EX BIBLIOTHECA FRANCES A. YATES Philosophies Ancient and Modern PLATO NOTE As a consequence of the success of the series of Religions Ancient and Modern, Messrs. Constable have decided to issue a set of similar primers, with brief introductions, lists of dates, and selected authorities, presenting to the wider public the salient features of the Philosophies of Greece and Rome and of the Middle Ages, as well as of modern Europe. They will appear in the same handy Shilling volumes, with neat cloth bindings and paper envelopes, which have proved so attractive in the case of the Religions. The writing in each case will be confided to an eminent authority, and one who has already proved himself capable of scholarly yet popular exposition within a small compass. Among the first volumes to appear will be : — Early Greek Philosophy. By A. W. Benn, author of The Philo- ^ sophy of Greece, Rationalism in the Nineteenth Century. Stoicism. By Professor St. George Stock, author of Deduc- tive Logic, editor of the Apology of Plato, etc. Plato. By Professor A. E. Taylor, St. Andrews University, author of The Problem of Conduct. Scholasticism. By Father Rickaby, S.J. Hoobes. By Professor A. E. Taylor. Locke. By Professor Alexander, of Owens College. Comte and Mill. By T. W. Whittaker, author of The Neoplatonists, Apollonius of Tyana and other Essays. Herbert Spencer. By W. H. Hudson, author of An Intro- duction to Spencer's Philosophy. Schopenhauer. By T. W. Whittakek. Berkeley. By Professor Campbell Eraser, D.C.L., LL.D. Bergsen. By Father Tyrrell. PLATO By A. E. TAYLOR LONDON ARCHIBALD CONSTABLE £sf CO Ltd I908 FOREWORD The following sketch makes no claim to be considered as a complete account of the philo- sophy of Plato. Many topics of importance have been omitted altogether, and others only treated with the utmost attainable brevity. I have also thought it necessary to avoid, as far as possible, all controversial discussion, and have therefore in many cases followed my own judgment on disput- able points without attempting to support it by the detailed reasoning which would be indispens- able in a work of larger scope. My object has been to sit as loose as possible to all the tradi- tional expositions of Platonism, and to give in broad outline the personal impression of the philosopher's thought which I have derived from repeated study of the Platonic text. The list of works useful to the student, though it merely comprises a few of those which I have myself found useful or important, will give my reader the opportunity to form his own judgment by comparing my interpretations with those of v PLATO others. Those who are most competent to con- demn the numerous defects of my little book will, I hope, be also most indulgent in their verdict on an attempt to compress into so small a compass an account of the most original and influential of all philosophies. A. E. T. vi CONTENTS CHAP. TAOE i. Life and Writings 1 n. Knowledge and its Objects .... 34 in. The Soul of Man— Psychology, Ethics, and Politics 73 iv. Cosmology 137 Select Bibliography 149 vii PLATO CHAPTER I LIFE AND WRITINGS The traditional story of the life of Plato is one in which it is unusually difficult to distinguish between historical fact and romantic fiction. Of the ' Lives ' of Plato which have come down to us from ancient times, the earliest in date is that of the African rhetorician and romance-writer Apuleius, who belongs to the middle and later half of the second century a.d. There is a longer biography in the scrap-book commonly known as the Lives of the Philosophers by Diogenes of Laerte, a compilation which dates, in its present form, from a time not long before the middle of the third century a.d., though much of its material is taken from earlier and better sources. The remaining 'Lives' belong to the latest age of Neo-Platonism, i.e. the sixth century after Christ and later. Thus the earliest extant bio- A I PLATO graphy of the philosopher comes to us from a time four hundred years after his death, and must be taken to represent the Platonic legend as it was current in a most uncritical age. When we try to get behind this legend to its basis in well-accredited fact, the results we obtain are singularly meagre. Plato himself has recorded only two facts about his own life. He tells us, in the Apology, that he was present in court at the trial of his master Socrates, and that he was one of the friends who offered to be surety for the payment of any fine which might be imposed on the old philosopher. In the Phaedo he adds that he was absent from the famous death-scene in the prison, owing to an illness, a statement which may, however, be no more than an artistic literary fiction. His contemporary Xenophon merely mentions him once in passing as a mem- ber of the inner Socratic circle. From Aristotle we further learn that Plato, as a young man, apparently before his intimacy with Socrates, had been a pupil of the Heraclitean philosopher Cratylus. A few anecdotes of an unfavourable kind are related by Diogenes of Laerte on the authority of Aristoxenus of Messene, a pupil of Aristotle, and a well-known writer on music, whose credibility is, however, impaired by his 2 LIFE AND WRITINGS unmistakable personal animus against Socrates and Plato, and his anxiety to deny them all philosophical originality. The dates of Plato's birth and death are, moreover, fixed for us by the unimpeachable authority of the Alexandrian chronologists, whose testimony has been pre- served by Diogenes. We may thus take it as certain that Plato was born in the year 427 B.C., early in the great Peloponnesian war, and died in 346, at the age of eighty- one. The way in which Xenophon, in his one solitary statement, couples the name of Plato with that of Charmides, a leader of the oligarchy of the 1 Thirty/ set up by the Spartans in Athens at the close of the Pelo- ponnesian war, taken together with the promi- nence given in the Platonic dialogues to Charmides and Critias as friends of Socrates, confirms the later tradition, according to which Plato himself was a near relative of the two ' oligarchs,' a fact which has to be borne in mind in reading his severe strictures upon Athenian democracy. There remains, indeed, a further source of in- formation, which, if its authenticity could be regarded as established, would be of the very highest value. Among the writings ascribed to Plato and preserved in our ancient manuscripts there is a collection of thirteen letters, purport- 3 PLATO ing to be written by the philosopher himself, some of which ostensibly contain a good deal of autobiographical detail. In particular the seventh letter, the longest and most important of the group, professes to contain the philosopher's own vindication of his life-long abstention from taking part in the public life of his country, and, if genuine, absolutely confirms the later story, pre- sently to be narrated, of his political relations with the court of Syracuse. As to the history of this collection of letters, all that we know for certain is that they were in existence and were regarded as Platonic early in the first century A.D., when they were included by the scholar Thrasyllus in his complete edition of the works of Plato. This, however, is not of itself proof of their genuineness, since the edition of Thrasyllus contained works which we can now show to be spurious, such as the Theages and Erastae. We further know from Diogenes of Laerte that cer- tain 'letters' had been included in the earlier edition of Plato by the famous scholar Aristo- phanes, who was librarian of the great museum of Alexandria towards the end of the second century B.C. ; but we are not told which or how many of our present collection Aristophanes recognised. When we examine the extant letters 4 LIFE AND WRITINGS themselves, we seem led to the conclusion that they can hardly all be genuine works of Plato, since some of them appear to allude to character- istic doctrines of the Neo-Pythagoreanism which arose about the beginning of the first century before Christ. It is not surprising, therefore, that Grote has stood alone, or almost alone, among recent scholars in maintaining the genuineness of the whole set of thirteen letters admitted into the collection of Thrasyllus. It is another question whether some at least of the collection, and notably the seventh, the only letter of real importance, may not be the work of Plato, and the problem must be said to be one upon which competent scholars are not as yet agreed. On the one side, it may be urged that the incidents related in the seventh letter are in no way incredible, and that their occur- rence, as we shall see directly, would explain a certain increase of pessimism in Plato's later writings on political philosophy. On the other, it is suspicious that the letter appears to quote directly from at least four Platonic dialogues (the Apology, Phaedo, Republic, Lysis), and that, apart from the account of Plato's relations with Syracuse, it contains nothing which might not have been put together with the help of the 5 PLATO dialogues. And we must remember that the desire to exhibit Plato, the great political theorist, actually at work on the attempt to construct a state after his own heart, would at any time have been a sufficient motive for the fabrication. Still, the style of the composition shows that, if a forgery, it is at least an early forgery, and we shall hardly be wrong in treating the narrative as being, at any rate, based upon a trustworthy tradition. Having premised this much as to the sources of our information, we may now proceed to nar- rate in outline the biography of Plato, as it was current early in the Christian era, omitting what is evidently myth or mere improving anecdote. Plato, the son of Ariston and Perictione, was born either in Athens or, according to another account, in Aegina, in the year 427 B.C. On the mother's side he was closely related to Critias and Char- mides, members of the oligarchy of the * Thirty/ and the former the leader among its more violent spirits, the family going back through Dropides, a relative of the great lawgiver Solon, to a divine first ancestor, the god Poseidon. On the father's side, too, his origin was no less illustrious, since Ariston was a descendant of Codrus, the last king of Athens, who was himself sprung from Poseidon. 