Neither here nor in the Phaedrus or Symposium, nor yet in the Philebus or Sophist, does he give any clear explanation of his meaning. He would probably have described his method as proceeding by regular steps to a system of universal knowledge, which inferred the parts from the whole rather than the whole from the parts. This ideal logic is not practised by him in the search after justice, or in the analysis of the parts of the soul; there, like Aristotle in the Nicomachean Ethics, he argues from experience and the common use of language. But at the end of the sixth book he conceives another and more perfect method, in which all ideas are only steps or grades or moments of thought, forming a connected whole which is self-supporting, and in which consistency is the test of truth. He does not explain to us in detail the nature of the process. Like many other thinkers both in ancient and modern times his mind seems to be filled with a vacant form which he is unable to realize. He supposes the sciences to have a natural order and connexion in an age when they can hardly be said to exist. He is hastening on to the ‘end of the intellectual world’ without even making a beginning of them.
In modern times we hardly need to be reminded that the process of acquiring knowledge is here confused with the contemplation of absolute knowledge. In all science a priori and a posteriori truths mingle in various proportions. The a priori part is that which is derived from the most universal experience of men, or is universally accepted by them; the a posteriori is that which grows up around the more general principles and becomes imperceptibly one with them. But Plato erroneously imagines that the synthesis is separable from the analysis, and that the method of science can anticipate science. In entertaining such a vision of a priori knowledge he is sufficiently justified, or at least his meaning may be sufficiently explained by the similar attempts of Descartes, Kant, Hegel, and even of Bacon himself, in modern philosophy. Anticipations or divinations, or prophetic glimpses of truths whether concerning man or nature, seem to stand in the same relation to ancient philosophy which hypotheses bear to modern inductive science. These ‘guesses at truth’ were not made at random; they arose from a superficial impression of uniformities and first principles in nature which the genius of the Greek, contemplating the expanse of heaven and earth, seemed to recognize in the distance. Nor can we deny that in ancient times knowledge must have stood still, and the human mind been deprived of the very instruments of thought, if philosophy had been strictly confined to the results of experience.