terms admit the probability that one's interpretation is incorrect, one must in each particular case make the most of the best light that one possesses. If any of Plato's arguments appear to the student to be patently unsound, they must pass with him for unsound until he has been convinced of the contrary. And while in the midst of one's reading it may be necessary to lose oneself by a species of dramatic illusion in the thoughts and feelings of the past; still there must be times of afterthought when one attempts to bring past and present together,—to sum up the permanent contributions of by-gone schools to one's own world-view. Nevertheless, as we must in the second place remark, the significance of these admissions is modified by the fact that the present world-view, by which we judge the past, is itself in process of development. It is not even as if we possessed a fixed body of scientific doctrine, which could be modified only by accretion; that is to say, by the addition of new facts and prinIciples which should leave the old unchanged and undisturbed. If that were the case, an objective and final criticism of earlier theories would not be so impracticable. But, on the contrary, the progress of science is a true evolution, an organic growth, in which no part is wholly unaffected. Time-honored formulæ, even if unrefuted, are narrowed in their field of application, or, by inclusion in more comprehensive generalizations, become possessed of a new significance. Thus, while two and two still make four and doubtless will continue to do so, the science of arithmetic has had a new birth and the general conception of number itself has been transformed, since the establishment by Cantor of the existence of distinct 'transfinite' numbers.

In the third place, the chief motive which we have for studying the thought of the past is such as to make sympathetic criticism of the greatest possible importance. For that motive is selfknowledge, the analysis of the categories of contemporary thought in the light of their development. The method of analysis to be employed is fundamentally the same as is used in genetic investigations of every sort. As a moving object is easier to

distinguish than one at rest, so the developing organism reveals the intricacies of its structure far more easily than the same organism studied at only a single stage of its life-history. And so also the philosophical conceptions, which to our direct examination appear to be inexplicable intuitions of the human mind, may exhibit their hidden content with the greatest clearness when the record of their various metamorphoses lies before us. Thus the question of deepest interest is not: "How far can Plato's thought be made to square with the science of today?" but rather: "How far has Plato's thought entered into the living tissue of the science of today?" The most valuable criticism, therefore, is contained in a plain and clear exposition. The best refutation of a theory is the unvarnished history of its transformations.

To many of our readers all that we have just now been saying must appear to be sheer truism; and very few will question its substantial correctness. It may, therefore, be thought worthy of note, that not one of the writers whom we have mentioned would have found a word of truth in the whole discussion. No feature, in fact, is more characteristic of the old dogmatism than the general incapacity of thinkers of both schools to recognize the fact (or the possibility) of an evolutionary progress of human knowledge. If science should advance, it must be by the addition of new truths to old. That half-truths might grow into whole ones was unsuspected. As truth was absolute truth, so error was absolute error; and as the former was most advantageous, so the latter (whether avoidable or unavoidable) was most detrimental, to the acquirement of further truth.

The

Of the rationalists this holds true as a matter of course. very essence of anti-evolutionism is expressed by Spinoza in his letter to a recreant pupil: "I do not presume that I have found the best philosophy, I know that I understand the true philosophy. If you ask in what way I know it, I answer: In the same way as you know that the three angles of a triangle are equal to two right angles." It is not a question of comparisons! But

1Letter LXXIV, Elwes tr

the most complete illustration of which we are aware is to be found in the second part of the Discourse on Method. The very first reflection which is there recorded is upon the fact, "that there is seldom so much perfection in works composed of many separate parts, upon which different hands have been employed, as in those completed by a single master." This might be seen in buildings, cities, religions, and civil constitutions. The same is true of the "sciences contained in books"—at least the nonmathematical sciences. And finally the very development of each one of us from infancy is a most unfortunate necessity. "It is almost impossible that our judgments can be so correct and solid as they would have been, had our reason been mature from the moment of our birth, and had we always been guided by it alone." It is under the influence of this reflection that he determines upon a clean sweep of his previously entertained opinions, and that he adopts as the first maxim of his future scientific endeavors: “never to accept anything for true which I did not clearly know to be such."