6 LIFE AND WRITINGS Even this origin, however, was not thought exalted enough for the philosopher by his admirers, and Plato's own nephew, Speusippus, is cited as an authority for the belief that the real father of Perictione's son was the god Apollo. (The relationship between Plato and the family of Critias and Charmides is, as we have said, made probable by the philosopher's own utter- ances, and he is also himself the authority for the descent of Critias from Dropides. The further assertions about the eminent descent of Dropides are hardly worthy of credit, since it seems clear that Solon the lawgiver was really a middle-class merchant. But the connection with Solon of itself shows that the family was one of the highest distinction as families went in the Athens of the late fifth century.) As a lad, the future philo- sopher was ambitious of poetical fame, and had even composed a tragedy for public performance. But when he came under the influence of Socrates, he devoted himself entirely to philo- sophy and burned all his poems. (That so great an imaginative writer as Plato should have begun his literary career as a poet is likely enough, and there is no reason why some of the epigrams ascribed to him in the Greek Anthology should not be genuine, but the story of the burnt tragedy 7 PLATO looks like a fabrication based upon the severe condemnation of poetry in general and the drama in particular in the Republic ; nor must we for- get that, according to Aristotle's statement, Plato got his introduction to philosophy not from Socrates, but from Cratylus.) The first association between Plato and Socrates took place when Plato was twenty years old, and their con- nection lasted eight years, since the death of Socrates falls in 399 B.C. After the death of the master, Plato retired from Athens and spent some years in foreign travel. The accounts of the extent of these travels become more and more exaggerated as the narrators are increas- ingly removed in date from the actual events. The seventh ■ letter ' speaks merely of a voyage to Italy and Sicily undertaken apparently in con- sequence of the writer's disgust with the proceed- ings of the restored Athenian democracy, which had inaugurated its career by the condemnation of Socrates. Cicero, who is the earliest authority for the story of the travels, apart from the ' letters,' makes Plato go first to Egypt, afterwards to Italy and Sicily. The later Platonic legend professes to know more, and relates an entire romance on the subject of Plato's adventures. According to this story Plato withdrew from 8 LIFE AND WRITINGS Athens on the death of Socrates, and resided for a while at the neighbouring city of Megara with his friend and fellow-disciple Eucleides. He then visited Cyrene, to enjoy the society of the mathe- matician Theodorus, Egypt, where he learned the wisdom of the priests, and Italy, where he asso- ciated vrith the members of the Pythagorean school who had survived the forcible dissolution of the political power of the sect. (The tale ran that he further purposed to visit the Persian Magi, but that this scheme failed, though some writers professed to know that Plato had met with Magians and learned their doctrines in Phoenicia.) From Italy the legend brings Plato to Sicity, where he is said, on the authority of the seventh letter, to have arrived at the age of forty; i.e. after twelve years of continuous travel. Here he visited the court of the vigorous ruler of Syracuse, Dionysius L, and so displeased that arbitrary monarch by his freedoms of speech that he caused him to be kidnapped by a Spartan ambassador who put him up for sale in the slave-market of Aegina, where, by a singular coincidence, the people had passed a resolution that the first Athenian who should land on the island should be put to death. Plato was, however, saved from his danger by a man of Cyrene, who ransomed 9 PLATO him and sent him home to Athens. (How much truth there may be in this story, the details of which are differently given by the different narrators, it is impossible to say with certainty. The story of the kidnapping, in particular, is told with a good deal of discrepancy in the details ; many features of it are highly singular, and it appears entirely unknown to both Cicero and the writer of the seventh ' letter/ Hence there is every ground to regard it as pure romance. The same must be said of the story of the twelve years' unbroken travel, and the association of Plato with Oriental priests and magicians. Stories of this kind were widely circulated from the beginning of the first century before Christ onward, when the gradual intermingling of East and West in great cities like Alexandria had given rise to the fancy that Greek science and philosophy had been originally borrowed from Oriental theosophy, a notion invented by Alex- andrian Neo-Pythagoreans and eagerly accepted by Jews and Christians, whom it enabled to represent the Greek sages as mere pilferers from the Hebrew scriptures. Even the alleged resi- dence in Megara and the voyage to Cyrene, may be no more than inventions based on the facts that the dialogue Theaetetus is dedicated to Platans 10 LIFE AND WRITINGS friend Eucleides of Megara and that the Cyrenian mathematician Theodoras is one of its dramatis personae. So again the frequent allusions in the dialogues to Egypt and Egyptian customs may be due to reminiscences of actual travel in Egypt, but can hardly be said to show more knowledge than an Athenian might have acquired at home by reading Herodotus and conversing with traders from the Nile Delta. On the other hand, the story of the visit to Italy and Sicily is con- firmed by the fact that Plato's works, as is well known, show considerable familiarity both with Pythagorean science and with the Pythagorean and Orphic theological ideas, and that the first dialogue in which this influence is particularly noticeable is the Gorgias, the work in which, as is now generally recognised. Plato speaks for the first time in the tone of the head of a philo- sophical school or sect. It is thus probable that Plato's final settlement at Athens as a philo- sophical teacher was actually preceded, as the tradition dating at least from the seventh ' letter ' asserts, by a visit to the home of Pythagoreanism, the Greek cities of Sicily and Southern Italy.) We have to think of Plato, then, as definitely established, from about 387 B.C., in Athens as the recognised head of a permanent seat of learning, 1 1 PLATO a university, as we might call it in our modern terminology. The home of this institution was in the north-western suburb of Athens known as the Academy, in consequence of the presence there of a shrine of the local hero Academus. Here Plato possessed a small property, which was per- haps (the words of the legend as preserved by Diogenes are obscure,) purchased for him by his foreign friends. From this circumstance the philosophical school founded by Plato came to be known in later days as the ' Academy/ It was not the first institution of the kind ; Plato's contem- porary and rival, the rhetorician and publicist Isocrates, had already gathered round him a similar group of students, and the writings of both authors bear traces of the rivalry between them. Their educational aims were, in fact, markedly different. Isocrates desired, first and foremost, to turn out accomplished and capable men of action, successful orators and politicians. Plato, on the other hand, was convinced that though the trained intelligence ought to direct the course of public life in a well-ordered society, the equipment requisite for such a task must first be obtained by a thorough mastery of the prin- ciples of science and philosophy, and was not to be derived from any superficial education in 12 LIFE AND WRITINGS 'general culture/ Thus, while Plato, in well- known passages, describes the pupils of Isocrates as ' sinatterers ' and ' pretenders to philosophy,' Isocrates, on his side, depreciates those of Plato as unpractical theorists. Of the precise nature of the teaching in Plato's Academy, unfortunately, little is known, but the reports of later tradition, such as they are, indicate that the author of the Re- public carried his theories of education into practice, and made the thorough and systematic study of exact mathematical science the founda- tion of all further philosophic instruction. The story that the door of the Academy bore the inscription ' Let none unversed in geometry come under this roof,' is, indeed, first found in the works of a medieval Byzantine, but its spirit is thoroughly Platonic. The outward peculiarity, it must be remem- bered, by which the education given by both Plato and Isocrates differed from that afforded by the eminent ' sophists ' of the last half of the fifth century, was that their teaching was more con- tinuous, and that it was, in theory at least, gratuitous. The great sophist of the past had usually been a distinguished foreigner whose task of making his pupils 1 good men, able to manage their own private affairs and the affairs of the 13 PLATO nation well/ had to be accomplished in the course of a flying visit of a few weeks or months, and he had also been a professional educator, depending upon his professional fees for his livelihood, and therefore inevitably exposed to the temptation to make his instruction attractive and popular, rather than thorough. Plato and Isocrates, on the other hand, were the heads of permanent schools, in which the education of the pupil could be steadily carried on for a protracted period, and where he could remain long after his time of pupilage proper was over, as an associate in the studies of his master. They were, moreover, not dependent for subsistence upon payments by their pupils, and were hence free from the neces- sity to make their teaching popular in the bad sense of the term, though it is only fair to add that neither had any objection to the occasional reception of presents from friends or pupils, and that Isocrates, at least, required a fee from foreign students. It is in virtue of this permanent and organised pursuit of intellectual studies, and this absence of 1 professionalism ' from their teaching, that we may call Plato and Isocrates the joint creators of the idea of what we now understand by university education. The remark I have just made about the absence of ' professionalism ' from 14 LIFE AND WRITINGS their scheme of instruction will, I hope, explain that persistent objection to the sophists' practice of demanding a fee for their courses which Grote found so unreasonable on the part of Plato and Aristotle. Plato's long life of quiet absorption in his self- chosen task as a director of scientific studies and a writer on philosophy, was destined to be once at least disastrously interrupted. The details of his abortive attempt to put his theories of govern- ment into practice at Syracuse must be sought in the histories of Greece. Here it must suffice to recapitulate the leading facts, as related in the 'letters,' and, apparently without any other authority than the ' letters,' in Plutarch's life of Dion. In the year 367 B.C., when Plato