly true though partly erring myth.' The value and instructiveness of a myth depends, therefore, on its being 'probable.' Probability, so used, does not mean so much that the chances are in favour of its being true, as that it reflects certain truth, or embodies it in the concrete, and consequently will prepare the way for the reception of the same principles in a more abstract shape.

An acute German critic* has endeavoured to show that Plato only resorts to the mythical form when he is met by the necessity of explaining the origin or growth (yéveo) of a thing. The theory of Ideas, he argues, is a theory of the existent as necessary and immutable; the process of becoming has logically no place in it: Plato intended his myths to do for philosophy what the popular mythology did for religion-to express a fundamental series of relations in a narrative form, as something which is, and also which has come to be what it is. Thus (to take

the most prominent example), the myth in the 'Phædrus' reconciles the eternity of the mind and of knowledge with the rise and progress of knowledge in the individual. The theory, however, although it is highly suggestive, and opens up a new and interesting side from which to compare the ancient opposition of the real and the phenomenal with the modern idea of development, can hardly be applied to all the myths in the Platonic dialogues. A more adequate account is suggested in Mr. Jowett's remarks on the second book of the Republic' (vol. ii. pp. 37 ff.). 'Art' (under which the composition of myths is included) may be another aspect of reason;' and 'this conception of art is not limited to strains of music or the forms of plastic art, but pervades all nature.' Mythology, in short, is made (like the mathematical sciences) a universal type; it represents the effort of the philosophic imagination to find modes of conceiving the unknown. In this wide sense there are myths taking the form, not only of history, geography, and cosmogony, but even of arithmetic and etymology. Thus the number in the Republic' expresses an undiscovered numerical relation, which is believed by Plato to govern the periodical decay inevitable in all human society. And the derivations in the 'Cratylus express an equally undiscovered relation between the sounds of words and the things which they represent. In neither case is the truth of the myth maintained; only its probability or 'likeness' to truth; as we should say, its fitness to suggest truth.

2. The description of the pursuit of truth under the figure of a divine madness is found along with the mythical imagery of the Phædrus,' but it exemplifies a distinct mode of representing the true instincts which yet fall short of knowledge. Of madness Plato there says there are four kinds: that of prophets, of the mysteries, of poetry, and of love; and of these the last is also the best. The enthusiasm of the lover is a lower form, a shadow,' of that of the philosopher; the object of the passion is desired because of the true relations which (like the productions of true art) it embodies in a concrete form. Thus there is a progress from sense to reason; the erotic madness passes, if rightly directed, into that enthusiastic anticipation of knowledge (called the love of wisdom,' ptλooopía) which animates the search for absolute truth. At the end of the Meno,' the right opinion by which statesmen have guided cities is said to be in politics what divination is in religion' (p. 99). The same theory, applied to poetry, is drawn out in the 'lon,' and in a

passage of the Republic,' which prescribes the manner of treating the multiform' or imitative poet. We will fall down and worship him as a sweet and holy and wonderful being, but we must also inform him that there is no place for such as he is in our state-the law will not allow them. And so when we have anointed him with myrrh, and set a garland of wool upon his head, we shall send him away to another city.' (p. 398). The tone of this passage, and of the Ion,' is that of a gentle contempt for the irrational element. In other places, however, the same thing is treated with the ut most respect. Thus, in the Laws,' it is said that Athenians, when good, are so in spite of their constitution, by a divinelygiven nature. Hence it is not necessary to suppose that the theory in the 'Meno,' unPlatonic as it seems, is proposed in irony; of which, Mr. Jowett remarks, there is no trace. 'A person may have some skill or latent experience, which he is able to use himself, and is yet unable to teach others, because he has no principles and is not able to collect and arrange his ideas. He has practice, but not theory; art, but not science. This a true fact of psychology, which is recognised by Plato in this passage' (vol. i. p. 253.). We may add that it is a fact which the Socratic doctrine and that of Plato's earlier writings had ignored; so that the Meno' may be thought to mark Plato's first attempts to place the relation of virtue and knowledge in a truer light. Plato, we may suppose, felt the difficulties of the Socratic identification, and had not yet gained the higher point of view-that of Dialectic upon which his own identification ultimately rests.