Among empiricists the same blindness to the possibility of a true evolution of knowledge prevails. "Nothing," says Hume, "is more usual and more natural for those who pretend to discover anything new to the world in philosophy and the sciences, than to insinuate the praises of their own systems, by decrying all those which have been advanced before them." This attitude is typical of pre-evolutionary thought; and Hume does not deny that it is substantially his own. He is only concerned to excuse the implied effrontery of his new pretensions. And his excuse is the usual one. He has found a new mode of attack, a new avenue of approach; he is applying new methods, or is radically enlarging the scope of old ones. Thus he hopes to succeed where others have failed. That his own philosophy is an almost inevitable outgrowth of the speculations of Locke, Berkeley, and Hutcheson, he does not for a moment suspect.

The certainty of immediate experience "seeing is believing"

1 Treatise of Human Nature, Introduction.

is a principle which in the history of philosophy may be said to date from Empedocles of Acragas, but which common sense has no doubt held from time immemorial. Not that either philosophy or common sense has always been agreed upon the matter. Indeed, as a philosophical dogma, nobody would ever have thought of asserting such a proposition, had it not previously been denied. With Empedocles it was simply a reassertion of the trustworthiness of clear perception which in the previous period of Greek philosophy had become more and more deeply suspected. The physical theories of the early cosmologists had been so utterly out of accord with ordinary observation that they (or their followers) had inevitably been led to exalt the authority of discursive reason as over against that of direct observation; until with Heraclitus, and still more with Parmenides, an absolute scepticism of the senses resulted.

In the generation following Empodocles this scepticism took on a new and more subtle form. According to a theory ascribed to Protagoras, the senses are indeed absolutely trustworthy in so far as they simply make each man aware of his own perceptive states of consciousness; but they give him no insight into the sensations of other men or into the nature of things. The notion of an indubitable immediate experience is thus preserved; but the range of its significance is seriously restricted. This doctrine of the relativity of sense-perception was maintained by all of the more important thinkers of antiquity (except the Stoics); and in modern times it has been held by both rationalists and empiricists. Certain of the latter, indeed, have made it a ground for doubting, or denying altogether, the existence of any object over and above the sensations themselves.

A second form of immediatism philosophy owes to Plato. Desiring, as a constructive social reformer, to found the theory and practice of politics upon a basis of indubitable truth, it appeared clear to him that Protagoras had taken away the hope of discovering such a basis in the evidence of the senses. Yet he saw that as a matter of our life-history all knowledge starts from

sensation. A partial solution of the difficulty he found in the example of geometry. Without sensible diagrams the geometrician could accomplish little, and yet the most exquisitely constructed diagram was far from conforming to the exact requirements of the science. It was evidently as a suggestion that the diagram was useful,—a suggestion of a perfect prototype which it weakly imitated. Was this not true of all our scientific ideas, including those of morality and statecraft (in which Plato was most deeply interested)? Is not the good man whom we seebrave, wise, temperate, and just as he may be a very imperfect illustration of the ideal courage, wisdom, temperance, and justice of which we can conceive, and of which the philosopher attempts to frame adequate definitions? But if the conceptions of the mathematical and moral sciences are not logically derived from sense-impressions, but only suggested by them, what logical ground have they? It seems to have been properly held by the geometricians, that the fundamental conceptions of their science were self-evident and needed no further warrant. But Plato saw that this was not so. He perceived that all these conceptions involved assumptions that might perfectly well be questioned and that the geometricians had no way of defending; and he believed the like to be true of the moral sciences.

In order properly to found both classes of sciences, one must, he thought, adopt a course directly the reverse of deduction. Frankly recognizing their fundamental assumptions as mere hypotheses, one must seek for more comprehensive hypotheses which shall unite and explain the former. And the new hypotheses must be similarly treated; and the process must be repeated again and again until it is no longer necessary or possible. That is to say, the process must be repeated until a conception is reached which is no longer hypothetical, but which is indeed self-explanatory and capable of explaining and justifying all the conceptions that have led up to it. The content of this highest conception Plato called the Good; and because the conception was itself incapable of being explained in simpler terms, but