was a man of sixty and had presided over the Academy for twenty years, Dionysius i of Syracuse died, leaving his kingdom to his son Dionysius il, a weak but impressionable youth. The actual direction of affairs was, at the time, mostly in the hands of Dion, the brother-in-law of Dionysius i. and an old friend and admirer of Plato. Plato himself had written in his Republic that truly good government will only be possible when a king becomes a philosopher, or a philosopher a king, i.e. when the knowledge of sound prin- 15 PLATO ciples of government and the power to embody them in fact are united in the same person. Dion seems to have thought that the circum- stances at Syracuse offered a favourable oppor- tunity for the realisation of this ideal. Why should not Dionysius, under the instructions of the great master, become the promised philoso- pher-king, and employ the unlimited power at his command to convert Syracuse into something not far removed from the ideal state of Plato's dream? To us, such a project seems chimerical enough, but, as Professor Bury has properly reminded us, the universal belief of Hellas was that a not very dissimilar task had actually been achieved by Lycurgus for Sparta, and there was no a priori reason for doubting that what Lycurgus had done for Sparta could be done for Syracuse by Plato. Plato was accordingly invited to Syra- cuse to undertake the education of the young prince. His reception was, at first, most promis- ing, but the thoroughness with which he set about accomplishing his work foredoomed it to failure. It was the first principle of his political system that nothing but the most thorough training of intelligence in the ideas and methods of science will ever fit a man for the work of governing mankind with true insight. Accord- 16 LIFE AND WRITINGS ingly he insisted upon beginning by putting his pupil through a thorough course of geometry. Dionysius, naturally enough, soon grew weary of this preliminary drill, and began to revolt against the control of his preceptors. An opportunity was found for banishing Dion, and though Diony- sius would have liked to keep Plato with him, the philosopher recognised that his scheme had failed, and speedily pressed for permission to return to Athens. A year or two later he paid another visit to Syracuse, apparently in the hope of reconciling Dion and Dionysius, but without result. The sequel of the story, the rapid development of Dionysius into a reckless tyrant, the expedition of Dion which led to the downfall and flight of Dionysius, the assassination of Dion by Callippus, another pupil of Plato, who then set himself up as tyrant, but was speedily overthrown in his turn by the half-brother of Dionysius, belongs to the history of Sicily, not to the bio- graphy of Plato. It is not unlikely that the disastrous failure of the Syracusan enterprise, and the discredit which subsequent events cast upon the members of the Academy, have much to do with the relatively disillusioned and pes- simistic tone of Plato's political utterances in the Theaetetus and Politicus as contrasted with B 17 PLATO the serenity and hopeful spirit of the greater part of the Republic. Yet, even in his old age, Plato seems to have clung to the belief that the experiment which had failed at Syracuse might be successful elsewhere. In his latest work, the Laws, which was possibly not circulated until after his death, he still insists that the one chance for the establishment of a really sound form of government lies in the association of a young and high-spirited prince with a wise law- giver. Nothing is recorded of the life of Plato after his last return from Syracuse, except that he died — legend says at a wedding-feast — in the year 347-6, at the age of eighty-one. His will, which is preserved by Diogenes, and is likely enough to be genuine, provides for a 'child Adeimantus,' who was probably a relative, as the same name had been borne by one of his half-brothers. Nothing further is known which throws any light on the question whether Plato was ever married or left any descendants. The scurrilous gossip collected by writers like Athenaeus, and the late Neo-Platonic traditions which make him into a celibate ascetic, are equally worthless. The headship of the Academy passed first for a few years to Speusippus, a nephew of Plato, and 18 LIFE AND WRITINGS then to Xenocrates of Chalcedon, another of the master's immediate pupils. The one man of real genius among the disciples, Aristotle of Stageira, took an independent course. For ten or eleven years after the death of the master, of whose school he had been a member from about 367-6 B.C. until 346, he was absent from Athens, being employed for part of the period (343-336 B.C.) as tutor to the future Alexander the Great, then Crown Prince of Macedonia. On his return in 335 he broke away from the Academy, and organised a new school with himself as its head. The formal reverence which Aristotle expresses in his writings for his predecessor was combined with a pugnacious determination to find him in the wrong on every possible occasion. Yet, in spite of the carping and unpleasantly self-satisfied tone of most of the Aristotelian criticism of Plato, the thought of the later philosopher on all the ultimate issues of speculation is little more than an echo of the larger utterance of his master, and it is perhaps as much by inspiring the doctrine of Aristotle as by his own utterances* that Plato has continued to our own day to exer- cise an influence in every department of philo- sophic thought, which is not less potent for being most often unsuspected. Of the direct and enor- 19 PLATO mously important influence of Platonism on the development of Christian theology this is perhaps hardly the place to speak. The works of Plato, we have reason to believe, have come down to us absolutely entire and com- plete. This is, no doubt, to be explained by the fact that the original manuscripts were carefully preserved in the Platonic Academy ; thence copies, as Grote has argued, would naturally find their way into the great library at Alexandria. It does not, however, follow that everything which our extant manuscripts of Plato contain must necessarily be Platonic. It would be quite easy, in course of time, for works incorrectly ascribed to Plato, or deliberately forged in his name, to be imposed upon the Alexandrian librarians, and to acquire a standing in the library, side by side with genuine writings derived directly from the original manuscripts preserved at first in the Academy at Athens. Indeed, the very anxiety of the Ptolemies, and their imitators the kings of Pergamus, to make their great collections of books as complete as possible, would furnish a powerful incentive to the unscrupulous to pro- duce alleged copies of works by famous authors. As it happens, we do not know either how long the original manuscripts of Plato continued to 20 LIFE AND WRITINGS exist undispersed (indeed, the very statement that they were kept in the Academy is an infer- ence from the probabilities of the case, and does not rest upon direct ancient testimony), nor what works of Plato were originally included in the Alexandrian library. The first trace which has been preserved of the existence of an edition of Plato in that library is the statement of Diogenes that the scholar Aristophanes of Byzantium made an arrangement of the works of Plato, in which certain of the dialogues were grouped together in 1 trilogies,' or sets of three, after the fashion of the tragic dramas of the fifth century. Diogenes gives the names of fourteen dialogues, which, together with a collection of 1 letters/ had been thus divided by Aristophanes into five tri- logies, and adds that the grouping was not carried out ' for the rest/ Unfortunately, he does not tell us the titles of the 1 rest/ so that we have no right to assert that everything now included in our manuscripts was recognised as Platonic at Alexandria in the time of Aristophanes. At a much later date, the grammarian Thrasyllus, who lived in the reigns of Augustus and Tiberius (i.e. in the early part of the first century a.d.), made a new classification of the Platonic dialogues into 1 tetralogies/ or groups of four, on the analogy of the 21 PLATO old tragic tetralogy of three tragedies followed by a satyric play. The Platonic canon of Thrasyllus contained nine of these tetralogies, i.e. thirty- five dialogues with a collection of thirteen ' letters/ the same as those we now possess. Of works im- properly ascribed to Plato he reckoned ten, five of which are still extant. No one now supposes that anything which was rejected by Thrasyllus is a genuine work of Plato, but there has been during the last sixty years a good deal of discussion as to whether all that was included by Thrasyllus may safely be accepted. The extreme view that nothing contained in the canon of Thrasyllus is spurious has found no im- portant defender except Grote, whose reasoning is vitiated by the double assumption that every- thing accepted as genuine by Thrasyllus must have been guaranteed by the Alexandrian library, and that the Alexandrian librarians themselves cannot have been misled or imposed upon. On the other hand, the scepticism of those German critics of the last half of the nineteenth century, who rejected as spurious many of the most im- portant dialogues, including, in some instances, works (e.g. the Laws) which are specifically named as Plato's by Aristotle, has proved itself even more untenable. Our surest guide in the matter, 22 LIFE AND WRITINGS wherever obtainable, is the evidence of Aristotle, and an increasingly careful study of the Aris- totelian text has now enabled us to say that, though some of the chief dialogues are never actually cited by Aristotle in express words, there is none of them, with the doubtful excep- tion of the Parmenides, which is not alluded to by him in a way in which, so far as we can dis- cern, he never makes use of any works except those of Plato. There is thus at present a general agreement among scholars that no considerable work in the canon of Thrasyllus is spurious. The few dialogues of his list which are either certainly or possibly spurious are all of them, from the philosophical point of view, insignificant, and no difference is made to our conception of Platonism by our judgment upon them. A more important question than that of the genuineness or spuriousness of the few minor dialogues about which it is still permissible to doubt is presented by the problem of the order of composition of the leading dialogues. Until some conclusion has been established as to the order in which Plato's principal works were com- posed, it is impossible to form any intelligible theory of the development of Plato's thought. Now it so happens that the only positive piece of 23 PLATO information on this point which has come down to us from antiquity, is the statement of Aristotle that the Laws was a later work than the Republic. The dialogues themselves enable us to supplement this statement to a slight extent. Thus the Sophistes and Politicus are expressly repre- sented as continuations of the conversation con- tained in the Theaetetus, and must therefore be later than that dialogue, and for a similar reason the Timaeus must be later than the earlier books of the Republic, since it recapitulates in its open- ing the political and educational theories of Republic ii.