เ

3. In the Republic,' the progress from sense to knowledge is represented by means of a psychology from which mythical and allegorical elements are finally excluded. The efficacy of the various means of moral education in preparing the way for the higher or scientific morality is now ascribed, as in the Ethics of Aristotle, to the influence of habit. Rhythm and harmony find their way into the secret places of the soul, on which they mightily fasten, bearing grace in their movements and making the soul graceful of him who is rightly educated, or ungraceful if ill-educated;' and he who is thus trained will justly blame and hate the bad now in the days of his youth, even before he is able to know the reason of the thing; and when reason comes he will recognise and salute her as a friend with whom his education has made him long familiar.' (p. 402 Steph.). In the scheme of the seventh book this training is referred to as the music

solve, the difficulty of the sufficiency of justice for happiness. 'The two brothers ask Socrates to prove to them that the just is happy when they have taken from him all that in which happiness is ordinarily supposed to consist' (Ibid., p. 20). And Socrates undertakes this proof. His answer in substance amounts to this, that under favourable circumstances, i. e. in the perfect State, justice and happiness will coincide; and that when justice has been once found, happiness may be left to take care of itself (Ibid., p. 22). This, however, is only the happiness of the State. The happiness of the individual depends, according to the sequel of the Republic,' not upon the perfect State, but upon the perfect or just individual. The royally constituted man' is especially happy when he is king in the ideal State, and the tyrannical man especially miserable when he is also a tyrant; but this is, in both cases, an exceptional enhancing of their position. In reality, as Aristotle perceived, the question is not so much, 'What is justice?' as 'What is happiness?' If happiness consists in external goods, then justice (or the rule of society), in the strong man's view, is that he should get as much as he desires; in that of weak men, that they should combine to keep what they can. Or, if happiness consists in the pleasure of the greatest number, then justice depends upon the conduct by which that pleasure may be best secured. But if happiness is an idea-the application to human life of a higher abstraction, the Idea of good, or the realization of human perfection, or under whatever form an ideal philosophy of ethics may conceive it-then it is the task of such a philosophy to harmonise this idea with its conception of the world and of knowledge. If Plato fails, as Mr. Grote says, by representing (in the just man of the 'Gorgias') a superhuman or transcendent virtue; or, again, as Mr. Mill points out, by finding no worthy place for an Aristides, a man whose justice consists in implicit obedience to law and traditional morality; the reason is, that in his ethics, as in other parts of his system, the highest truth is made to reside in the purest attainable abstraction. The notion of happiness, apart from ingredients, is parallel to the notion of an astronomy without the visible heavens, or of harmonics without audible harmony.'

The manner in which Plato treats the question of pleasure varies in the different dialogues, but always exhibits the tendency to make light of that which presents itself as a fact or process rather than as an idea. In the 'Protagoras' Socrates begins, indeed,

by assuming that pleasure is merely another name for good; but he soon shows that the choice of pains and pleasures involves comparison between them, and therefore an 'art of measurement.' Pains and pleasures, it follows, are only, as it were, the material out of which the Good (or 'useful' or 'happiness') may be formed; whereas knowledge is the formative element. This mode of stating the theory of Socrates is hardly to be distinguished from the latest form of Utilitarianism; but with Plato, to whom the form or idea is always the real element, it led directly to the inference that pleasure is something transient and 'unreal,-a view which naturally acquired strength and consistency with the development of the theory of Ideas. In the Theaetetus,' again, Socrates shows that the apprehension of the useful, by bringing in the consideration of consequences, involves comparison, and therefore the universal element. In both cases, the difficulty which we feel in rightly understanding the issue arises from the extreme form in which the opposite doctrines are found. All philosophers, even the most opposed, would now agree in giving a value both to experience and to abstractions, and also in recognising pleasure as an element to be brought under regulation by a principle of some kind. Modern psychology lies wholly within the ancient extremes, Sense is the only knowledge,' 'Sense is delusive;' just as modern ethics lies within the analogous extremes, Pleasure is the good,' Pleasure is worthless.'

The 'confusion of ethics and politics' is not, strictly speaking, the Platonic confusion of the State and the individual as moral agents, but rather a confusion of the relations in which an individual stands to the State with those in which he stands to other groups or to the whole of mankind, to other sentient beings, and to his own character and prospects. The place which the organisation of the State has held in this general fabric of moral duty has varied in different periods of history; but the tendency has been, on the whole, towards diminishing its importance. The duties enforced by law, or by a custom having the stringency of law-though never in Greece, perhaps, co-extensive with morality much less nearly so than they were in Plato's time. The State, moreover, does not now make so exclusive a claim on the regard of its citizens. Other forms of common action and sentiment-the town or district, the Church, the European public, the brotherhood of mankind-divide with it the interest once concentrated on the Hellenic