-v. And further, a dialogue which quotes from another, as the Republic appears to do from the Phaedo, and the Phaedo from the Meno, must, of course, be later than the dialogue quoted. But the results which can be won by considerations of this kind carry us only a little way, and, in the main, students of Plato were until forty years ago about as devoid of the means of forming a correct conception of the develop- ment of Plato's thought as students of Kant would have been of the means of writing the history of Kantianism, if the works of Kant had come down to us entirely undated. Each scholar had his own theory of the order of the dialogues, founded upon some fanciful principle of arrange- 24 LIFE AND WRITINGS ment for which no convincing grounds could be given. The first step towards the definite solution of the problem by rational methods was taken by Professor Lewis Campbell in 1867 in his edition of the Sophistes and Politicus. Starting from the universally recognised fact that the Laivs must, on linguistic grounds, as well as on the strength of ancient tradition, be regarded as Plato's latest composition, Professor Campbell proposed to treat the amount of stylistic resem- blance between a given dialogue and the Laws, as ascertained by minute linguistic statistics, as a criterion of relative date. The method of investi- gation thus pointed out has been since followed by a number of other scholars, and notably, and with the greatest wealth of detail, by W. Lutos- lawski in his work on The Origin and Growth of Plato s Logic. At the same time, much additional light has been thrown on the subject by the more careful investigation of the numerous half- concealed polemical references in Plato to Isocrates, and in Isocrates to Plato. The result is that while we are still by no means able to arrange the works of Plato in an absolutely certain serial order, there is, in spite of some indi- vidual points of disagreement, a growing consensus among scholars as to the relative order of succes- 25 PLATO sion of the principal dialogues. For a full account of the methods just referred to and the results to which they lead, the reader may be referred to the recent work of Hans Raeder, Platon's Philo- sophische Entwickelung. I shall content myself here with a statement of what appear to be the main results. Plato's genuine writings fall on examination into four main classes. These are: (1) Early dialogues, marked by the freshness of the dramatic portraiture, the predominant preoccupa- tion with questions of ethics, and the absence of the great characteristic Platonic psychological, epistemological, and metaphysical conceptions, particularly of the famous theory of 1 Ideas.' To this group belong the dialogues which have often been called par excellence ' Socratic,' such as the Apology, Crito, Charmides, Laches, Euthyphro, Euthydemus, and probably Cratylus. The most important members of the group are the Prota- goras and Gorgias, the latter being almost certainly the last of the series. There is reason, as already said, to regard the Gorgias as probably composed soon after 387, when Plato was beginning his career as president of the Academy. (2) A group of great dialogues in which Plato's literary power is at its height, and which are all marked by 26 LIFE AND WRITINGS the central position given in them to the ' theory of Ideas/ with its corollary, the doctrine that scientific knowledge is recollection. The Meno ap- pears to furnish the connecting link between this group and the preceding ; the other members of it are the Symposium, Phaedo, Republic, Phaedrus. Of these the Phaedrus has been shown, conclu- sively as I think, by Raeder, and, on independent grounds, by Lutoslawski, to be later than the Republic, which, in its turn, is pretty certainly later than the other two. Since the Phaedrus appears to allude to the Panegyricus of Isocrates, which was published in 380 B.C., the ' second period ' of Plato's activity as a writer must have extended at least down to that year. (3) A group of dialogues of a ' dialectical ' kind, in which the primary objects of consideration are logical questions, the nature of true and of false predication, the problem of the categories, the meaning of negation, the processes of logical division and definition. An external link is provided between the dialogues of this group by the exceptional prominence given in them all to the doctrines of the great Eleatic philosopher Parmenides. The group consists of four great dialogues, Theaetetus, Parmenides, Sophistes, Politicus. The last two are undoubtedly later than the others. They are, 27 PLATO in form, continuations of the conversation begun in the Theaetetus, and are shown to be later than the Parrnenides both by linguistic evidence and by the presence in the Sophistes of an explicit allusion to the arguments of the Parrnenides. Whether the Theaetetus is also later than the Parrnenides is still an open question, though it also contains what looks like a distinct reference to that dialogue. (4) Three important works remain which form, linguistically, a group by themselves and must be referred to the latest years of Plato's life : the Philebus, the maturest exposition of Platonic ethics; the Timaeus (with its fragmentary con- tinuation, the Critias), concerned in the main with cosmology and physics, but including a great deal that is of high metaphysical and ethical importance; and the Laws, in which the aged philosopher, without abandoning the ideals of the Republic, undertakes the construction of such a ' second-best ' form of society as might be actually practicable not for ' philosophers,' but for average fourth-century Greeks. Actual dates can hardly be determined in connection with these two last groups. We can only say that the seven works mentioned must have been written between 380 (the earliest possible date for the PJiaedrus) and 28 LIFE AND WRITINGS 347-6, the year of Plato's death, and that the difference of tone between the third and second groups of dialogues makes it almost certain that the earliest works of the third group fall at least some years later than the Phaedrus. It is tempt- ing to go a step further, and say, with Lutos- lawski, that the bitter expressions of the Theaetehis about the helplessness of the philo- sopher in practical affairs contain a personal allusion to the failure of Plato's own intervention at Syracuse, in which case the Theaetetus and all the following dialogues must be later than 367-6, but the inference is far from certain. This chapter may conveniently end with somo brief observations on the form of the Platonic writings, and the difficulties which that form creates for the interpreter of Plato's thought. In form, the philosophical works of Plato are all dramatic ; they are, one and all, SidXoyoi, conver- sations. This is true even of the Apology, which is, in point of fact, no set speech, but a series of colloquies of Socrates with his accuser and his judges. It is true that the dramatic element becomes less prominent as we pass from the earlier works to the later. In the dialogues of our last two groups, the function of the minor personages becomes less and less important. 29 PLATO They tend, more and more, to serve as mere instruments for giving the chief speaker his cue, until in the Timaeus the conversation becomes a mere prelude to the delivery of a consecutive and unbroken cosmological discourse, and in the Laws the two minor characters have little more to do than to receive the instructions of their com- panion with appropriate expressions of agreement. We note, too, that in general the position of chief speaker is assigned to Socrates, though in three of the later dialogues (the Sophistes, Politicus, and Timaeus) he recedes into the background, as though Plato felt that he was passing in these works definitely beyond the bounds of the Socratic influence, while in the Laws he disappears altogether (probably because the scene of the dialogue is laid in Crete, where the introduction of the home-keeping son of Sophroniscus would have been incongruous), and his place is taken by a 'stranger from Athens,' who is palpably no other than Plato himself. Plato's reasons for choosing the dialogue as the most appropriate vehicle of philosophical thought are not hard to discover. It was the natural mode of expres- sion for a philosophic movement which originated in the searching and incisive conversation of Socrates. Most of the * Socratic men 9 expressed 30 LIFE AND WRITINGS their ideas in the guise of Socratic dialogue, and Plato may not have been the originator of the practice. Moreover, Plato, as he himself tells us, had a poor opinion of written books as provoca- tive of thought, in comparison with the actual face-to-face discussion of problems and examina- tion of difficulties between independent seekers for truth. The dialogue form recommended itself to him as the nearest literary approximation to the actual contact of mind with mind ; it enabled him to examine a doctrine successively from the points of view of its adherents and its opponents, and thus to ensure thoroughness in the quest for truth. And finally, the dialogue, more than any other form of composition, gives full play to the dramatic gifts of portrayal of character and humorous satire in which Plato takes rank with the greatest comic and tragic masters. At the same time, Plato's choice of the dialogue as his mode of expression has created a source of fallacy for his interpreters. If we would avoid serious errors, it is necessary always to remember that the personages of one of Plato's philosophical dialogues are one and all characters in a play. ' Protagoras ' or % Gorgias,' in a Platonic dialogue, is not the historical Professor of that name, but a fictitious personage created by 3i PLATO Plato as a representative of views and tendencies which he wishes to criticise. Mingled with traits drawn from the actual persons whose names these characters bear, we can often find in the picture others which can be known or suspected to belong to the writer's contemporaries. And the same thing is true, though the fact is com- monly forgotten, of the protagonist of the drama, the Platonic ' Socrates.' 1 Socrates ' in Plato is neither, as some of the older and more uncritical expositors used to assume, the historical Socrates, nor, as is too often taken for granted to-day, the historical Plato, but the hero of the Platonic drama. The hero's character is largely modelled on that of the actual Socrates, his opinions are often those of the historical Plato, but he is still distinct from them both. In particular, it is a grave mistake of interpretation to assume that a proposition put forward by ' Socrates ' must neces- sarily represent the views of his creator, or that where ' Socrates ' declares himself baffled by a pro- blem, Plato must always have been equally at a loss. Plato shares to the full that gift of Attic 1 irony ' which is so characteristic of the great Athenian tragedians, and, as any attentive read- ing of the Protagoras will show, he has no objection to exercising it, on occasion, at the 32 LIFE AND WRITINGS expense of his principal personage. In determin- ing which of the views of his hero are put forward as his own, we, who are deprived of the oral instructions dispensed to the students of the Academy, have to observe much the same con- ditions and practise much the same precautions as are required for similar interpretation of a great dramatist or novelist. c 53 CHAPTER II KNOWLEDGE AND ITS OBJECTS The word 'philosophy,' which to us has come to mean no more than a body of theories and inquiries, has for Plato a more living and subjec- tive sense. Philosophy is, as its name declares, the love of wisdom, the passionate striving after truth and light which is, in some degree, the dower of every human soul. It belongs, the Symposium tells us, neither to the mind that is wholly wise, nor to that which is merely and complacently stupid. It is the aspiration of the partly illuminated, partly confused and perplexed, soul towards a complete vision in which its pre- sent doubts and difficulties may vanish. Accord- ing to the Theaetetus and Republic, philosophy begins in wonder, or more precisely in the mental distress we feel when confronted by conflicting perceptions, each apparently equally well ac- credited. In a famous passage this state of distress, in which the soul is, so to say, in travail 34 KNOWLEDGE AND ITS OBJECTS with a half-formed idea, is likened to the pains of child-birth, and the philosopher is presented, in his relation to his disciples, as the midwife of the spirit. His task is not to think for other men, but to help them to bring their own thoughts to the birth. This conception of philosophy and its function is far from being narrowly 'intellectualist' in a bad sense. Philo-; sophy is, in Plato's eyes, a ' wayj)f life/ a discipline for character no less than for understanding. But it is his conviction that there is a deep truth enshrined in the crude saying of the old physi- ologists that ' like is known by like/ His theory iof education is dominated by the thought that; the mind itself inevitably ' imitates ' the character of the things it habitually contemplates. Just because the aspiration after wisdom is the funda- mental expression of the mind's true nature, it cannot be followed persistently without resulting in a transfiguration of our whole character; its ultimate effect is to reproduce in the individual soul those very features of law, order, and rational purpose which the philosopher's contemplation reveals as omnipresent in the world of genuine knowledge. Yet the starting-point of the whole process is an intellectual emotion, a passion for insight into truth. The upward pilgrimage of 35 PLATO the soul begins for Plato not, as for Bunyan, with ' conviction of sin,' but with that humiliating s ense of ignorance w hich Socrates aimed at pro- ducing in those who submitted to his cross- questioning. Insight and enlightenment are the first requisites for sound morality, no less than for science. In action as well as in speculation, what distinguishes the 'philosopher' from other men is the fact that where they have mere 'opinions' he has 'knowledge/ i.e. convictions which have been won by free intellectual inquiry and can be justified at the bar of reason. The ' theory of knowledge 1 is thus the very centre of Plato's philosophy. He takes his stand upon the fundamental assumption that there really is such a thing as ' science/ i.e. as a body of knowable truth which is valid always and absol- utely and for every thinking mind. The problem he sets before himself in his metaphysics is to find the answer to the question ' How is science possible ? ' 1 What is the general character which must be ascribed to the objects of our scientific knowledge ? ' Plato may, therefore, in spite of Kant's hasty inclusion of him among the dog- matists, be truly said to be a great 'critical' philosopher, and, indeed, with a partial reservation in favour of his revered predecessor Parmenides, 36 KNOWLEDGE AND ITS OBJECTS the earliest critical philosopher of Europe. Indeed, it is not too much to say that Plato's fundamental problem is essentially identical with that of Kant in the Critique of Pure Reason, though Plato's solution of it differs strikingly in some respects from Kant's. Like Kant, he finds his point of departure in the broad contrast between the world of everyday unsystematised ' experience,' and that of science. The world as it appears to the everyday unscientific man is a scene of strange disorder and confusion; his so-called experience is made up of what Plato calls c opinions/ a multitude of conflicting and changing beliefs, some of which are often actually contra- dictory of others; he can give no satisfactory grounds for regarding them as true, and can often be persuaded out of them by appeals to irrational emotion. Science, on the other hand, is a body of consistent and fixed convictions, a system of truths, valid absolutely, always, and for every one, in which the various members are connected by a bond of logical necessity — in a word, a body of reasoned deductions from true principles. What then is the relation between these apparently so diverse worlds, that of ' opinion 1 and that of ' science ' ? In more modern lan- guage, of what nature are the objects cognised by 37 PLATO the universal propositions of science, and how are they related to the particular percepts of sense? Plato's answer to this question is con- tained in his famous ' Theory of Ideas/ which is thus, according to its author's intention, neither 1 dogmatic 1 metaphysics nor poetical imagery, but a logical doctrine of the import of universal propositions. The real character of this central Platonic doctrine, as primarily a theory of pr.QdloatjQii, is bttmrnm^ mx* * % -jf-n.-w- -"■11*7-1 — i -'^fc^M^ 3 * mm - well brought out by the succinct account of its meaning and its logical connection with previous Greek thought given by Aristotle in his Meta- physics. According to Aristotle, the doctrine was a logical consequence from two premises, taken one from Heracliteanism, the other from Socrates. From Heracliteanism Plato had learned that all the kinds of things which our senses perceive are 1 in flux/ i.e. are constantly undergoing all sorts of incalculable changes, and consequently that no universal truths can be formulated about them. (Cf. Locke's doctrine that all our certain know- ledge of 'nature' is 'barely particular.') From | Socrates, whose methods, though used by himself ; only in the discussion of 'matters of conduct,' I were really of universal application, he further \ learned that without universal truths there can 38 KNOWLEDGE AND ITS OBJECTS be no science. Hence, since there is such, a thing as science, Plato inferred that the objects which science defines, and about which she undertakes to prove universally valid conclusions, cannot be the indefinitely variable things of the sensible physical world. There is therefore a supra- physical world of entities, eternal and immutable, and it is these unchanging entities, called by Plato ' Ideas/ which are the objects with which the definitions and universal truths of exact science are concerned. The relation between this world of pure logical concepts and the world of every- day sensible experience is that the things of the sensible world are approximate and imperfect resemblances of the corresponding conceptual entities from which they get their various class- names. This relation Plato calls ' participation in ' the Ideas, a phrase to which Aristotle objects that it is no more than a misleading imaginative metaphor. Such is the preliminary account which Aristotle prefixes to his 'smashing 1 attack on the Platonic metaphysics. When we turn to the great dialogues, such as the Phaedo, Republic, Timaeus, in which the doctrine of Ideas is most prominent, we find this account, so far as it goes, fully borne out. In the Timaeus, in the only passage where Plato ever 39 PLATO I directly raises the question whether the ' Ideas ' actually exist or not, their existence is said to be a necessary implication of the reality of the •distinction between 'true opinion ' and 'science/ If science is no more than true opinion, there need be no objects except those of the physical and sensible world ; if science is other than true opinion, there must be a corresponding difference between objects of which we can only have true opinion and objects of which we can have scientific knowledge. But science is assuredly something more than true opinion ; it deals with things which cannot be perceived by the senses, but only conceived by thought j it is eternally and immutably valid; it rests on rational grounds and logical proof. ' Ideas ' therefore exist. So again, in the three dialogues alike, Ave find that there is a standing contrast between the unity of the ' Idea 1 and the multiplicity of the things which, as Plato puts it, * participate \ in the \ Idea/ or, as we should say, of whipk- tha corresponding term can be predicated. There are a countless host of beings whom we call men or oxen, of things which we speak of as just or beautiful, but the humanity we predicate of one man is identi- cal with the humanity we predicate of any other : the justice or beauty in virtue of which we call 40 KNOWLEDGE AND ITS OBJECTS different persons or things just or beautiful is always one and the same. Again, a thing or person who is beautiful may become unbeautiful, may cease, in Platonic language, to ' participate in' Beauty, but Beauty itself never begins nor ceases, but simply is eternally identical with itself. And, once more, the pure logical concept is never fully embodied in any sensible example : two things, for instance, which at the first blush appear equal, on closer comparison will be found to be only approximately so ; the visible diagram which we take to stand for a triangle, in studying geometry, has never really the properties which we attribute to ' the triangle 9 in our definition ; the conduct we praise as just may, on close scrutiny, turn out to be only imperfectly just. Thus Socrates, ABC, the conduct of an Aris- tides, ' partake ' of humanity, of triangularity, of justice; they are not humanity, triangu- larity, justice 'themselves/ 'What man is in itself/ 'what justice is in itself/ is always something other than any one man or any one just deed. Considerations like these show us clearly what is the entity to which Plato gives the name of an I ' Idea/ It is what we should now call the ' signi- fication' or 'intension' of a class-name, as dis- 4i PLATO tinguished from its ' extension/ The extension of the name is what Plato means when he speaks of the 'many things which partake of* the one Idea or class-concept. Consequently he some- times says that there exists an Idea for every group of many things which 'have a common name/ and we find Aristotle using the expression ' the One over the Many ' as a synonym for the Platonic Idea. But this restriction of the range of Ideas to classes of many things with a com- mon class-concept is not really involved in the general theory of the nature of the Idea, since, as the existence of significant singular terms shows us, classes with only one member are just as common in logic as classes with many, and so we find Plato in the Timaeus explicitly recognising one such concept or Idea which is ' partaken of ' by only one sensible thing, viz. the Idea or con- cept of the physical universe itself as a whole (the so-called avroggoy). By the 'Ideas,' then, Plato means the system of terms or concepts of fixed and determinate intension which would form the contents of an ideally perfect science, and which form the content of our existing science in so far as it is completely and rigidly 'scientific/ the system of universal meanings. Before we go further, it may be as well to call 42 KNOWLEDGE AND ITS OBJECTS attention to one or two points in regard to which his doctrine is capable of being and has often actually been misunderstood. (1) Plato's theory of ' Ideas/ as the true objects of knowledge, is not at all a doctrine of ' Idealism ' in the modern sense of the word. He calls the concepts of science ' Ideas 1 simply because they constitute the forms or types in accordance with which the universe of things is constructed ; the words ISea, eZ£o?, simply mean 1 shape/ 1 form/ and nothing more. We merely miss his meaning if we allow the Berkeleyan notion of an 1 idea 9 as a state of mind to affect our interpretation of him. The suggestion that an ' Idea ' is something which only exists ' in a soul/ and therefore is a 1 thought/ is only made once in Plato's writings, in a passage of the Parrnenides, and is only put forward there to be promptly rejected. With Plato the 1 Ideas ' are not ' states ' of the knowing mind, but objects distinct from and independent of itself, about which it has knowledge. It is only with the Neo-Platonists, who taught that 'objects of thought have no subsistence outside the tkinET ing mind/ that we come within 'measurable dis- tance of any form of modern ' Idealism.' Hence ' conceptual realism' is a much better and less ambigi^s'name ttan 1 Idealism ' for the type of PLATO doctrine of which Plato is the most illustrious exponent. (2) It follows also at once that, since the ' Ideas ' are not processes of thought but objects of thought, we must not conceive of them as the thoughts of the divine mind, ' creative conceptions ' of God. This interpretation of Plato is as old, at least, as the Alexandrian Jewish philosopher Philo, a contemporary of Christ, and has found notable support in modern times, but is, none the less, thoroughly un-Platonic. It is not easy to tell how far, when Plato speaks of personal gods or a personal God, he is using the language of exact philosophy, or how far he is merely accom- modating himself to the current phraseology of his time ; but this much, at least, is clear. When j God is spoken of in connection with the ' Ideas/ \ as in the Timaeus, where he is imaginatively portrayed as shaping the physical universe on the model of the ' Ideas/ the ' Ideas ' are always referred to as objects existing independently of God and known by Him, never as owing their exist- ence to His thought about them. In fact, whatever may have been Plato's precise concep- tion of God, God appears in the language of the dialogues as altogether secondary in his system ; it is the ' Ideas/ and not, as in so many modern 44 KNOWLEDGE AND ITS OBJECTS systems, God, which are, for Plato, the ens realis- simum. As we shall see further in the next chapter, it is just because God and ' the soul ' are not for Plato entia realissima that he has to employ imaginative myths when he would speak of them, whereas his language, when he deals with the ' Ideas/ is as devoid of mythical traits as the multiplication table. (3) In speaking of the relation between the members of the extension of a class-name and the common intension or class-concept or 'Idea' which corresponds to them, Plato employs not only the expression, regarded by Aristotle as specially characteristic of him, that the various things 4 partake of the Idea, but a number of equivalent phrases. Thus it is said (Phaedo) that the things ' have communion with ' (tcoLvcovel) the Idea, or that the Idea ' is present to 1 {irdpeaTi) them ; and it is explicitly declared that it does not matter which of these expressions we use, so long as we understand the relation which they all denote. Yet another way of expressing the same relation is to say that the things are ' imita- tions ' (jLLLfjLijfiaTa) or ' copies ' of Ideas. This form of expression naturally meets us more particularly in the semi-mythical cosmogony of the Timaeus, but it is found also side by side with the language 45 PLATO about ' participation 9 in the Republic ; and in the Parmenides, where a number of difficulties are being raised about the nature of 'participation/ it is expressly asked whether the ' participation 1 of things in the Ideas may not be explained more exactly by the view that they are ' likenesses ' of them. The same metaphor is, of course, implied in all passages which dwell upon the imperfect and merely approximate character of the embodi- ment of Ideas in sensible things. Hence it is clear that there is no ground for the recent inter- pretation which distinguishes between an earlier version of Platonism, according to which things 'participate' in Ideas, and a later version in which they merely 'resemble' them. In fact, as has well been shown by Professor Shorey, the whole conception of a marked difference between an earlier and a later Platonic metaphysic has no tenable foundation. All these different meta- phors are intended to express one and the same relation, viz. that which subsists between the subject and predicate of such propositions as 'Socrates is a man,' ( ABG is a triangle,' the relation, that is, between the individual member of a class and the class to which it belongs. The peculiarity of Plato's view is that, whereas modern exact logic treats this relation (denoted in the 46 KNOWLEDGE AND ITS OBJECTS I symbolism of Peano by e) as a relation between . the individual and the extension of the class \ (' Socrates is a man ' = ' Socrates is one member of the group men '), Plato treats it as a relation be- • tween the individual and the intension of the class- name (' Socrates is a man ' = ' Socrates possesses humanity/ or 'humanity is found in Socrates'). (4) From the time of Aristotle to the present day the point which has given rise to the sharp- est criticism of Plato has always been his in- sistence on the ' tmn s c ft ^gnt ' character of the Ideas. This character is expressed in Plato by his reiterated assertion that the Ideas are some- thing 'separate from' (^go/hs V the things which ' participate in 9 or ' resemTfilethem, and are called by their names. It is upon this point that Aris- totle's most incisive attacks upon the Platonic theory turn. He treats Plato's doctrine as amounting to the assertion that, e.g. ' humanity ' and 1 triangularity ' are things which exist apart from and outside of all actually existing men or triangles, and objects that, if this is so, there can be no intelligible connection between the sup- posed world of Ideas and the world of concrete realities. If the Idea is ' outside ' or ' apart from 1 6 all sensible things, how can it be their inmost reality or 1 substance ' ; and again, how can know-. 47 PLATO ledge of Ideas and their relations contribute in any way to our scientific knowledge of the real world ? Aristotle thus leads the way in regarding Plato's doctrine as a 'reification of concepts/ a fallacious attribution of"Substantive existence to universal predicates, and condenses his objection to it in the statement that what science requires is not that there should be 'Ideas, or a One which is something over and above the Many,' but merely that one attribute should be predi- cable of many subjects. The difficulty has been felt so strongly by modern interpreters that many of them have endeavoured, in the face of Plato's plainest declarations, to explain it away, and thus to bring Plato's theory of predication into accord with that of his great disciple. Plato's language, however, is too explicit to permit of any such interpretation, as Aristotle was well aware. More careful consideration will, I think, both explain its true meaning and throw some light on a probable source of the Platonic theory which Aristotle's analysis leaves only imperfectly indi- cated. If we consider the passages, from the Pltaedo onwards, in which Plato insists most strongly on the ' transcendent ' and ' separate ' character of the Idea and the imperfection of its sensible embodiment, we shall find that his illus- 48 KNOWLEDGE AND ITS OBJECTS trations are drawn from two spheres, those of mathematics and of ethics. It is primarily in mathematics and in ethics that the Idea most obviously appears as an Ideal, a conceptual limit to which experience only presents imperfect approximation. And when we remember the importance attached by Plato to measure, order, and proportion as characteristics of the morally good, we may see reason to reduce the two cases to one. It is primarily from mathematics that Plato has derived his conception of science and its concepts and their relation to the world of experience. Now, as Plato himself reminds us in the Republic, the visible diagrams of the mathematician are only aids to the imagination ; tihey are not themselves the true objects of his reasoning. He may represent a point by a visible dot, or a line by a stroke drawn with chalk; but these dots and strokes are not really the points and lines about which he is reasoning, and have not really the properties which he ascribes to the point and the line {e.g. the visible dot is not, like the true point, a thing without parts or magnitude ; the visible stroke is never devoid of breadth or absolutely straight, and so on). The real objects of mathematical study are a system of pure logical concepts which can be thought D 49 PLATO with exact precision but cannot be adequately- represented to sense or imagination. In Plato's theory of Ideas we have a conception of science which rests upon the view that mathematics is the one and only true science, a consistent work- ing out of the thought expressed by Kant in his saying that every study contains only so much of science as it contains of pure mathematics. That Plato's doctrine of knowledge should thus have arisen primarily from reflection upon the concepts and methods of pure mathematics is in accord not only with the special prominence given both in the dialogues and, so far as we can learn, in the oral teaching of the Academy, to mathematical study, but also with the historical fact that pure mathematics was in Plato's time the only scientific study in which certain and well-established results had been attained. These same considerations also explain why the answer given by Plato to the question 1 How is scientific truth possible ? ' differs so greatly from the answer given by Kant and his followers to the same problem. For Plato the great Kantian problem 'How is pure a priori natural science possible ? 1 does not exist. ' Natural science/ in the sense of proved universal laws of physical process, had for him no being. A true ' science KNOWLEDGE AND ITS OBJECTS of physical nature 9 could, of course, not exist in the fourth century B.C., which possessed neither the appliances requisite for precise determination of physical magnitudes nor the mathematical methods necessary for the establishment of general laws on the basis of individual observa- tions. But even had the physics of the twentieth century a.d. been known to Plato, it is pretty clear that he would still have refused to bestow the title of science upon our knowledge of actual nature. He would have called attention to the merely approximate character of all actual physi- cal measurements, and the necessity of admitting that the course of any actual physical process may be influenced by the presence of conditions which are neglected in our formulation of 1 general laws of sequence,' in justification of the view that our results are only strictly proved, and therefore rigidly scientific, so long as we confine ourselves within the domains of pure conceptual mathematics. For him the actual physical world, just because it cannot be completely analysed into combinations of logical concepts, but involves a factor of irrational sensible fact, is incapable of being an object of science proper. Any conclu- sions we may form as to its structure and history must be put forward, not as proved results of 5i PLATO science, but as, at best, a 'probable account.' The physical world is thus the proper object of ' opinion,' and any account of its development must be, like the narrative of Timaeus, largely mythical. (Compare once more Locke's doctrine of the extent of our knowledge of 1 real existence ' of sensible things, and the position of those eminent logicians who hold that 1 induction ' from observed facts is unable to lead to results which are more than probable.) Hence while Kant denounces all 1 transcendent ' employment of the fundamental concepts of science, and confines knowledge within the limits of 'possible experi- ence,' Plato, to put the matter qui te^lSSftty, holds that all true science is 1 transcendent ' and deals with objects which lie entirely beyond the range of any possible 'experience' of sense. Where 1 experience ' begins, science, in his opinion, leaves off. That Kant does not come to the same con- clusion seems to be due to his assumption that the sciences of Arithmetic and Geometry deal with objects which are not analysable into purely logical concepts, but involve an element of irra- tional sensuous 1 intuition.' It is this assumption which Plato is really denying by anticipation when he says in the Republic that the diagrams of the geometers are mere aids to the imagination, 52 KNOWLEDGE AND ITS OBJECTS and not themselves the objects of geometrical reasoning, and again in the Timaeus that space is apprehended not by sense but by a 'kind of bastard thought/ We have seen, then, what is the general char- acter of the system of Platonic Ideas, the true object of scientific knowledge. It is a world of exactly defined logical concepts, each standing in immutable relations to the rest. Further light is thrown upon the internal structure of this system by a famous passage at the end of the sixth book of the Rejmblic, which, taken along with the exposition of it in the following book, is by; far the most important single text for thej whole of Plato's epistemology. In this passage' Plato is concerned to distinguish four grades of cognition, and to provide each with its appropriate class of objects. He illustrates his meaning by what is, in point of fact, a diagram. He takes a vertical line, and begins by dividing it into an upper and a lower segment, the upper segment representing knowledge or science, the lower 'opinion/ Each segment is then, in turn, once more divided into an upper and a lower part, in the same ratio in which the whole line was originally divided. We thus get an inferior and a superior form of 'opinion' and of knowledge S3 PLATO respectively, the inferior, in each case, standing in respect of its truth and certainty to the higher, in the same relation in which 'opinion* as a whole stands to knowledge. The lowest type of cognition of all, the inferior form of ' opinion/ Plato calls eucaala, ' g^ss^voA/ with a punning allusion to the el/coves or ' images ' which are its appropriate objects. It is the state of mind in which reflections in water and the imagery of dreams are not as yet distinguished from the solid physical realities of which they are the images, t he mental condition of the savage or child at the mercy of ' primitive credujity,' who accepts every presentation, so long as it lasts, as equally true with any other, and has not yet learned to know the shadow from the substance. A more developed and truer form of cognition is represented by ttigtis, belief, the state of mind of the man who, while still recognising the existence of nothing but the sensible, has learned to dis- tinguish physical things from their mere shadows or reflections or dream-images, and thus to make a distinction between the truth- values of the two kinds of presentation. Such a man, though as yet not possessed of proved and universal scientific truth, has already a fair stock of tolerably system- atised and trustworthy convictions about the 54 KNOWLEDGE AND ITS OBJECTS empirical course of things, and, as we have seen, Plato holds that this is the highest degree of truth which can be attained in the study of the actual physical world. Thus ttLcttis corresponds exactly to what we should call sensible experience, and the knowledge based on induction from such experience. A further step is taken towards the ideal of genuine knowledge when we pass from the higher form of 'opinion' to the lower form of science. Plato's name for this inferior grade of science is hiavoia, which we may loosely render ' understanding/ and it is declared to be the knowledge supplied by ' geometry and the kindred arts/ i.e. mathematics as usually studied. Being ' science/ these studies have for their object concepts of a purely rational kind, and hence Plato observes that they use diagrams and models, which for 1 opinion ' are realities as mere images of the higher realities with which they are con- cerned. But he finds two defects in the procedure i of ordinary mathematics: the mathematician employs sensible aids, even if he uses them only as aids to his imagination ; he also makes use i of a host of notions which he has not defined | and postulates which he has not proved. Hence Plato maintains that there must be a still more perfect realisation of the ideal of knowledge, 55 PLATO which is given by a science called by him ' dia- lectic/ which has for its objects • Ideas 1 or ' Forms 9 themselves, and studies them without the aid of any sensuous representations whatever. The procedure of ' dialectic/ as he describes it, is two- fold: a process of analysis followed by one of synthesis. The dialectician will start in his turn with the axioms and indefinables of the ordinary mathematician, but he will not regard them as ultimate. He will treat them as literally ' hypo- theses/ bases or starting-points from which he may ascend to a supreme first principle which is ■ unhypothetical ' ; then from the cognition of this first principle he will once more descend by a regular gradation to the knowledge of its con- sequences, proceeding throughout ' from forms to forms without the aid of anything sensible/ * That is, it seems, the dialectitian is to compare the principles assumed as ultimate by the various ; branches of mathematics, and as a result of the ; comparison to arrive at some still more ultimate first principle of a logical character which is self- evidently true. Having done this, he is then to deduce the supposed ultimates of the ordinary mathematician, and through them their conse- quences, from his own supreme and self-evident axiom. Only when this has been done shall we 56 KNOWLEDGE AND ITS OBJECTS have realised the ideal of scientific investigation, the reduction of known truth to a systematic body of logical deductions from true and ultimate premises. It is clear from this that Plato's conception is closely akin to the ideal of the growing school of mathematicians who maintain that the whole of pure mathematical science is a body of deduc- tions from a few ultimate premises which are all of a purely logical kind, and require for their statement no primary notions except those of formal logic. Could he have met with such a work as the Formulaire of Professor Peano, or the Grunclgesetze der Arithmetik of Professor Frege, he would plainly have felt that his con- ception of 'dialectic' was there very largely justi- fied and realised. But there are also very important differences between the 'dialectic' of Plato and the 'logistic' of our contemporary philosophical mathematicians. For one thing, Plato, like Leibniz after him, dreamed of the deduction of all pure science from a single ultimate principle, while the development of exact logic has definitely shown that the prin- ciples of logic themselves form a body of mutually independent postulates, the number of which the old traditional logic, with its three 57 PLATO laws of thought, seriously underestimated. A still more important difference appears when we ask what, in Plato's opinion, is the character of the supreme principle itself. He tells us that it is ' the good ' or the ' Idea of the good/ which is to the world of concepts what ' its offspring ' the sun is to the sensible world. Now, in the sensible world, the sun has a double function. It is the source of the light by which the eye beholds both the sun itself and everything else ; it is also, as the source of heat, the cause of growth and vitality. So, in the world of concepts, the c good ' is at once the source of knowledge and illumination to the knowing mind, and the 1 source of reality and being to the objects of its I knowledge. And all the time, just as the sun is not itself light or growth, so the ' good ' is not itself Being or Truth, but the transcendent source of both. Plato's meaning in this famous passage is far from easy to grasp with precision, as he himself seems to admit, but his general sense may perhaps be divined by a comparison with well- known passages of other dialogues. There is a famous page of the SJui^o which professes, it is hard to say with what degree of accuracy, to trace the mental biography of Socrates. After re- counting his youthful dissatisfaction with the 58 KNOWLEDGE AND ITS OBJECTS mutually conflicting theories of various early physicists about the universe, Socrates is made to say that he hailed with rapture the saying of Anaxagoras that it is Mind which is the cause of order and structure in the universe. This he tookjgjn^ is responsible for the universe, its existing arrangements must be those which ar e bes t, and he eagerly procured the book of Anaxagoras in the hope that he would find there a theory of the universe in which every detail would be justified by a proof that it was better for things to be as the writer said than to be in any other way. On reading the work he was, however, disgusted to find that Anaxagoras did not live up to his own principles, but for the most part accounted for existing facts by hypo- theses of mechanical causation, only appealing to Mind as the universal cause when he was at a loss for some more specific mechanical explana- tion of a fact. This criticism of Anaxagoras, in i which Aristotle emphatically concurs, is then j made the opportunity for drawing an impor- tant distinction between the true cause of a thing and subordinate accessory conditions, * without which the cause would not be a cause.' I The true cause of every arrangement in nature - is that 'it is best that things should be so'; 59 PLATO the alleged mechanical 1 causes ' of the men of science are merely accessory conditions in the absence of which the efficacy of the true cause , would be destroyed. We find this distinction carefully observed in the half-mythical cosmo- gony of Plato's own Timaeus. The true reason or cause of the existence of the universe is the goodness of God, who, being good Himself, desired that His work too should be as good as possible ; the accessory conditions are provided for by the character of the disorderly material out of which the universe is moulded by God. Thus we see that for Plato, as for th£_Gr ee k m ind in general, to bejipod means to be good for some end or purpose, to be the expression of a rational aim or interest. Evil, on the other hand, is pre- cisely that which is disorderly, which hampers or frustrates the execution of rational purpose. (And hence, by the way, there is, on Plato's principles, an irreducible element of evil in the physical universe, precisely because that uniy ersg con - t ains, as ^ e ^stem of pj^e concepts does not, an Putting all this together, we may say that the recognition of the ' good/ as the supreme source from which the 1 Ideas ' derive their being, would appear to mean that the whole body of true irrational and incaJgHJ f ) V C _ f KNOWLEDGE AND ITS OBJECTS scientific concepts forms an organic unity in which each member is connected with the rest teleologically by the fact that some of them point forward, or logically lead up, to it, while it, in its turn, leads up to others. The objective unity of I the system of scientific concepts is thus the [ counterpart of the unity of aim and purpose ' which it is the mission of philosophy, according , to Plato, to bring about in the philosophers inner life. In the great series of dialectical dialogues, which we may safely follow the all but unanimous opinion of the latest scholars in regarding as posterior to the Phaedo- Republic group, Plato in the main turns away from his original problem of the relation between the individual thing and the intension of the class to which we refer it, to deal with further questions of logic and epistemology. This does not mean, as has sometimes been supposed, that he has abandoned or come to make serious modifications in his doctrine of ' Ideas/ as may be seen both from the reappearance of ihe\ familiar theory in the Tirnaeus and from the \ manifestly bona fide ignorance of Aristotle as to any difference in principle between an earlier and a later Platonism. WhaJ it means is simply that the whole theory of knowledge is not exhausted PLATO for Plato by any single doctrine. It is precisely in this group of dialogues that we find Plato anticipating the achievement of Aristotle in the creation of a scientific logical terminology to a degree which entitles him fairly to be called, rather than any other one man, the creator of logic. In the Parmenides, Theaetetus, and Sophistes we meet, among other things, the first attempt to construct a table of the categories, or leading scientific conceptions required for the ordering of experience. Plato's list varies slightly according as his immediate purpose demands greater or less completeness. In the Theaetetus, where his object is primarily to argue that the categories are not products of sense-perception, but are perceived 1 directly by the soul herself without the aid of bodily instruments/ i.e. are, as we should say, purely intellectual a priori forms of relation, in accord with which mind organises the material of its experience, we find him including in the list being, sameness, difference, likeness,, unlikeness, beauty, ugliness, goodness, badness,] number. All these, just because they can be pre- dicated of subjects of all kinds, he contends, cannot be cognised by the activity of any special sense. The same dialogue provides us, among other contributions to logical theory, with the 62 KNOWLEDGE AND ITS OBJECTS important distinction between two kinds of ' change 9 (/elvrjo-is), local motion, or change of \ position (cfrojpij, and alteration, or change of quality (aXXo tWig) , and with a searching and acute inquiry into the nature of definition, and the conditions under which definition is possible. Among its contributions to terminology we note the new words 'quality' (ttolott]^ = what-like- , ' organ, m the sense of a bodily instrument of perception, ' criterion/ ' difference/ in the specifically logical sense afterwards to be made classic by Aristotle. In the Sojihistes, where the particular distinction between that which the soul perceives through the body, and that which she perceives 'by herself' is in abeyance, the list of ' chief kinds ' or classes is given in a briefer form as being, sameness, difference, rest (or changeless- ness), motion (or change, /civrjais). And that dialogue and its continuation the Politicus are notable for the prominence given in them, as well as in the introduction to the Philebus, to the process of exhaustive logical division of a class into sub-classes by successive acts of dichotomy as a means towards exact definition. Still more interesting, as a contribution to the theory of knowledge, is the main problem with which both the Theaetetus and Sophistes are really concerned. 63 PLATO Formally the Theaetetus deals with the question 'What is knowledge?' and aims at showing that knowledge can neither be identified with sense- perception, as, according to Plato, had been held by Protagoras, nor more generally with 'true belief ; the Sophistes professes to be an attempt to illustrate the process of logical definition by finding a satisfactory definition of the class } ' sophists.' The actual ' knot ' of both dialogues is, J however, provided by a paradox of Plato's fellow- Socratic, Antisthenes the Cynic, who had main- tained that no term can be truly predicated of i any other, i.e. that the only true propositions I are identities. Plato had already touched upon this paradox in earlier dialogues, the Euthydemus and Cratylus, where, however, he treats it as a mere extravagance and a fit subject for banter and parody. In the 'dialectical' dialogues he shows himself aware of its real significance for the whole theory of knowledge. Perceiving that the very possibility of science depends upon the possibility of making true propositions in which the subject and predicate are not identical, he sets himself to work to furnish a serious refu- tation of the doctrine of Antisthenes. An immediate consequence of that doctrine is that genuine error, or 1 false opinion,' is impossible. 6 4 KNOWLEDGE AND ITS OBJECTS You cannot think 1 what is not 9 (i.e. of any- given subject A you can only think ' A is A 9 ; you cannot think 'A is B 9 or 'A is C if all predica- tion is strictly identical). Hence the problem as to the nature of error becomes fundamental for the inquiry of the Theaetetus. In suggesting that knowledge is the same thing as true belief, it is implied that there may be false beliefs, but this is exactly what the doctrine of Antisthenes denied. So in trying to define the ' sophist/ we find ourselves obliged to speak of him as a man who inculcates false beliefs for purposes of gain ; but what if the sophist should protest that there is no such thing as a false belief? In the Theaetetus the question how error is possible is left unsolved, with the consequence that the dialogue reaches no positive conclusion. We are found, in fact, to have been committing an illogicality in discussing the nature of false belief before arriving at any insight into the nature of truth. One important result is, however, obtained. It is elaborately shown that error may occur not only in judgments which involve a reference to facts of sense-experience, but in those in which both terms belong to the class which the soul 'perceives by herself, 1 as e.g. if a man should mistakenly believe the proposition '5 + 7 = 11/ E 65 PLATO /